From fabiano at fidencio.org  Sun Jul  1 19:35:49 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Sun, 1 Jul 2012 16:35:49 -0300
Subject: [virt-tools-list] [PATCH 10/10] Add desktop profile for Windows
In-Reply-To: <20120613102558.GB25661@redhat.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-11-git-send-email-fabiano@fidencio.org>
	<20120613102558.GB25661@redhat.com>
Message-ID: <CAK9pz9+_7=Hj6Ct9ysp87nbc__EzQ1uW4qoqPuMqOt_eUz=jqA@mail.gmail.com>

On Wed, Jun 13, 2012 at 7:25 AM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Mon, Jun 11, 2012 at 11:23:11PM -0300, Fabiano Fid?ncio wrote:
>> +     <xsl:template match="/install-script-config">
>> +sc config TlntSvr start= auto
>> +net user <xsl:value-of select="config/user-realname"/> <xsl:value-of select="config/admin-password"/> /add /passwordreq:no
>> +net localgroup administrators <xsl:value-of select="config/user-realname"/> /add
>> +net accounts /maxpwage:unlimited
>> +copy a:\<xsl:value-of select="config/user-realname"/>.bmp "c:\Documents and Settings\All Users\Application Data\Microsoft\User Account Pictures"
>> +REGEDIT /S a:\windows.reg
>> +EXIT
>
>
> ...<continued my reply from the fedora patch>
>
> where as to use the Windows desktop profile, an application needs to know
> that they must provide a floppy disk containing an icon file in the BMP
> format.
>
> I'm wondering how we can avoid the need for applications to know about
> these OS specific requirements.
>
>
>
>> diff --git a/data/oses/windows.xml b/data/oses/windows.xml
>> index 8769511..2fb18be 100644
>> --- a/data/oses/windows.xml
>> +++ b/data/oses/windows.xml
>> @@ -308,7 +308,10 @@
>>      </resources>
>>
>>      <installer>
>> -      <script id='http://microsoft.com/windows/sif'/>
>> +      <script id='http://microsoft.com/windows/sif/jeos'/>
>> +      <script id='http://microsoft.com/windows/sif/desktop'/>
>> +      <script id='http://microsoft.com/windows/reg/desktop'/>
>> +      <script id='http://microsoft.com/windows/cmd/desktop'/>
>>      </installer>
>>    </os>
>
> I'm wondering if we should have one <install-script> that contains
> data for all 3 files, instead of one <install-script> per file.

Agree!

>
>
>> @@ -103,15 +103,18 @@ static void script_file_name_get(char *distro)
>>  {
>>      if (!(strcmp(distro, "win2k")) || !(strcmp(distro, "winxp")) ||
>>          !(strcmp(distro, "win2k3")) || !(strcmp(distro, "win2k3r3"))) {
>> -        output = g_strdup("windows.sif");
>> +        outputScripts = g_slist_append(outputScripts, g_strdup("windows.sif"));
>> +        outputScripts = g_slist_append(outputScripts, g_strdup("windows.reg"));
>> +        outputScripts = g_slist_append(outputScripts, g_strdup("windows.cmd"));
>
>
> I'm also wondering if we can change the XML database, such that we don't
> need to hardcode these filenames.

The same that I asked in another mail.
How could we set/get it from the XML?

>
>>      } else if (!(strcmp(distro, "win2k8")) || !(strcmp(distro, "win2k8r2")) ||
>>                 !(strcmp(distro, "vista")) || !(strcmp(distro, "win7")) ||
>>                 !(strcmp(distro, "win8"))) {
>> -        output = g_strdup("windows.xml");
>> +        outputScripts = g_slist_append(outputScripts, g_strdup("windows.xml"));
>>      } else {
>>          gsize len = sizeof(distro) + sizeof(".ks");
>> -        output = g_malloc(len);
>> +        gchar *output = g_malloc(len);
>>          g_snprintf(output, len, "%s.ks", distro);
>> +        outputScripts = g_slist_append(outputScripts, output);
>
> Can be better written as:
>
>     gchar *output = g_strdup_printf("%s.ks", distro)
>
>
>
> Considering all of this data, what if we were to change the <install-script>
> XML syntax to allow multiple output files, and also provide metadata about
> the external disk that needs to be provided, and the images (or other files)
> within it.
>
> eg something like:
>
>   <install-script id='http://microsoft.com/windows/sif'>
>     <profile>desktop</profile>
>
>     <assets>
>        <disk media='floppy'/>
>
>        <resource type='image'/>
>           <path>/user.bmp</path>
>           <mime-type>image/bmp</mime-type>
>           <width>64</width>
>           <height>64</height>
>        </image>
>     </assets>
>
>     <template filename='windows.sif'>
>       <xsl:stylesheet>
>         ....
>       </xsl:stylesheet>
>     </template>
>
>     <template filename='windows.cmd'>
>       <xsl:stylesheet>
>         ....
>       </xsl:stylesheet>
>     </template>
>
>     <template filename='windows.reg'>
>       <xsl:stylesheet>
>         ....
>       </xsl:stylesheet>
>     </template>
>   </install-script>
>
>
>
>
>   <install-script id='http://fedoraproject.org/kickstart'>
>     <profile>desktop</profile>
>
>     <assets>
>        <disk media='harddisk'/>
>
>        <resource type='image'/>
>           <path>/user.png</path>
>           <mime-type>image/png</mime-type>
>           <width>128</width>
>           <height>128</height>
>        </image>
>     </assets>
>
>     <template filename='fedora.ks'>
>       <xsl:stylesheet>
>         ....
>       </xsl:stylesheet>
>     </template>
>   </install-script>
>
>
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From fabiano at fidencio.org  Sun Jul  1 19:36:01 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Sun, 1 Jul 2012 16:36:01 -0300
Subject: [virt-tools-list] [PATCH 08/10] Add desktop profile for Fedora
 15 and newer
In-Reply-To: <20120613101620.GA25661@redhat.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-9-git-send-email-fabiano@fidencio.org>
	<20120613101620.GA25661@redhat.com>
Message-ID: <CAK9pz9JVi9G6gD0Da-y1j=829Rh0o0QUUmNCzzfHGasMbLq-rw@mail.gmail.com>

On Wed, Jun 13, 2012 at 7:16 AM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Mon, Jun 11, 2012 at 11:23:09PM -0300, Fabiano Fid?ncio wrote:
>> +# Set user avatar
>> +mkdir /mnt/unattended-media
>> +mount /dev/sda /mnt/unattended-media
>> +cp /mnt/unattended-media/<xsl:value-of select="config/user-login"/> /var/lib/AccountsService/icons/
>> +umount /mnt/unattended-media
>
> Interesting. So to use the Fedora desktop profile, an application needs
> to already know that they must provide a 'sda' hard disk, containing
> a filesystem containing a specially named icon file.....

To be honest, it is part of set_direct_boot_params() function, that
will be moved from gnome-boxes to libosinfo, IIUC.
I think we can leave it as is, at least for now. But, of course, I
really want to read your opinion.

>
> ...<continued in my reply to the windows patch>
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From fabiano at fidencio.org  Sun Jul  1 19:36:20 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Sun, 1 Jul 2012 16:36:20 -0300
Subject: [virt-tools-list] [PATCH 10/10] Add desktop profile for Windows
In-Reply-To: <CAAa3hFMsZ2JmO1bbrcPsxZGJieo=z5VD9qPAVJzQ7enW5xzeHQ@mail.gmail.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-11-git-send-email-fabiano@fidencio.org>
	<20120613102558.GB25661@redhat.com>
	<CAAa3hFPZ80EwzDeg2wTPNS=+ByM2VumvRLe3c63pGxvyAFqDaw@mail.gmail.com>
	<20120613154905.GH29641@redhat.com>
	<CAAa3hFMsZ2JmO1bbrcPsxZGJieo=z5VD9qPAVJzQ7enW5xzeHQ@mail.gmail.com>
Message-ID: <CAK9pz9JVRQi3wH7T90QYHxcBp6PrSdcew3z08UOSEQ9g=hz1tw@mail.gmail.com>

On Wed, Jun 13, 2012 at 4:47 PM, Zeeshan Ali (Khattak)
<zeeshanak at gnome.org> wrote:
> On Wed, Jun 13, 2012 at 6:49 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
>> On Wed, Jun 13, 2012 at 06:44:29PM +0300, Zeeshan Ali (Khattak) wrote:
>>>
>>> >  <install-script id='http://microsoft.com/windows/sif'>
>>> >    <profile>desktop</profile>
>>> >
>>> >    <assets>
>>> >       <disk media='floppy'/>
>>> ..
>>> >  <install-script id='http://fedoraproject.org/kickstart'>
>>> >    <profile>desktop</profile>
>>> >
>>> >    <assets>
>>> >       <disk media='harddisk'/>
>>>
>>> How do we make sure that we assume the correct device name/number/path
>>> in the scripts? Why not have these as configuration parameters as well
>>> so that app tells us where to find these? Some parameters could be
>>> mandatory: app gets error if it doesn't set these params.
>>
>> I'd say the rule is that the disk external "data" disk should be the
>> first available unit number, after the primary VM disks. As written
>> the install scripts assume only a single primary VM disk is used.
>> If we wanted to allow more, we'd have to add a new parameter to the
>> install script API to allow the app to specify how many disks it is
>> attaching.
>
> How does libosinfo knows what is the primary disk? Let me explain by
> example: For fedora (or linux in general) the path to primary target
> disk could be either /dev/vda or /dev/sda, depending on whether virtio
> is being used or not. This will also impact the name/path of the
> unattended disk: /dev/sda or /dev/sdb depending on whether /dev/sda is
> being used for primary disk or not.

To be honest, it doesn't matter.
IIUC, Boxes will support only "newer" distros (Fedora 15+, for example).
And these distros, always, have support to virtio. So, we don't need
to worry about it.
Our first disk will always be /dev/vda and the disk with the image
will always be /dev/sda.

Daniel, Zeeshan, it makes sense for you?

>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124



-- 
Fabiano Fid?ncio



From berrange at redhat.com  Mon Jul  2 10:02:38 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 2 Jul 2012 11:02:38 +0100
Subject: [virt-tools-list] [PATCH 10/10] Add desktop profile for Windows
In-Reply-To: <CAK9pz9JVRQi3wH7T90QYHxcBp6PrSdcew3z08UOSEQ9g=hz1tw@mail.gmail.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-11-git-send-email-fabiano@fidencio.org>
	<20120613102558.GB25661@redhat.com>
	<CAAa3hFPZ80EwzDeg2wTPNS=+ByM2VumvRLe3c63pGxvyAFqDaw@mail.gmail.com>
	<20120613154905.GH29641@redhat.com>
	<CAAa3hFMsZ2JmO1bbrcPsxZGJieo=z5VD9qPAVJzQ7enW5xzeHQ@mail.gmail.com>
	<CAK9pz9JVRQi3wH7T90QYHxcBp6PrSdcew3z08UOSEQ9g=hz1tw@mail.gmail.com>
Message-ID: <20120702100237.GG18636@redhat.com>

On Sun, Jul 01, 2012 at 04:36:20PM -0300, Fabiano Fid?ncio wrote:
> On Wed, Jun 13, 2012 at 4:47 PM, Zeeshan Ali (Khattak)
> <zeeshanak at gnome.org> wrote:
> > On Wed, Jun 13, 2012 at 6:49 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> >> On Wed, Jun 13, 2012 at 06:44:29PM +0300, Zeeshan Ali (Khattak) wrote:
> >>>
> >>> >  <install-script id='http://microsoft.com/windows/sif'>
> >>> >    <profile>desktop</profile>
> >>> >
> >>> >    <assets>
> >>> >       <disk media='floppy'/>
> >>> ..
> >>> >  <install-script id='http://fedoraproject.org/kickstart'>
> >>> >    <profile>desktop</profile>
> >>> >
> >>> >    <assets>
> >>> >       <disk media='harddisk'/>
> >>>
> >>> How do we make sure that we assume the correct device name/number/path
> >>> in the scripts? Why not have these as configuration parameters as well
> >>> so that app tells us where to find these? Some parameters could be
> >>> mandatory: app gets error if it doesn't set these params.
> >>
> >> I'd say the rule is that the disk external "data" disk should be the
> >> first available unit number, after the primary VM disks. As written
> >> the install scripts assume only a single primary VM disk is used.
> >> If we wanted to allow more, we'd have to add a new parameter to the
> >> install script API to allow the app to specify how many disks it is
> >> attaching.
> >
> > How does libosinfo knows what is the primary disk? Let me explain by
> > example: For fedora (or linux in general) the path to primary target
> > disk could be either /dev/vda or /dev/sda, depending on whether virtio
> > is being used or not. This will also impact the name/path of the
> > unattended disk: /dev/sda or /dev/sdb depending on whether /dev/sda is
> > being used for primary disk or not.
> 
> To be honest, it doesn't matter.
> IIUC, Boxes will support only "newer" distros (Fedora 15+, for example).
> And these distros, always, have support to virtio. So, we don't need
> to worry about it.
> Our first disk will always be /dev/vda and the disk with the image
> will always be /dev/sda.
> 
> Daniel, Zeeshan, it makes sense for you?

No, we can't assume that we'll only need to handle "new" distros. One of
the major selling points of virt is that you can utilization legacy
distros. Perhaps it might not be the top priority of Boxes, but users
will certainly expect to use old distros.


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From zeeshanak at gnome.org  Mon Jul  2 23:37:57 2012
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Tue, 3 Jul 2012 02:37:57 +0300
Subject: [virt-tools-list] [PATCH 08/10] Add desktop profile for Fedora
 15 and newer
In-Reply-To: <CAK9pz9JVi9G6gD0Da-y1j=829Rh0o0QUUmNCzzfHGasMbLq-rw@mail.gmail.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-9-git-send-email-fabiano@fidencio.org>
	<20120613101620.GA25661@redhat.com>
	<CAK9pz9JVi9G6gD0Da-y1j=829Rh0o0QUUmNCzzfHGasMbLq-rw@mail.gmail.com>
Message-ID: <CAAa3hFOiES9+hz1CAt0=fjaC7w4FqUPeN78+bUiu1L3=DXZuaQ@mail.gmail.com>

On Sun, Jul 1, 2012 at 10:36 PM, Fabiano Fid?ncio <fabiano at fidencio.org> wrote:
> On Wed, Jun 13, 2012 at 7:16 AM, Daniel P. Berrange <berrange at redhat.com> wrote:
>> On Mon, Jun 11, 2012 at 11:23:09PM -0300, Fabiano Fid?ncio wrote:
>>> +# Set user avatar
>>> +mkdir /mnt/unattended-media
>>> +mount /dev/sda /mnt/unattended-media
>>> +cp /mnt/unattended-media/<xsl:value-of select="config/user-login"/> /var/lib/AccountsService/icons/
>>> +umount /mnt/unattended-media
>>
>> Interesting. So to use the Fedora desktop profile, an application needs
>> to already know that they must provide a 'sda' hard disk, containing
>> a filesystem containing a specially named icon file.....
>
> To be honest, it is part of set_direct_boot_params() function, that
> will be moved from gnome-boxes to libosinfo, IIUC.
> I think we can leave it as is, at least for now. But, of course, I
> really want to read your opinion.

He already gave you his opinion: We can't make those assumptions in
libosinfo. While Boxes may or may not be the primary user of
libosinfo, we need to keep the API generic enough for other apps too.
We can't leave this as is since this will be broken for every other
app but Boxes. Also things don't remain the same in Boxes so the next
time the VM configuration code changes, nobody would remember to
update this code in libosinfo and we'll have mysterious issues at our
hand.

-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124



From fabiano at fidencio.org  Tue Jul  3 03:22:23 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 3 Jul 2012 00:22:23 -0300
Subject: [virt-tools-list] Fwd: [PATCH 09/10] Change
 osinfo_install_script_generate to write the output in a file
In-Reply-To: <CAK9pz9JA7Tnf-Kjcpn5g=YTFmK++addyrHBdjV7Efxcpb3g0zQ@mail.gmail.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-10-git-send-email-fabiano@fidencio.org>
	<20120613103556.GJ25661@redhat.com>
	<CAK9pz9JA7Tnf-Kjcpn5g=YTFmK++addyrHBdjV7Efxcpb3g0zQ@mail.gmail.com>
Message-ID: <CAK9pz9+NfdCDxCmk+yD+Zt=2m891sg-DOnfiShtY5oJ1TFYw4g@mail.gmail.com>

---------- Forwarded message ----------
From: Fabiano Fid?ncio <fabiano at fidencio.org>
Date: Sun, Jul 1, 2012 at 4:36 PM
Subject: Re: [virt-tools-list] [PATCH 09/10] Change
osinfo_install_script_generate to write the output in a file
To: "Daniel P. Berrange" <berrange at redhat.com>


On Wed, Jun 13, 2012 at 7:35 AM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Mon, Jun 11, 2012 at 11:23:10PM -0300, Fabiano Fid?ncio wrote:
>>     For libosinfo API:
>>         Now, instead of return a string, writes that string in a file,
>>         passed as argument
>>
>>     For osinfo-install-script tool:
>>         The output will be wrote in:
>>             - Linuxes: <short_id>,ks
>>             - Windows 2k3r2 and older: windows.sif
>>             - Windows 2k8 and newer: windows.xml
>>
>>     It will be used to create, easily, multiple scripts, as used in:
>>     http://bugzilla-attachments.gnome.org/attachment.cgi?id=214681
>
> I was a bit dubious about this one at first, but seeing what you
> needed to do with the windows scripts, I understand now.
>
> If we follow my suggestion, of actually putting the required
> filename in the XML metadata, this API will need to change
> slightly too.

Sorry, It is not so clear to me. How could I set/get the filename in the XML?
Example: <template filename=<xsl:value-of select="config/filename"/>>?
Didn't test but I think I can't do this.

>
>> +gboolean osinfo_install_script_generate(OsinfoInstallScript *script,
>> +                                        OsinfoOs *os,
>> +                                        OsinfoInstallConfig *config,
>> +                                        gchar *path,
>> +                                        GCancellable *cancellable,
>> +                                        GError **error);
>
>
> I'd suggest it should take a 'GFile *outputdir' parameter,
> and then for each template filename listed in the XML,
> we generate  'outputdir/filename'.  Using GFile is preferrable
> to gchar *path, since then we're not tied to local paths - apps
> can use any URI scheme supported by GIO.

Thinking here, we really need this change?
osinfo_install_script_generate can just return the string. Why?
Because it is what the apps will use, commonly.
If some application (as osinfo-install-scirpt tool) would write the
output in the file, we could do it in their own code, adopting your
suggestion.
Another thing is: if we will set/get the filename we could not pass
the "filename" as argument, too?

>
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



--
Fabiano Fid?ncio


-- 
Fabiano Fid?ncio



From zeeshanak at gnome.org  Tue Jul  3 17:18:44 2012
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Tue, 3 Jul 2012 20:18:44 +0300
Subject: [virt-tools-list] Fwd: [PATCH 09/10] Change
 osinfo_install_script_generate to write the output in a file
In-Reply-To: <CAK9pz9+NfdCDxCmk+yD+Zt=2m891sg-DOnfiShtY5oJ1TFYw4g@mail.gmail.com>
References: <1339467791-13443-1-git-send-email-fabiano@fidencio.org>
	<1339467791-13443-10-git-send-email-fabiano@fidencio.org>
	<20120613103556.GJ25661@redhat.com>
	<CAK9pz9JA7Tnf-Kjcpn5g=YTFmK++addyrHBdjV7Efxcpb3g0zQ@mail.gmail.com>
	<CAK9pz9+NfdCDxCmk+yD+Zt=2m891sg-DOnfiShtY5oJ1TFYw4g@mail.gmail.com>
Message-ID: <CAAa3hFM76exvNTa00FMRjsKUy7X5FYVHbgWp1--H5tXVReQdHg@mail.gmail.com>

On Tue, Jul 3, 2012 at 6:22 AM, Fabiano Fid?ncio <fabiano at fidencio.org> wrote:
> ---------- Forwarded message ----------
> From: Fabiano Fid?ncio <fabiano at fidencio.org>
> Date: Sun, Jul 1, 2012 at 4:36 PM
> Subject: Re: [virt-tools-list] [PATCH 09/10] Change
> osinfo_install_script_generate to write the output in a file
> To: "Daniel P. Berrange" <berrange at redhat.com>
>
>
> On Wed, Jun 13, 2012 at 7:35 AM, Daniel P. Berrange <berrange at redhat.com> wrote:
>> On Mon, Jun 11, 2012 at 11:23:10PM -0300, Fabiano Fid?ncio wrote:
>>> ? ? For libosinfo API:
>>> ? ? ? ? Now, instead of return a string, writes that string in a file,
>>> ? ? ? ? passed as argument
>>>
>>> ? ? For osinfo-install-script tool:
>>> ? ? ? ? The output will be wrote in:
>>> ? ? ? ? ? ? - Linuxes: <short_id>,ks
>>> ? ? ? ? ? ? - Windows 2k3r2 and older: windows.sif
>>> ? ? ? ? ? ? - Windows 2k8 and newer: windows.xml
>>>
>>> ? ? It will be used to create, easily, multiple scripts, as used in:
>>> ? ? http://bugzilla-attachments.gnome.org/attachment.cgi?id=214681
>>
>> I was a bit dubious about this one at first, but seeing what you
>> needed to do with the windows scripts, I understand now.
>>
>> If we follow my suggestion, of actually putting the required
>> filename in the XML metadata, this API will need to change
>> slightly too.
>
> Sorry, It is not so clear to me. How could I set/get the filename in the XML?
> Example: <template filename=<xsl:value-of select="config/filename"/>>?
> Didn't test but I think I can't do this.
>
>>
>>> +gboolean osinfo_install_script_generate(OsinfoInstallScript *script,
>>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?OsinfoOs *os,
>>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?OsinfoInstallConfig *config,
>>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?gchar *path,
>>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?GCancellable *cancellable,
>>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?GError **error);
>>
>>
>> I'd suggest it should take a 'GFile *outputdir' parameter,
>> and then for each template filename listed in the XML,
>> we generate ?'outputdir/filename'. ?Using GFile is preferrable
>> to gchar *path, since then we're not tied to local paths - apps
>> can use any URI scheme supported by GIO.
>
> Thinking here, we really need this change?
> osinfo_install_script_generate can just return the string. Why?
> Because it is what the apps will use, commonly.

For that you can create add a utility function
osinfo_install_script_generate_to_path() that creates the GFile from
the path, calls  osinfo_install_script_generate and then frees the
GFile. Whats more important for apps is for all these I/O functions to
have async variants.

-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124



From zeeshanak at gnome.org  Wed Jul  4 19:45:29 2012
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 4 Jul 2012 22:45:29 +0300
Subject: [virt-tools-list] Tracking VM installations
Message-ID: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>

Hi,

We have a requirement in Boxes for showing progress bars on VMs under
installation:

https://github.com/gnome-design-team/gnome-mockups/raw/master/boxes/boxes-install6.png
https://bugzilla.gnome.org/show_bug.cgi?id=679107

However, we are bit clueless on how that could be achieved. The only
idea I have is to know the number of passes (reboots) and typical disk
usage in each pass for "every" (we only do this for express installs
so we don't need this for all) OS and estimate the progress based on
that knowledge. If that idea is worth looking into, we'll need this
information in libosinfo and hence me cross-posting this to
virt-tools-list.

Other ideas welcome! Keep in mind that we don't need to be very
accurate, just accurate enough to not annoy the user.

-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124



From rjones at redhat.com  Wed Jul  4 21:09:46 2012
From: rjones at redhat.com (Richard W.M. Jones)
Date: Wed, 4 Jul 2012 22:09:46 +0100
Subject: [virt-tools-list] Tracking VM installations
In-Reply-To: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
References: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
Message-ID: <20120704210946.GA20861@amd.home.annexia.org>

On Wed, Jul 04, 2012 at 10:45:29PM +0300, Zeeshan Ali (Khattak) wrote:
> Hi,
> 
> We have a requirement in Boxes for showing progress bars on VMs under
> installation:
> 
> https://github.com/gnome-design-team/gnome-mockups/raw/master/boxes/boxes-install6.png
> https://bugzilla.gnome.org/show_bug.cgi?id=679107
> 
> However, we are bit clueless on how that could be achieved. The only
> idea I have is to know the number of passes (reboots) and typical disk
> usage in each pass for "every" (we only do this for express installs
> so we don't need this for all) OS and estimate the progress based on
> that knowledge. If that idea is worth looking into, we'll need this
> information in libosinfo and hence me cross-posting this to
> virt-tools-list.
> 
> Other ideas welcome! Keep in mind that we don't need to be very
> accurate, just accurate enough to not annoy the user.

It's a tricky one :-)

You can take a screenshot of the guest (assuming a visual
installation).  Oz does this.  Displaying an icon-sized screenshot
might be a surrogate for displaying a progress bar, since many
installers will display a progress bar themselves.
https://github.com/clalancette/oz/blob/74d99497840e3bade4d679ed94cf9a4628827bcc/oz/Guest.py#L624

You can use libguestfs (specifically, virt-df) to display the amount
of disk space used in the guest.  For many installers, disk space will
monotonically increase as it proceeds, although of course you've no
idea what the final disk usage will be.  You can turn this into a
pulse-mode progress bar.
http://libguestfs.org/virt-df.1.html

Oz also uses network and disk activity to tell if the installer is
making progress.  This again doesn't tell you when the installer is
going to finish, but it allows you to do a pulse-mode bar.
https://github.com/clalancette/oz/blob/74d99497840e3bade4d679ed94cf9a4628827bcc/oz/Guest.py#L569

Finally I've always thought it would be a good idea if guests
communicated information down to the host about progress, whether that
is progress booting or progress installing.  There is (on PC hardware)
even an I/O port reserved for this purpose (port 0x80)!  You'd have to
get buy-in and get it upstream in qemu and every installer out there.
ttp://stackoverflow.com/questions/6793899/what-does-the-0x80-port-address-connects/6796109#6796109

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://et.redhat.com/~rjones/virt-top



From zeeshanak at gnome.org  Thu Jul  5 01:49:49 2012
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Thu, 5 Jul 2012 04:49:49 +0300
Subject: [virt-tools-list] [libvirt]  Tracking VM installations
In-Reply-To: <20120704210946.GA20861@amd.home.annexia.org>
References: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
	<20120704210946.GA20861@amd.home.annexia.org>
Message-ID: <CAAa3hFMGHED3_ybtc3tu_bGynQLQqks1JNOyuPx+p+N41h6rqg@mail.gmail.com>

On Thu, Jul 5, 2012 at 12:09 AM, Richard W.M. Jones <rjones at redhat.com> wrote:
> On Wed, Jul 04, 2012 at 10:45:29PM +0300, Zeeshan Ali (Khattak) wrote:
>> Hi,

Hey Richard,

>> We have a requirement in Boxes for showing progress bars on VMs under
>> installation:
>>
>> https://github.com/gnome-design-team/gnome-mockups/raw/master/boxes/boxes-install6.png
>> https://bugzilla.gnome.org/show_bug.cgi?id=679107
>>
>> However, we are bit clueless on how that could be achieved. The only
>> idea I have is to know the number of passes (reboots) and typical disk
>> usage in each pass for "every" (we only do this for express installs
>> so we don't need this for all) OS and estimate the progress based on
>> that knowledge. If that idea is worth looking into, we'll need this
>> information in libosinfo and hence me cross-posting this to
>> virt-tools-list.
>>
>> Other ideas welcome! Keep in mind that we don't need to be very
>> accurate, just accurate enough to not annoy the user.
>
> It's a tricky one :-)

Indeed. :)

> You can take a screenshot of the guest (assuming a visual
> installation). ?Oz does this. ?Displaying an icon-sized screenshot
> might be a surrogate for displaying a progress bar, since many
> installers will display a progress bar themselves.
> https://github.com/clalancette/oz/blob/74d99497840e3bade4d679ed94cf9a4628827bcc/oz/Guest.py#L624

Well they usually don't do it from start to finish. E.g Anaconda show
separate progress bars at different stages but mainly at package
installations. Although our progressbar won't be very different from
that if we base it on disk allocation but a. ours will be a lot more
visible and it'll be there from the start.

> You can use libguestfs (specifically, virt-df) to display the amount
> of disk space used in the guest.

For target disk, we use libvirt's storage volume API and that provides
us API to get current allocation.

> ?For many installers, disk space will
> monotonically increase as it proceeds, although of course you've no
> idea what the final disk usage will be. ?You can turn this into a
> pulse-mode progress bar.
> http://libguestfs.org/virt-df.1.html

Well, that would be too easy then. :) If we fail to find any way to do
what we want, we can think about doing pulsemode progressbar instead.

> Oz also uses network and disk activity to tell if the installer is
> making progress.
> https://github.com/clalancette/oz/blob/74d99497840e3bade4d679ed94cf9a4628827bcc/oz/Guest.py#L569

Oh, this would be another thing we'd want to do as well: Inform the
user if installer is hung for too long and provide option to delete
the VM or ignore..

> Finally I've always thought it would be a good idea if guests
> communicated information down to the host about progress, whether that
> is progress booting or progress installing. ?There is (on PC hardware)
> even an I/O port reserved for this purpose (port 0x80)! ?You'd have to
> get buy-in and get it upstream in qemu and every installer out there.
> ttp://stackoverflow.com/questions/6793899/what-does-the-0x80-port-address-connects/6796109#6796109

That would indeed be awesome but it seems way too much work and fight
(not to mention, its impossible to fix this in proprietary OSs if they
don't already support this) for just one progress bar.

Thanks for your input btw!

-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124



From rjones at redhat.com  Thu Jul  5 07:54:20 2012
From: rjones at redhat.com (Richard W.M. Jones)
Date: Thu, 5 Jul 2012 08:54:20 +0100
Subject: [virt-tools-list] [libvirt]  Tracking VM installations
In-Reply-To: <CAAa3hFMGHED3_ybtc3tu_bGynQLQqks1JNOyuPx+p+N41h6rqg@mail.gmail.com>
References: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
	<20120704210946.GA20861@amd.home.annexia.org>
	<CAAa3hFMGHED3_ybtc3tu_bGynQLQqks1JNOyuPx+p+N41h6rqg@mail.gmail.com>
Message-ID: <20120705075420.GO2552@amd.home.annexia.org>

On Thu, Jul 05, 2012 at 04:49:49AM +0300, Zeeshan Ali (Khattak) wrote:
> On Thu, Jul 5, 2012 at 12:09 AM, Richard W.M. Jones <rjones at redhat.com> wrote:
> > You can use libguestfs (specifically, virt-df) to display the amount
> > of disk space used in the guest.
> 
> For target disk, we use libvirt's storage volume API and that provides
> us API to get current allocation.

libvirt storage API gives you a different number (blocks allocated) so
it won't work for VMs backed with LVs or preallocated files.  virt-df
looks inside the filesystem(s).  However a guest agent will give you
the most accurate figures when the guest is running, and I imagine a
guest agent may even be able to query the installer somehow.

> > Finally I've always thought it would be a good idea if guests
> > communicated information down to the host about progress, whether that
> > is progress booting or progress installing. ?There is (on PC hardware)
> > even an I/O port reserved for this purpose (port 0x80)! ?You'd have to
> > get buy-in and get it upstream in qemu and every installer out there.
> > http://stackoverflow.com/questions/6793899/what-does-the-0x80-port-address-connects/6796109#6796109
> 
> That would indeed be awesome but it seems way too much work and fight
> (not to mention, its impossible to fix this in proprietary OSs if they
> don't already support this) for just one progress bar.

I wouldn't necessarily discount this one.  It's a standard of sorts,
and useful in other scenarios that we very much care about,
eg. getting a libvirt event when a guest has started booting, finished
booting, is ready for login, etc.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
New in Fedora 11: Fedora Windows cross-compiler. Compile Windows
programs, test, and build Windows installers. Over 70 libraries supprt'd
http://fedoraproject.org/wiki/MinGW http://www.annexia.org/fedora_mingw



From berrange at redhat.com  Thu Jul  5 09:19:10 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 5 Jul 2012 10:19:10 +0100
Subject: [virt-tools-list] [libvirt]  Tracking VM installations
In-Reply-To: <20120705075420.GO2552@amd.home.annexia.org>
References: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
	<20120704210946.GA20861@amd.home.annexia.org>
	<CAAa3hFMGHED3_ybtc3tu_bGynQLQqks1JNOyuPx+p+N41h6rqg@mail.gmail.com>
	<20120705075420.GO2552@amd.home.annexia.org>
Message-ID: <20120705091910.GB13884@redhat.com>

On Thu, Jul 05, 2012 at 08:54:20AM +0100, Richard W.M. Jones wrote:
> On Thu, Jul 05, 2012 at 04:49:49AM +0300, Zeeshan Ali (Khattak) wrote:
> > On Thu, Jul 5, 2012 at 12:09 AM, Richard W.M. Jones <rjones at redhat.com> wrote:
> > > You can use libguestfs (specifically, virt-df) to display the amount
> > > of disk space used in the guest.
> > 
> > For target disk, we use libvirt's storage volume API and that provides
> > us API to get current allocation.
> 
> libvirt storage API gives you a different number (blocks allocated) so
> it won't work for VMs backed with LVs or preallocated files.  virt-df
> looks inside the filesystem(s).  However a guest agent will give you
> the most accurate figures when the guest is running, and I imagine a
> guest agent may even be able to query the installer somehow.
> 
> > > Finally I've always thought it would be a good idea if guests
> > > communicated information down to the host about progress, whether that
> > > is progress booting or progress installing. ?There is (on PC hardware)
> > > even an I/O port reserved for this purpose (port 0x80)! ?You'd have to
> > > get buy-in and get it upstream in qemu and every installer out there.
> > > http://stackoverflow.com/questions/6793899/what-does-the-0x80-port-address-connects/6796109#6796109
> > 
> > That would indeed be awesome but it seems way too much work and fight
> > (not to mention, its impossible to fix this in proprietary OSs if they
> > don't already support this) for just one progress bar.
> 
> I wouldn't necessarily discount this one.  It's a standard of sorts,
> and useful in other scenarios that we very much care about,
> eg. getting a libvirt event when a guest has started booting, finished
> booting, is ready for login, etc.

IMHO this is all a job for a guest agent to do using virtio-serial.

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Thu Jul  5 09:21:08 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 5 Jul 2012 10:21:08 +0100
Subject: [virt-tools-list] Tracking VM installations
In-Reply-To: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
References: <CAAa3hFPMaCAdvjf9qTkrxu1AgY7hdY5G1t8Y6VTtE4AeU872cA@mail.gmail.com>
Message-ID: <20120705092108.GC13884@redhat.com>

On Wed, Jul 04, 2012 at 10:45:29PM +0300, Zeeshan Ali (Khattak) wrote:
> Hi,
> 
> We have a requirement in Boxes for showing progress bars on VMs under
> installation:
> 
> https://github.com/gnome-design-team/gnome-mockups/raw/master/boxes/boxes-install6.png
> https://bugzilla.gnome.org/show_bug.cgi?id=679107
> 
> However, we are bit clueless on how that could be achieved. The only
> idea I have is to know the number of passes (reboots) and typical disk
> usage in each pass for "every" (we only do this for express installs
> so we don't need this for all) OS and estimate the progress based on
> that knowledge. If that idea is worth looking into, we'll need this
> information in libosinfo and hence me cross-posting this to
> virt-tools-list.
> 
> Other ideas welcome! Keep in mind that we don't need to be very
> accurate, just accurate enough to not annoy the user.

I don't see any general purpose way todo a progress feedback. The best
you can do is to show whether /something/ is happening (by looking at
disk & net I/O stats).  If you want actual progress feedback then you're
looking at modifying the guest OS init process / installer to send feedback
over a virtio-serial device directly, or indirectly via a guest agent.


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From crobinso at redhat.com  Mon Jul  9 15:08:12 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Mon, 09 Jul 2012 11:08:12 -0400
Subject: [virt-tools-list] ANNOUNCE: virt-manager 0.9.2 and virtinst 0.600.2
	released
Message-ID: <4FFAF3DC.3000301@redhat.com>

I'm happy to announce two new releases:

virt-manager 0.9.2: virt-manager is a desktop application for managing
KVM and Xen virtual machines via libvirt.

virtinst 0.600.2: virtinst is a collection of command line tools for
provisioning libvirt virtual machines, including virt-install and
virt-clone.

The releases can be downloaded from:

http://virt-manager.org/download.html

The direct download links are:

http://virt-manager.org/download/sources/virt-manager/virt-manager-0.9.2.tar.gz
http://virt-manager.org/download/sources/virtinst/virtinst-0.600.2.tar.gz

The virt-manager release includes:

- Convert to gtkbuilder: UI can now be edited with modern glade tool
- virt-manager no longer runs on RHEL5, but can manage a remote RHEL5
  host
- Option to configure spapr net and disk devices for pseries (Li Zhang)
- Many bug fixes and improvements

The virtinst release includes:

- virt-install: New --memballoon option (Eiichi Tsukata)
- virt-install: Improved qemu/kvm pseries support (Li Zhang, Qing Lin)
- virt-install: Support setting BIOS path using -boot loader=PATH
- Various bug fixes and improvements

Also, unless there are serious snags, the next virt-manager release will have
some major organizational changes: port to GTK3, virtinst will no longer be a
public API, merging virtinst and virt-manager git repos, maybe even converting
to python3. While this will facilitate a ton of future improvements, the
downside is that a much more recent distro will be required to run
virt-manager. More info forthcoming on virt-tools-list

Thanks,
Cole



From crobinso at redhat.com  Mon Jul  9 21:28:46 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Mon, 09 Jul 2012 17:28:46 -0400
Subject: [virt-tools-list] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <4FFAF3DC.3000301@redhat.com>
References: <4FFAF3DC.3000301@redhat.com>
Message-ID: <4FFB4D0E.7050409@redhat.com>

On 07/09/2012 11:08 AM, Cole Robinson wrote:
> I'm happy to announce two new releases:
> 
> virt-manager 0.9.2: virt-manager is a desktop application for managing
> KVM and Xen virtual machines via libvirt.
> 

The 0.9.2 tar.gz was busted for reasons I haven't quite figured out yet:
running ./configure was complaining about missing files. Either way I put up a
new 0.9.3 release which fixes things for me.

http://virt-manager.org/download.html
http://virt-manager.org/download/sources/virt-manager/virt-manager-0.9.3.tar.gz

Thanks,
Cole




From hutao at cn.fujitsu.com  Wed Jul 11 08:33:07 2012
From: hutao at cn.fujitsu.com (Hu Tao)
Date: Wed, 11 Jul 2012 16:33:07 +0800
Subject: [virt-tools-list] [libvirt] [PATCHv4] add-vcpu-usage
In-Reply-To: <1336553311-27311-1-git-send-email-hutao@cn.fujitsu.com>
References: <1336553311-27311-1-git-send-email-hutao@cn.fujitsu.com>
Message-ID: <20120711083307.GC7370@hut>

ping

On Wed, May 09, 2012 at 04:48:29PM +0800, Hu Tao wrote:
> show `vcpu usages' by `virt-top -1'
> 
> Before this patch, `virt-top -1' shows total cpu usages
> which euqal to `vcpu usages' + `hypervisor usages'. This
> patch adds another column for domains showing `vcpu
> usages'. An example is:
> 
> PHYCPU %CPU example_domain
>    0   10.4 10.4  0.8
>    1    1.6  1.6  1.4
>    2    2.6  2.6  2.6
>    3    0.0  0.0  0.1
> ---
>  virt-top/virt_top.ml |   72 ++++++++++++++++++++++++++++++++++++-------------
>  1 files changed, 53 insertions(+), 19 deletions(-)
> 
> diff --git a/virt-top/virt_top.ml b/virt-top/virt_top.ml
> index e2fe554..0dcb170 100644
> --- a/virt-top/virt_top.ml
> +++ b/virt-top/virt_top.ml
> @@ -448,6 +448,7 @@ let collect, clear_pcpu_display_data =
>  
>    (* Save pcpu_usages structures across redraws too (only for pCPU display). *)
>    let last_pcpu_usages = Hashtbl.create 13 in
> +  let last_vcpu_usages = Hashtbl.create 13 in
>  
>    let clear_pcpu_display_data () =
>      (* Clear out pcpu_usages used by PCPUDisplay display_mode
> @@ -652,12 +653,17 @@ let collect, clear_pcpu_display_data =
>  	      (try
>  		 let domid = rd.rd_domid in
>  		 let maplen = C.cpumaplen nr_pcpus in
> -		 let cpu_stats = D.get_cpu_stats rd.rd_dom nr_pcpus in
> -		 let rec find_usages_from_stats = function
> +		 let cpu_stats = D.get_cpu_stats rd.rd_dom false in
> +		 let rec find_cpu_usages = function
>  		   | ("cpu_time", D.TypedFieldUInt64 usages) :: _ -> usages
> -		   | _ :: params -> find_usages_from_stats params
> +		   | _ :: params -> find_cpu_usages params
>  		   | [] -> 0L in
> -		 let pcpu_usages = Array.map find_usages_from_stats cpu_stats in
> +		 let rec find_vcpu_usages = function
> +		   | ("vcpu_time", D.TypedFieldUInt64 usages) :: _ -> usages
> +		   | _ :: params -> find_vcpu_usages params
> +		   | [] -> 0L in
> +
> +		 let pcpu_usages = Array.map find_cpu_usages cpu_stats in
>  		 let maxinfo = rd.rd_info.D.nr_virt_cpu in
>  		 let nr_vcpus, vcpu_infos, cpumaps =
>  		   D.get_vcpus rd.rd_dom maxinfo maplen in
> @@ -669,11 +675,19 @@ let collect, clear_pcpu_display_data =
>  		 (* Update last_pcpu_usages. *)
>  		 Hashtbl.replace last_pcpu_usages domid pcpu_usages;
>  
> -		 (match prev_pcpu_usages with
> -		  | Some prev_pcpu_usages
> +                 (* vcpu usages *)
> +		 let vcpu_usages = Array.map find_vcpu_usages cpu_stats in
> +                 let prev_vcpu_usages =
> +                   try Some (Hashtbl.find last_vcpu_usages domid)
> +                   with Not_found -> None in
> +		 Hashtbl.replace last_vcpu_usages domid vcpu_usages;
> +
> +		 (match prev_pcpu_usages, prev_vcpu_usages with
> +		  | Some prev_pcpu_usages, Some prev_vcpu_usages
>  		      when Array.length prev_pcpu_usages = Array.length pcpu_usages ->
> -		      Some (domid, name, nr_vcpus, vcpu_infos, pcpu_usages,
> -			    prev_pcpu_usages, cpumaps, maplen)
> +                      Some (domid, name, nr_vcpus, vcpu_infos, pcpu_usages,
> +                      prev_pcpu_usages, vcpu_usages, prev_vcpu_usages,
> +                      cpumaps, maplen)
>  		  | _ -> None (* ignore missing / unequal length prev_vcpu_infos *)
>  		 );
>  	       with
> @@ -691,13 +705,24 @@ let collect, clear_pcpu_display_data =
>  
>  	List.iteri (
>  	  fun di (domid, name, nr_vcpus, vcpu_infos, pcpu_usages,
> -		  prev_pcpu_usages, cpumaps, maplen) ->
> +		  prev_pcpu_usages, vcpu_usages, prev_vcpu_usages,
> +                  cpumaps, maplen) ->
>  	    (* Which pCPUs can this dom run on? *)
>  	    for p = 0 to Array.length pcpu_usages - 1 do
>  	      pcpus.(p).(di) <- pcpu_usages.(p) -^ prev_pcpu_usages.(p)
> -	    done
> +            done
>  	) doms;
>  
> +        let vcpus = Array.make_matrix nr_pcpus nr_doms 0L in
> +        List.iteri (
> +	  fun di (domid, name, nr_vcpus, vcpu_infos, pcpu_usages,
> +		  prev_pcpu_usages, vcpu_usages, prev_vcpu_usages,
> +                  cpumaps, maplen) ->
> +	    for p = 0 to Array.length vcpu_usages - 1 do
> +	      vcpus.(p).(di) <- vcpu_usages.(p) -^ prev_vcpu_usages.(p)
> +            done
> +        ) doms;
> +
>  	(* Sum the CPU time used by each pCPU, for the %CPU column. *)
>  	let pcpus_cpu_time = Array.map (
>  	  fun row ->
> @@ -709,7 +734,7 @@ let collect, clear_pcpu_display_data =
>  	    Int64.to_float !cpu_time
>  	) pcpus in
>  
> -	Some (doms, pcpus, pcpus_cpu_time)
> +	Some (doms, pcpus, vcpus, pcpus_cpu_time)
>        ) else
>  	None in
>  
> @@ -913,7 +938,7 @@ let redraw =
>  	 loop domains_lineno doms
>  
>       | PCPUDisplay -> (*---------- Showing physical CPUs ----------*)
> -	 let doms, pcpus, pcpus_cpu_time =
> +	 let doms, pcpus, vcpus, pcpus_cpu_time =
>  	   match pcpu_display with
>  	   | Some p -> p
>  	   | None -> failwith "internal error: no pcpu_display data" in
> @@ -922,9 +947,9 @@ let redraw =
>  	 let dom_names =
>  	   String.concat "" (
>  	     List.map (
> -	       fun (_, name, _, _, _, _, _, _) ->
> +	       fun (_, name, _, _, _, _, _, _, _, _) ->
>  		 let len = String.length name in
> -		 let width = max (len+1) 7 in
> +		 let width = max (len+1) 12 in
>  		 pad width name
>  	     ) doms
>  	   ) in
> @@ -941,18 +966,27 @@ let redraw =
>  	     addch ' ';
>  
>  	     List.iteri (
> -	       fun di (domid, name, _, _, _, _, _, _) ->
> +	       fun di (domid, name, _, _, _, _, _, _, _, _) ->
>  		 let t = pcpus.(p).(di) in
> +                 let tv = vcpus.(p).(di) in
>  		 let len = String.length name in
> -		 let width = max (len+1) 7 in
> -		 let str =
> +		 let width = max (len+1) 12 in
> +		 let str_pcpu =
>  		   if t <= 0L then ""
>  		   else (
>  		     let t = Int64.to_float t in
>  		     let percent = 100. *. t /. total_cpu_per_pcpu in
> -		     sprintf "%s " (Show.percent percent)
> +		     sprintf "%s" (Show.percent percent)
>  		   ) in
> -		 addstr (pad width str);
> +                 let str_vcpu =
> +                    if tv <= 0L then ""
> +                    else (
> +                      let tv = Int64.to_float tv in
> +                      let percent = 100. *. tv /. total_cpu_per_pcpu in
> +                      sprintf "%s" (Show.percent percent)
> +                    ) in
> +                 let str = sprintf "%s %s" str_pcpu str_vcpu in
> +                 addstr (pad width str);
>  		 ()
>  	     ) doms
>  	 ) pcpus;
> -- 
> 1.7.1
> 
> --
> libvir-list mailing list
> libvir-list at redhat.com
> https://www.redhat.com/mailman/listinfo/libvir-list

-- 
Thanks,
Hu Tao



From eblake at redhat.com  Wed Jul 11 14:38:31 2012
From: eblake at redhat.com (Eric Blake)
Date: Wed, 11 Jul 2012 08:38:31 -0600
Subject: [virt-tools-list] [libvirt-users] I can't virt-clone into an
 existing LVM now (ERROR: clone onto existing storage volume is not
 supported:) - i could in previous lbvirt versions? (deployment scripts no
 longer work...)
In-Reply-To: <CAGfHYAS43gT4ZsrOrC5O5VSvXaNJrExriwOzynPZeXiLJbFgDg@mail.gmail.com>
References: <CAGfHYAS43gT4ZsrOrC5O5VSvXaNJrExriwOzynPZeXiLJbFgDg@mail.gmail.com>
Message-ID: <4FFD8FE7.8000104@redhat.com>

[adding virt-tools-list]

On 07/11/2012 04:36 AM, Morgan Cox wrote:
> Hi
> 
> To deploy KVM vms I use a KVM template + script - which is cloned, then
> cloned again and resized - using virt-resize.
> 
> i.e
> 
> ----------------------------------------------------
> 
> virt-clone --original debian6template-DONTSTART --name NAME.test -f
> /dev/vgpool/NAME.test
> 
> virsh vol-create-as vgpool NAME 8G
> 
> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
> 
> virt-resize --expand /dev/sda1 /dev/vgpool/NAME.test /dev/vgpool/NAME
> 
> ----------------------------------------------------
> 
> The reason I clone twice is because I know that virt-clone does things to
> prep it for a new install, also the whole point was to be able to specify a
> size to be clone into (hence the use of virt-resize)
> 
> This worked fine on our Ubuntu 11.10 server, however in Ubuntu 12.04 when I
> try to do the same I get the error
> 
> 'ERROR    Clone onto existing storage volume is not supported:
> '/dev/vgpool/test'

virt-clone is a separate package from libvirt, I'm hoping that someone
on the virt-tools-list has more insight into whether this was an
accidental regression.

> 
> If I can't virt-clone into an existing LVM I know I can clone once - then
> resize with LVM, then virt-resize into the partition (then virt-sysprep) -
> however the issue with that is that it is far more fiddly to mange...
> 
> Previously could just make a 20GB partition and use virt-resize to expand
> into - using the above method I would have to increase the LVM to the value
> I want minus the size of the template..
> i.e I have to specify - the  amount to increase by, rather than the final
> size I want.... The templates also change size (when I update them, etc)
> which would mean re-writting the script after every update.
> 
> Is there a way I can re-enable being able to virt-clone to an existing LVM
> partition ? As mentioned it works 100% fine in Ubuntu 10.04
> 


-- 
Eric Blake   eblake at redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 620 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120711/c15879cb/attachment.sig>

From morgancoxuk at gmail.com  Thu Jul 12 08:55:32 2012
From: morgancoxuk at gmail.com (Morgan Cox)
Date: Thu, 12 Jul 2012 09:55:32 +0100
Subject: [virt-tools-list] Ubuntu 12.04. I can't virt-clone into an existing
 LVM now (ERROR: clone onto existing storage volume is not supported:) - i
 could in previous lbvirt versions? (deployment scripts no longer work...)
Message-ID: <CAGfHYATtzTXfv6Qxi+Z=G6Pon9gWORGOA9zstvU4LVSzeBpLSA@mail.gmail.com>

Hi

To deploy KVM vms I use a KVM template + script - which is cloned, then
cloned again and resized - using virt-resize.

i.e

----------------------------------------------------

virt-clone --original debian6template-DONTSTART --name NAME.test -f
/dev/vgpool/NAME.test

virsh vol-create-as vgpool NAME 8G

virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME

virt-resize --expand /dev/sda1 /dev/vgpool/NAME.test /dev/vgpool/NAME

----------------------------------------------------

The reason I clone twice is because I know that virt-clone does things to
prep it for a new install, also the whole point was to be able to specify a
size to be clone into (hence the use of virt-resize)

This worked fine on our Ubuntu 11.10 server, however in Ubuntu 12.04 when I
try to do the same I get the error

'ERROR    Clone onto existing storage volume is not supported:
'/dev/vgpool/test'

If I can't virt-clone into an existing LVM I know I can clone once - then
resize with LVM, then virt-resize into the partition (then virt-sysprep) -
however the issue with that is that it is far more fiddly to mange...

Previously could just make a 20GB partition and use virt-resize to expand
into - using the above method I would have to increase the LVM to the value
I want minus the size of the template..
i.e I have to specify - the  amount to increase by, rather than the final
size I want.... The templates also change size (when I update them, etc)
which would mean re-writting the script after every update.

Is there a way I can re-enable being able to virt-clone to an existing LVM
partition ? As mentioned it works 100% fine in Ubuntu 10.04

Regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120712/61b10308/attachment.htm>

From morgancoxuk at gmail.com  Thu Jul 12 09:20:26 2012
From: morgancoxuk at gmail.com (Morgan Cox)
Date: Thu, 12 Jul 2012 10:20:26 +0100
Subject: [virt-tools-list] [v2v-qe] Fwd: Re: [libvirt-users] I can't
 virt-clone into an existing LVM now (ERROR: clone onto existing storage
 volume is not supported:)
In-Reply-To: <f233c079-dae0-493e-8e5d-30e920de124d@zmail18.collab.prod.int.phx2.redhat.com>
References: <4FFE351B.9000600@redhat.com>
	<f233c079-dae0-493e-8e5d-30e920de124d@zmail18.collab.prod.int.phx2.redhat.com>
Message-ID: <CAGfHYATswxPaES-3DYuaXs0PBBz861mcjcZ0Fi1LaCAsqJ8O0w@mail.gmail.com>

Hi

I'm pretty sure there is nothing wrong with the LVM partitioning - I
am using the same partitioning/methods on a older (ubuntu 10.04)
server.

Just a note it is a Ubuntu 12.04 server that has the issue...

-------------------------------
vgdisplay vgpool
  --- Volume group ---
  VG Name               vgpool
  System ID
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  55
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                6
  Open LV               3
  Max PV                0
  Cur PV                1
  Act PV                1
  VG Size               874.77 GiB
  PE Size               4.00 MiB
  Total PE              223941
  Alloc PE / Size       35789 / 139.80 GiB
  Free  PE / Size       188152 / 734.97 GiB
  VG UUID               uPK29Y-cygJ-irzN-4rX1-9MLk-WDQ5-3UWlRo

------------------------------

lvcreate -L 4000M -n lvol0 vgpool
  Logical volume "lvol0" created

mkfs -t ext4 -m 1 -v /dev/vgpool/lvol0

virt-clone -o centos6template-DONTSTART -n centos6.new -f /dev/vgpool/lvol0
ERROR    Clone onto existing storage volume is not supported:
'/dev/vgpool/lvol0'

Still getting the same thing

Any ideas how to troubleshoot this further ?

Regards




On 12 July 2012 07:04, Lei Cui <lcui at redhat.com> wrote:
> Hi Morgan,
> I could not reproduce your issue on RHEL6.3, would you please check your existing LVM is working fine?
> I doubt about your command:
>> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
> '/dev/vgpool/NAME' is it a logic volume?In this scenario, I think it should be the LVM device name following '-f'. As in man doc about virt-clone, here is the explanation of -f parameter
> "
>        -f DISKFILE, --file=DISKFILE
>          Path to the file, disk partition, or logical volume to use as the backing store for the new guest's virtual disk.
> "
> Here are my steps for clone a guest to existing LVM, the new guest could be created successfully and start normally. Please refer to
> 1. Create a LVM
> 1) fdisk /dev/sdb
> With a serious input in fdisk command to create a Linux LVM type partition(8e)
> 2) pvcreate /dev/sdb1
> 3) vgcreate -s 16M vg0 /dev/sdb1
> 4) lvcreate -L 60000M -n lvol0 vg0
> 5) mkfs -t ext3 -m 1 -v /dev/vg0/lvol0
> 6) vgdisplay vg0 (check the lvm status)
> 2. Clone a guest to existing LVM
> 1) virt-clone -o vm1 -n vm1-new -f /dev/vg0/lvol0
>
> after a while, pop up 'Clone vm1-new created successfully.
>
> thanks
> Lei
>
>
> ----- ???? -----
> ???: "yupzhang" <yupzhang at redhat.com>
> ???: v2v-qe-list at redhat.com
> ????: ???, 2012? 7 ? 12? ?? 10:23:23
> ??: [v2v-qe] Fwd: Re: [virt-tools-list] [libvirt-users] I can't virt-clone into an existing LVM now (ERROR: clone onto existing storage volume is not supported:) - i could in previous lbvirt versions? (deployment scripts no longer work...)
>
> Hi Lei,
>
> Please have a look this problem of virt-clone.
>
> Thanks
> Yuping
> -------- Original Message --------
> Subject:        Re: [virt-tools-list] [libvirt-users] I can't virt-clone into an existing LVM now (ERROR: clone onto existing storage volume is not supported:) - i could in previous lbvirt versions? (deployment scripts no longer work...)
> Date:   Wed, 11 Jul 2012 08:38:31 -0600
> From:   Eric Blake <eblake at redhat.com>
> Organization:   Red Hat
> To:     Morgan Cox <morgancoxuk at gmail.com>
> CC:     libvirt-users at redhat.com , virt-tools-list <virt-tools-list at redhat.com>
>
> [adding virt-tools-list]
>
> On 07/11/2012 04:36 AM, Morgan Cox wrote:
>> Hi
>>
>> To deploy KVM vms I use a KVM template + script - which is cloned, then
>> cloned again and resized - using virt-resize.
>>
>> i.e
>>
>> ----------------------------------------------------
>>
>> virt-clone --original debian6template-DONTSTART --name NAME.test -f
>> /dev/vgpool/NAME.test
>>
>> virsh vol-create-as vgpool NAME 8G
>>
>> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
>>
>> virt-resize --expand /dev/sda1 /dev/vgpool/NAME.test /dev/vgpool/NAME
>>
>> ----------------------------------------------------
>>
>> The reason I clone twice is because I know that virt-clone does things to
>> prep it for a new install, also the whole point was to be able to specify a
>> size to be clone into (hence the use of virt-resize)
>>
>> This worked fine on our Ubuntu 11.10 server, however in Ubuntu 12.04 when I
>> try to do the same I get the error
>>
>> 'ERROR    Clone onto existing storage volume is not supported:
>> '/dev/vgpool/test'
>
> virt-clone is a separate package from libvirt, I'm hoping that someone
> on the virt-tools-list has more insight into whether this was an
> accidental regression.
>
>>
>> If I can't virt-clone into an existing LVM I know I can clone once - then
>> resize with LVM, then virt-resize into the partition (then virt-sysprep) -
>> however the issue with that is that it is far more fiddly to mange...
>>
>> Previously could just make a 20GB partition and use virt-resize to expand
>> into - using the above method I would have to increase the LVM to the value
>> I want minus the size of the template..
>> i.e I have to specify - the  amount to increase by, rather than the final
>> size I want.... The templates also change size (when I update them, etc)
>> which would mean re-writting the script after every update.
>>
>> Is there a way I can re-enable being able to virt-clone to an existing LVM
>> partition ? As mentioned it works 100% fine in Ubuntu 10.04
>>
>
>
> --
> Eric Blake eblake at redhat.com +1-919-301-3266
> Libvirt virtualization library http://libvirt.org



From morgancoxuk at gmail.com  Thu Jul 12 10:01:28 2012
From: morgancoxuk at gmail.com (Morgan Cox)
Date: Thu, 12 Jul 2012 11:01:28 +0100
Subject: [virt-tools-list] [v2v-qe] Fwd: Re: [libvirt-users] I can't
 virt-clone into an existing LVM now (ERROR: clone onto existing storage
 volume is not supported:)
In-Reply-To: <CAGfHYATswxPaES-3DYuaXs0PBBz861mcjcZ0Fi1LaCAsqJ8O0w@mail.gmail.com>
References: <4FFE351B.9000600@redhat.com>
	<f233c079-dae0-493e-8e5d-30e920de124d@zmail18.collab.prod.int.phx2.redhat.com>
	<CAGfHYATswxPaES-3DYuaXs0PBBz861mcjcZ0Fi1LaCAsqJ8O0w@mail.gmail.com>
Message-ID: <CAGfHYATQsun04ZevTBip1gxU4pHrz_8ttMQtm4WrHr=Kkn2kcQ@mail.gmail.com>

Hi

I should also point out that I can clone

virt-clone --original centos6template-DONTSTART -n centos6.new -f
/dev/vgpool/centos6.new

If non existing LVM volume exists....

I can confirm that it 100% using LVM volumes I just can't clone into
existing LVM volumes (I can on Ubuntu 10.04)

Is this a config issue?

any help would be welcomed.

Regards



On 12 July 2012 10:20, Morgan Cox <morgancoxuk at gmail.com> wrote:

> Hi
>
> I'm pretty sure there is nothing wrong with the LVM partitioning - I
> am using the same partitioning/methods on a older (ubuntu 10.04)
> server.
>
> Just a note it is a Ubuntu 12.04 server that has the issue...
>
> -------------------------------
> vgdisplay vgpool
>   --- Volume group ---
>   VG Name               vgpool
>   System ID
>   Format                lvm2
>   Metadata Areas        1
>   Metadata Sequence No  55
>   VG Access             read/write
>   VG Status             resizable
>   MAX LV                0
>   Cur LV                6
>   Open LV               3
>   Max PV                0
>   Cur PV                1
>   Act PV                1
>   VG Size               874.77 GiB
>   PE Size               4.00 MiB
>   Total PE              223941
>   Alloc PE / Size       35789 / 139.80 GiB
>   Free  PE / Size       188152 / 734.97 GiB
>   VG UUID               uPK29Y-cygJ-irzN-4rX1-9MLk-WDQ5-3UWlRo
>
> ------------------------------
>
> lvcreate -L 4000M -n lvol0 vgpool
>   Logical volume "lvol0" created
>
> mkfs -t ext4 -m 1 -v /dev/vgpool/lvol0
>
> virt-clone -o centos6template-DONTSTART -n centos6.new -f /dev/vgpool/lvol0
> ERROR    Clone onto existing storage volume is not supported:
> '/dev/vgpool/lvol0'
>
> Still getting the same thing
>
> Any ideas how to troubleshoot this further ?
>
> Regards
>
>
>
>
> On 12 July 2012 07:04, Lei Cui <lcui at redhat.com> wrote:
> > Hi Morgan,
> > I could not reproduce your issue on RHEL6.3, would you please check your
> existing LVM is working fine?
> > I doubt about your command:
> >> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
> > '/dev/vgpool/NAME' is it a logic volume?In this scenario, I think it
> should be the LVM device name following '-f'. As in man doc about
> virt-clone, here is the explanation of -f parameter
> > "
> >        -f DISKFILE, --file=DISKFILE
> >          Path to the file, disk partition, or logical volume to use as
> the backing store for the new guest's virtual disk.
> > "
> > Here are my steps for clone a guest to existing LVM, the new guest could
> be created successfully and start normally. Please refer to
> > 1. Create a LVM
> > 1) fdisk /dev/sdb
> > With a serious input in fdisk command to create a Linux LVM type
> partition(8e)
> > 2) pvcreate /dev/sdb1
> > 3) vgcreate -s 16M vg0 /dev/sdb1
> > 4) lvcreate -L 60000M -n lvol0 vg0
> > 5) mkfs -t ext3 -m 1 -v /dev/vg0/lvol0
> > 6) vgdisplay vg0 (check the lvm status)
> > 2. Clone a guest to existing LVM
> > 1) virt-clone -o vm1 -n vm1-new -f /dev/vg0/lvol0
> >
> > after a while, pop up 'Clone vm1-new created successfully.
> >
> > thanks
> > Lei
> >
> >
> > ----- ???? -----
> > ???: "yupzhang" <yupzhang at redhat.com>
> > ???: v2v-qe-list at redhat.com
> > ????: ???, 2012? 7 ? 12? ?? 10:23:23
> > ??: [v2v-qe] Fwd: Re: [virt-tools-list] [libvirt-users] I can't
> virt-clone into an existing LVM now (ERROR: clone onto existing storage
> volume is not supported:) - i could in previous lbvirt versions?
> (deployment scripts no longer work...)
> >
> > Hi Lei,
> >
> > Please have a look this problem of virt-clone.
> >
> > Thanks
> > Yuping
> > -------- Original Message --------
> > Subject:        Re: [virt-tools-list] [libvirt-users] I can't virt-clone
> into an existing LVM now (ERROR: clone onto existing storage volume is not
> supported:) - i could in previous lbvirt versions? (deployment scripts no
> longer work...)
> > Date:   Wed, 11 Jul 2012 08:38:31 -0600
> > From:   Eric Blake <eblake at redhat.com>
> > Organization:   Red Hat
> > To:     Morgan Cox <morgancoxuk at gmail.com>
> > CC:     libvirt-users at redhat.com , virt-tools-list <
> virt-tools-list at redhat.com>
> >
> > [adding virt-tools-list]
> >
> > On 07/11/2012 04:36 AM, Morgan Cox wrote:
> >> Hi
> >>
> >> To deploy KVM vms I use a KVM template + script - which is cloned, then
> >> cloned again and resized - using virt-resize.
> >>
> >> i.e
> >>
> >> ----------------------------------------------------
> >>
> >> virt-clone --original debian6template-DONTSTART --name NAME.test -f
> >> /dev/vgpool/NAME.test
> >>
> >> virsh vol-create-as vgpool NAME 8G
> >>
> >> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
> >>
> >> virt-resize --expand /dev/sda1 /dev/vgpool/NAME.test /dev/vgpool/NAME
> >>
> >> ----------------------------------------------------
> >>
> >> The reason I clone twice is because I know that virt-clone does things
> to
> >> prep it for a new install, also the whole point was to be able to
> specify a
> >> size to be clone into (hence the use of virt-resize)
> >>
> >> This worked fine on our Ubuntu 11.10 server, however in Ubuntu 12.04
> when I
> >> try to do the same I get the error
> >>
> >> 'ERROR    Clone onto existing storage volume is not supported:
> >> '/dev/vgpool/test'
> >
> > virt-clone is a separate package from libvirt, I'm hoping that someone
> > on the virt-tools-list has more insight into whether this was an
> > accidental regression.
> >
> >>
> >> If I can't virt-clone into an existing LVM I know I can clone once -
> then
> >> resize with LVM, then virt-resize into the partition (then
> virt-sysprep) -
> >> however the issue with that is that it is far more fiddly to mange...
> >>
> >> Previously could just make a 20GB partition and use virt-resize to
> expand
> >> into - using the above method I would have to increase the LVM to the
> value
> >> I want minus the size of the template..
> >> i.e I have to specify - the  amount to increase by, rather than the
> final
> >> size I want.... The templates also change size (when I update them, etc)
> >> which would mean re-writting the script after every update.
> >>
> >> Is there a way I can re-enable being able to virt-clone to an existing
> LVM
> >> partition ? As mentioned it works 100% fine in Ubuntu 10.04
> >>
> >
> >
> > --
> > Eric Blake eblake at redhat.com +1-919-301-3266
> > Libvirt virtualization library http://libvirt.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120712/b9bf940a/attachment.htm>

From rjones at redhat.com  Thu Jul 12 22:02:15 2012
From: rjones at redhat.com (Richard W.M. Jones)
Date: Thu, 12 Jul 2012 23:02:15 +0100
Subject: [virt-tools-list] [libvirt-users] I can't virt-clone into an
 existing LVM now (ERROR: clone onto existing storage volume is not
 supported:) - i could in previous lbvirt versions? (deployment scripts no
 longer work...)
In-Reply-To: <4FFD8FE7.8000104@redhat.com>
References: <CAGfHYAS43gT4ZsrOrC5O5VSvXaNJrExriwOzynPZeXiLJbFgDg@mail.gmail.com>
	<4FFD8FE7.8000104@redhat.com>
Message-ID: <20120712220215.GA11400@amd.home.annexia.org>

On Wed, Jul 11, 2012 at 08:38:31AM -0600, Eric Blake wrote:
> On 07/11/2012 04:36 AM, Morgan Cox wrote:
> > 'ERROR    Clone onto existing storage volume is not supported:
> > '/dev/vgpool/test'
> 
> virt-clone is a separate package from libvirt, I'm hoping that someone
> on the virt-tools-list has more insight into whether this was an
> accidental regression.

I think an important thing to point out is that virt-clone is just a
fancy way of copying (ie. dd) the data, plus it makes a handful of
changes to the libvirt config.  You can do most of this by alternate
means.

> > Previously could just make a 20GB partition and use virt-resize to
> > expand into - using the above method I would have to increase the
> > LVM to the value I want minus the size of the template..  i.e I
> > have to specify - the amount to increase by, rather than the final
> > size I want.... The templates also change size (when I update
> > them, etc) which would mean re-writting the script after every
> > update.

So I'm not really clear why you don't just use virt-resize as before.
There's not much point in doing virt-clone + virt-resize, since
virt-resize itself does a copy.

Maybe you should explain exactly what overall goal you're trying to
achieve, instead of how you're currently trying to do it.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
New in Fedora 11: Fedora Windows cross-compiler. Compile Windows
programs, test, and build Windows installers. Over 70 libraries supprt'd
http://fedoraproject.org/wiki/MinGW http://www.annexia.org/fedora_mingw



From lcui at redhat.com  Thu Jul 12 06:04:40 2012
From: lcui at redhat.com (Lei Cui)
Date: Thu, 12 Jul 2012 02:04:40 -0400 (EDT)
Subject: [virt-tools-list] [v2v-qe] Fwd: Re: [libvirt-users] I can't
	virt-clone into an existing LVM now (ERROR: clone onto
	existing storage volume is not supported:)
In-Reply-To: <4FFE351B.9000600@redhat.com>
Message-ID: <f233c079-dae0-493e-8e5d-30e920de124d@zmail18.collab.prod.int.phx2.redhat.com>

Hi Morgan,
I could not reproduce your issue on RHEL6.3, would you please check your existing LVM is working fine?
I doubt about your command:
> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
'/dev/vgpool/NAME' is it a logic volume?In this scenario, I think it should be the LVM device name following '-f'. As in man doc about virt-clone, here is the explanation of -f parameter
"
       -f DISKFILE, --file=DISKFILE
         Path to the file, disk partition, or logical volume to use as the backing store for the new guest's virtual disk.
"
Here are my steps for clone a guest to existing LVM, the new guest could be created successfully and start normally. Please refer to
1. Create a LVM
1) fdisk /dev/sdb
With a serious input in fdisk command to create a Linux LVM type partition(8e)
2) pvcreate /dev/sdb1
3) vgcreate -s 16M vg0 /dev/sdb1
4) lvcreate -L 60000M -n lvol0 vg0
5) mkfs -t ext3 -m 1 -v /dev/vg0/lvol0
6) vgdisplay vg0 (check the lvm status)
2. Clone a guest to existing LVM
1) virt-clone -o vm1 -n vm1-new -f /dev/vg0/lvol0

after a while, pop up 'Clone vm1-new created successfully.

thanks
Lei


----- ???? -----
???: "yupzhang" <yupzhang at redhat.com>
???: v2v-qe-list at redhat.com
????: ???, 2012? 7 ? 12? ?? 10:23:23
??: [v2v-qe] Fwd: Re: [virt-tools-list] [libvirt-users] I can't virt-clone into an existing LVM now (ERROR: clone onto existing storage volume is not supported:) - i could in previous lbvirt versions? (deployment scripts no longer work...)

Hi Lei, 

Please have a look this problem of virt-clone. 

Thanks 
Yuping 
-------- Original Message -------- 
Subject: 	Re: [virt-tools-list] [libvirt-users] I can't virt-clone into an existing LVM now (ERROR: clone onto existing storage volume is not supported:) - i could in previous lbvirt versions? (deployment scripts no longer work...) 
Date: 	Wed, 11 Jul 2012 08:38:31 -0600 
From: 	Eric Blake <eblake at redhat.com> 
Organization: 	Red Hat 
To: 	Morgan Cox <morgancoxuk at gmail.com> 
CC: 	libvirt-users at redhat.com , virt-tools-list <virt-tools-list at redhat.com> 

[adding virt-tools-list]

On 07/11/2012 04:36 AM, Morgan Cox wrote:
> Hi
> 
> To deploy KVM vms I use a KVM template + script - which is cloned, then
> cloned again and resized - using virt-resize.
> 
> i.e
> 
> ----------------------------------------------------
> 
> virt-clone --original debian6template-DONTSTART --name NAME.test -f
> /dev/vgpool/NAME.test
> 
> virsh vol-create-as vgpool NAME 8G
> 
> virt-clone --original NAME.test --name NAME -f /dev/vgpool/NAME
> 
> virt-resize --expand /dev/sda1 /dev/vgpool/NAME.test /dev/vgpool/NAME
> 
> ----------------------------------------------------
> 
> The reason I clone twice is because I know that virt-clone does things to
> prep it for a new install, also the whole point was to be able to specify a
> size to be clone into (hence the use of virt-resize)
> 
> This worked fine on our Ubuntu 11.10 server, however in Ubuntu 12.04 when I
> try to do the same I get the error
> 
> 'ERROR    Clone onto existing storage volume is not supported:
> '/dev/vgpool/test'

virt-clone is a separate package from libvirt, I'm hoping that someone
on the virt-tools-list has more insight into whether this was an
accidental regression.

> 
> If I can't virt-clone into an existing LVM I know I can clone once - then
> resize with LVM, then virt-resize into the partition (then virt-sysprep) -
> however the issue with that is that it is far more fiddly to mange...
> 
> Previously could just make a 20GB partition and use virt-resize to expand
> into - using the above method I would have to increase the LVM to the value
> I want minus the size of the template..
> i.e I have to specify - the  amount to increase by, rather than the final
> size I want.... The templates also change size (when I update them, etc)
> which would mean re-writting the script after every update.
> 
> Is there a way I can re-enable being able to virt-clone to an existing LVM
> partition ? As mentioned it works 100% fine in Ubuntu 10.04
> 


-- 
Eric Blake eblake at redhat.com +1-919-301-3266
Libvirt virtualization library http://libvirt.org 



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:40 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:40 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 05/19] Destroy dialog
	immediately after run
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-6-git-send-email-marcandre.lureau@gmail.com>

If the parent is already destroyed, it looks like the dialog is
destroyed too. This avoids a crash when calling app_quit().
---
 src/virt-viewer-app.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 40d1ff7..1447ce7 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -283,6 +283,7 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
                                         GTK_BUTTONS_OK_CANCEL,
                                         _("This is the last visible display. Do you want to quit?"));
             gint result = gtk_dialog_run (GTK_DIALOG (dialog));
+            gtk_widget_destroy(dialog);
             switch (result) {
             case GTK_RESPONSE_OK:
                 virt_viewer_app_quit(self);
@@ -290,7 +291,6 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
             default:
                 break;
             }
-            gtk_widget_destroy(dialog);
             return FALSE;
         } else {
             virt_viewer_app_quit(self);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:45 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:45 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 10/19] Use
	virt_viewer_connect_object() for display
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-11-git-send-email-marcandre.lureau@gmail.com>

When display is released, detach signal automatically
---
 src/virt-viewer-app.c    |    4 ++--
 src/virt-viewer-window.c |   24 ++++++++++++------------
 2 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 758cfe9..e5c674a 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -658,8 +658,8 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
     }
 
     virt_viewer_window_set_display(window, display);
-    g_signal_connect(display, "notify::show-hint",
-                     G_CALLBACK(display_show_hint), window);
+    virt_viewer_signal_connect_object(display, "notify::show-hint",
+                                      G_CALLBACK(display_show_hint), window, 0);
     g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */
 }
 
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 4887e28..3f7cbbe 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1042,18 +1042,18 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
         gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
         gtk_widget_show_all(GTK_WIDGET(display));
 
-        g_signal_connect(display, "display-pointer-grab",
-                         G_CALLBACK(virt_viewer_window_pointer_grab), self);
-        g_signal_connect(display, "display-pointer-ungrab",
-                         G_CALLBACK(virt_viewer_window_pointer_ungrab), self);
-        g_signal_connect(display, "display-keyboard-grab",
-                         G_CALLBACK(virt_viewer_window_keyboard_grab), self);
-        g_signal_connect(display, "display-keyboard-ungrab",
-                         G_CALLBACK(virt_viewer_window_keyboard_ungrab), self);
-        g_signal_connect(display, "display-desktop-resize",
-                         G_CALLBACK(virt_viewer_window_desktop_resize), self);
-        g_signal_connect(display, "notify::show-hint",
-                         G_CALLBACK(display_show_hint), self);
+        virt_viewer_signal_connect_object(display, "display-pointer-grab",
+                                          G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-pointer-ungrab",
+                                          G_CALLBACK(virt_viewer_window_pointer_ungrab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-keyboard-grab",
+                                          G_CALLBACK(virt_viewer_window_keyboard_grab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-keyboard-ungrab",
+                                          G_CALLBACK(virt_viewer_window_keyboard_ungrab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-desktop-resize",
+                                          G_CALLBACK(virt_viewer_window_desktop_resize), self, 0);
+        virt_viewer_signal_connect_object(display, "notify::show-hint",
+                                          G_CALLBACK(display_show_hint), self, 0);
     }
 }
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:47 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:47 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 12/19] Prevent from adding the
	same display several time in the session
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-13-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-session.c |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
index 18b6922..9249a1f 100644
--- a/src/virt-viewer-session.c
+++ b/src/virt-viewer-session.c
@@ -282,6 +282,9 @@ virt_viewer_session_new(void)
 void virt_viewer_session_add_display(VirtViewerSession *session,
                                      VirtViewerDisplay *display)
 {
+    if (g_list_find(session->priv->displays, display))
+        return;
+
     session->priv->displays = g_list_append(session->priv->displays, display);
     g_object_ref(display);
     g_signal_emit_by_name(session, "session-display-added", display);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:54 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:54 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 19/19] Show status page by
	default if display not ready
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-20-git-send-email-marcandre.lureau@gmail.com>

GtkNotebook will use the currently visible widget as default page.
If we don't show status widget before we append the display, the
current page will be on display. Quoting Gtk+ documentation:

"Note that due to historical reasons, GtkNotebook refuses to switch to
a page unless the child widget is visible. Therefore, it is
recommended to show child widgets before adding them to a notebook."
---
 src/virt-viewer-notebook.c |    1 +
 src/virt-viewer-window.c   |    6 +++++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/src/virt-viewer-notebook.c b/src/virt-viewer-notebook.c
index 09e8ec4..cc7136b 100644
--- a/src/virt-viewer-notebook.c
+++ b/src/virt-viewer-notebook.c
@@ -86,6 +86,7 @@ virt_viewer_notebook_init (VirtViewerNotebook *self)
     priv->status = gtk_label_new("");
     gtk_notebook_set_show_tabs(GTK_NOTEBOOK(self), FALSE);
     gtk_notebook_set_show_border(GTK_NOTEBOOK(self), FALSE);
+    gtk_widget_show_all(priv->status);
     gtk_notebook_append_page(GTK_NOTEBOOK(self), priv->status, NULL);
     gdk_color_parse("white", &color);
     gtk_widget_modify_fg(priv->status, GTK_STATE_NORMAL, &color);
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 4bc427a..7c6e41f 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1040,8 +1040,12 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
         virt_viewer_display_set_zoom_level(VIRT_VIEWER_DISPLAY(priv->display), priv->zoomlevel);
         virt_viewer_display_set_auto_resize(VIRT_VIEWER_DISPLAY(priv->display), priv->auto_resize);
 
-        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
         gtk_widget_show_all(GTK_WIDGET(display));
+        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
+        /* switch back to non-display if not ready */
+        if (!(virt_viewer_display_get_show_hint(display) &
+              VIRT_VIEWER_DISPLAY_SHOW_HINT_READY))
+            gtk_notebook_set_current_page(GTK_NOTEBOOK(priv->notebook), 0);
 
         virt_viewer_signal_connect_object(display, "display-pointer-grab",
                                           G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:41 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:41 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn display:show-hint
	into flags type
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c     |    2 +-
 src/virt-viewer-display.c |   26 ++++++++++++++++----------
 src/virt-viewer-display.h |   11 ++++++-----
 src/virt-viewer-window.c  |    5 ++++-
 4 files changed, 27 insertions(+), 17 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 1447ce7..b93a978 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -619,7 +619,7 @@ display_show_hint(VirtViewerDisplay *display,
                  "show-hint", &hint,
                  NULL);
 
-    if (hint == VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE) {
+    if (hint == 0) {
         if (win != self->priv->main_window &&
             g_getenv("VIRT_VIEWER_HIDE"))
             virt_viewer_window_hide(win);
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 6d6052b..43ec45e 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -41,7 +41,7 @@ struct _VirtViewerDisplayPrivate
     guint zoom_level;
     gboolean zoom;
     gint nth_display;
-    gint show_hint;
+    guint show_hint;
     VirtViewerSession *session;
     gboolean auto_resize;
 };
@@ -151,13 +151,12 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
 
     g_object_class_install_property(object_class,
                                     PROP_SHOW_HINT,
-                                    g_param_spec_int("show-hint",
-                                                     "Show hint",
-                                                     "Show state hint",
-                                                     0,
-                                                     G_MAXINT32,
-                                                     0,
-                                                     G_PARAM_READABLE));
+                                    g_param_spec_flags("show-hint",
+                                                       "Show hint",
+                                                       "Show state hint",
+                                                       VIRT_VIEWER_TYPE_DISPLAY_SHOW_HINT_FLAGS,
+                                                       0,
+                                                       G_PARAM_READABLE));
 
     g_object_class_install_property(object_class,
                                     PROP_SESSION,
@@ -298,7 +297,7 @@ virt_viewer_display_get_property(GObject *object,
         g_value_set_int(value, priv->nth_display);
         break;
     case PROP_SHOW_HINT:
-        g_value_set_int(value, priv->show_hint);
+        g_value_set_flags(value, priv->show_hint);
         break;
     case PROP_SESSION:
         g_value_set_object(value, virt_viewer_display_get_session(display));
@@ -530,7 +529,14 @@ GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display)
     return VIRT_VIEWER_DISPLAY_GET_CLASS(display)->get_pixbuf(display);
 }
 
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, gint hint)
+guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
+{
+    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), 0);
+
+    return self->priv->show_hint;
+}
+
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
 {
     VirtViewerDisplayPrivate *priv;
     g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index ffbaf0e..225080d 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -25,6 +25,7 @@
 #define _VIRT_VIEWER_DISPLAY_H
 
 #include <gtk/gtk.h>
+#include "virt-viewer-enums.h"
 
 G_BEGIN_DECLS
 
@@ -54,10 +55,9 @@ typedef struct _VirtViewerDisplayPrivate VirtViewerDisplayPrivate;
 
 typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
 
-enum {
-    VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE = 0,
-    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY,
-};
+typedef enum {
+    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
+} VirtViewerDisplayShowHintFlags;
 
 /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
 struct _VirtViewerDisplay {
@@ -108,7 +108,8 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
 void virt_viewer_display_send_keys(VirtViewerDisplay *display,
                                    const guint *keyvals, int nkeyvals);
 GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, gint hint);
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
+guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
 void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
 gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 7b582a0..c873ebc 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1002,9 +1002,12 @@ display_show_hint(VirtViewerDisplay *display,
                   GParamSpec *pspec G_GNUC_UNUSED,
                   VirtViewerWindow *self)
 {
-    gboolean hint;
+    guint hint;
 
     g_object_get(display, "show-hint", &hint, NULL);
+
+    hint = (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
+
     gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-send")), hint);
     gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-file-screenshot")), hint);
     gtk_widget_set_sensitive(self->priv->toolbar_send_key, hint);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:49 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:49 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 14/19] Use SpiceDisplay:ready
	property instead of channel mark
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-15-git-send-email-marcandre.lureau@gmail.com>

The display can now check several conditions before the display can be
shown, use that instead of display mark, which was not high-level
enough.
---
 src/virt-viewer-display-spice.c |   38 +++++++++++++++-----------------------
 1 file changed, 15 insertions(+), 23 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 101abc5..53430dd 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -122,31 +122,25 @@ virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display)
 }
 
 static void
-display_mark(SpiceChannel *channel G_GNUC_UNUSED,
-             gint mark,
-             VirtViewerDisplay *display)
+display_ready(GObject *display,
+              GParamSpec *pspec G_GNUC_UNUSED,
+              VirtViewerDisplay *self)
 {
-    DEBUG_LOG("Toggle monitor visibility %p %d", channel, mark);
+    guint hint;
+    gboolean ready;
 
-    virt_viewer_display_set_show_hint(display, mark);
-}
+    g_object_get(display, "ready", &ready, NULL);
+    DEBUG_LOG("display %p ready:%d", self, ready);
 
-static void
-primary_create(SpiceChannel *channel G_GNUC_UNUSED,
-               gint format G_GNUC_UNUSED,
-               gint width,
-               gint height,
-               gint stride G_GNUC_UNUSED,
-               gint shmid G_GNUC_UNUSED,
-               gpointer imgdata G_GNUC_UNUSED,
-               VirtViewerDisplay *display)
-{
-    DEBUG_LOG("spice desktop resize %dx%d", width, height);
+    hint = virt_viewer_display_get_show_hint(self);
+    if (ready)
+        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
+    else
+        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
 
-    virt_viewer_display_set_desktop_size(display, width, height);
+    virt_viewer_display_set_show_hint(self, hint);
 }
 
-
 static void
 virt_viewer_display_spice_keyboard_grab(SpiceDisplay *display G_GNUC_UNUSED,
                                         int grabbed,
@@ -239,10 +233,8 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
     self->priv->display = spice_display_new(s, channelid);
     g_object_unref(s);
 
-    virt_viewer_signal_connect_object(channel, "display-primary-create",
-                                      G_CALLBACK(primary_create), self, 0);
-    virt_viewer_signal_connect_object(channel, "display-mark",
-                                      G_CALLBACK(display_mark), self, 0);
+    virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
+                                      G_CALLBACK(display_ready), self, 0);
 
     gtk_container_add(GTK_CONTAINER(self), g_object_ref(self->priv->display));
     gtk_widget_show(GTK_WIDGET(self->priv->display));
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:39 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:39 +0200
Subject: [virt-tools-list] =?utf-8?q?=5BPATCH_virt-viewer_04/19=5D_build?=
	=?utf-8?q?=3A_generate_enums_type_boilerplate?=
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-5-git-send-email-marcandre.lureau@gmail.com>

---
 configure.ac                      |    3 ++
 src/Makefile.am                   |   23 ++++++++++++--
 src/virt-viewer-enums.c.etemplate |   60 +++++++++++++++++++++++++++++++++++++
 src/virt-viewer-enums.h.etemplate |   41 +++++++++++++++++++++++++
 4 files changed, 124 insertions(+), 3 deletions(-)
 create mode 100644 src/virt-viewer-enums.c.etemplate
 create mode 100644 src/virt-viewer-enums.h.etemplate

diff --git a/configure.ac b/configure.ac
index a2471f4..548b945 100644
--- a/configure.ac
+++ b/configure.ac
@@ -72,6 +72,9 @@ AC_DEFINE_UNQUOTED([GETTEXT_PACKAGE],"$GETTEXT_PACKAGE", [GETTEXT package name])
 AM_GLIB_GNU_GETTEXT
 IT_PROG_INTLTOOL([0.35.0])
 
+PKG_PROG_PKG_CONFIG
+GLIB_MKENUMS=`$PKG_CONFIG --variable=glib_mkenums glib-2.0`
+AC_SUBST(GLIB_MKENUMS)
 
 PKG_CHECK_MODULES(GLIB2, glib-2.0 >= $GLIB2_REQUIRED gthread-2.0 gmodule-export-2.0)
 PKG_CHECK_MODULES(LIBXML2, libxml-2.0 >= $LIBXML2_REQUIRED)
diff --git a/src/Makefile.am b/src/Makefile.am
index d99b043..5daa98a 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -10,11 +10,28 @@ builderxml_DATA =				\
 	virt-viewer-auth.xml			\
 	$(NULL)
 
-EXTRA_DIST = $(builderxml_DATA) \
-	gbinding.c \
-	gbinding.h
+EXTRA_DIST =					\
+	$(builderxml_DATA)			\
+	virt-viewer-enums.c.etemplate		\
+	virt-viewer-enums.h.etemplate		\
+	gbinding.c				\
+	gbinding.h				\
+	$(NULL)
+
+ENUMS_FILES =					\
+	virt-viewer-display.h			\
+	$(NULL)
+
+BUILT_SOURCES =					\
+	virt-viewer-enums.h			\
+	virt-viewer-enums.c			\
+	$(NULL)
+
+$(BUILT_SOURCES): %: %.etemplate $(ENUMS_FILES)
+	$(AM_V_GEN)$(GLIB_MKENUMS) --template $^ | sed -e 's/VIRT_TYPE_VIEWER/VIRT_VIEWER_TYPE/' > $@
 
 COMMON_SOURCES =					\
+	$(BUILT_SOURCES)				\
 	virt-gtk-compat.h				\
 	virt-viewer-util.h virt-viewer-util.c		\
 	virt-viewer-auth.h virt-viewer-auth.c		\
diff --git a/src/virt-viewer-enums.c.etemplate b/src/virt-viewer-enums.c.etemplate
new file mode 100644
index 0000000..6f9925f
--- /dev/null
+++ b/src/virt-viewer-enums.c.etemplate
@@ -0,0 +1,60 @@
+/*** BEGIN file-header ***/
+/*
+ * Virt Viewer: A virtual machine console viewer
+ *
+ * Copyright (C) 2007-2012 Red Hat, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Author: Marc-Andr? Lureau <marcandre.lureau at redhat.com>
+ */
+
+#include "virt-viewer-enums.h"
+
+/*** END file-header ***/
+
+/*** BEGIN file-production ***/
+#include "@filename@"
+/*** END file-production ***/
+
+
+/*** BEGIN value-header ***/
+
+GType
+ at enum_name@_get_type (void)
+{
+  static volatile gsize g_define_type_id__volatile = 0;
+
+  if (g_once_init_enter (&g_define_type_id__volatile))
+    {
+      static const G at Type@Value values[] = {
+/*** END value-header ***/
+
+/*** BEGIN value-production ***/
+        { @VALUENAME@, "@VALUENAME@", "@valuenick@" },
+/*** END value-production ***/
+
+/*** BEGIN value-tail ***/
+        { 0, NULL, NULL }
+      };
+      GType g_define_type_id =
+        g_ at type@_register_static (g_intern_static_string ("@EnumName@"), values);
+      g_once_init_leave (&g_define_type_id__volatile, g_define_type_id);
+    }
+
+  return g_define_type_id__volatile;
+}
+
+/*** END value-tail ***/
diff --git a/src/virt-viewer-enums.h.etemplate b/src/virt-viewer-enums.h.etemplate
new file mode 100644
index 0000000..f33158d
--- /dev/null
+++ b/src/virt-viewer-enums.h.etemplate
@@ -0,0 +1,41 @@
+/*** BEGIN file-header ***/
+/*
+ * Virt Viewer: A virtual machine console viewer
+ *
+ * Copyright (C) 2007-2012 Red Hat, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Author: Marc-Andr? Lureau <marcandre.lureau at redhat.com>
+ */
+
+#ifndef VIRT_VIEWER_ENUMS_H
+#define VIRT_VIEWER_ENUMS_H
+
+#include <glib-object.h>
+
+G_BEGIN_DECLS
+/*** END file-header ***/
+
+/*** BEGIN value-header ***/
+GType @enum_name at _get_type (void) G_GNUC_CONST;
+#define @ENUMPREFIX at _TYPE_@ENUMSHORT@ (@enum_name at _get_type ())
+/*** END value-header ***/
+
+/*** BEGIN file-tail ***/
+G_END_DECLS
+
+#endif
+/*** END file-tail ***/
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:52 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:52 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 17/19] Make display submenu
	insensitive if they can't be modified
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-18-git-send-email-marcandre.lureau@gmail.com>

Only when the agent is running may a display be enabled/disabled.
---
 src/virt-viewer-app.c           |   28 ++++++++++++++++++++++++++-
 src/virt-viewer-display-spice.c |   40 ++++++++++++++++++++++++++++++++-------
 src/virt-viewer-display.c       |   23 ++++++++++++++++++++++
 src/virt-viewer-display.h       |    2 ++
 src/virt-viewer-session-spice.c |   18 ++++++++++++++----
 src/virt-viewer-session.c       |   10 ++++++++++
 src/virt-viewer-session.h       |    1 +
 src/virt-viewer-window.c        |   13 +++++++++++--
 src/virt-viewer-window.h        |    1 +
 9 files changed, 122 insertions(+), 14 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index f590641..def52c5 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -661,6 +661,7 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
     }
 
     virt_viewer_window_set_display(window, display);
+    virt_viewer_app_update_menu_displays(self);
     virt_viewer_signal_connect_object(display, "notify::show-hint",
                                       G_CALLBACK(display_show_hint), window, 0);
     g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */
@@ -684,6 +685,13 @@ virt_viewer_app_display_removed(VirtViewerSession *session G_GNUC_UNUSED,
         virt_viewer_app_remove_nth_window(self, nth);
 }
 
+static void
+virt_viewer_app_display_updated(VirtViewerSession *session G_GNUC_UNUSED,
+                                VirtViewerApp *self)
+{
+    virt_viewer_app_update_menu_displays(self);
+}
+
 int
 virt_viewer_app_create_session(VirtViewerApp *self, const gchar *type)
 {
@@ -734,6 +742,8 @@ virt_viewer_app_create_session(VirtViewerApp *self, const gchar *type)
                      G_CALLBACK(virt_viewer_app_display_added), self);
     g_signal_connect(priv->session, "session-display-removed",
                      G_CALLBACK(virt_viewer_app_display_removed), self);
+    g_signal_connect(priv->session, "session-display-updated",
+                     G_CALLBACK(virt_viewer_app_display_updated), self);
 
     g_signal_connect(priv->session, "session-cut-text",
                      G_CALLBACK(virt_viewer_app_server_cut_text), self);
@@ -1634,15 +1644,31 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
     while (tmp) {
         int *nth = tmp->data;
         VirtViewerWindow *vwin = VIRT_VIEWER_WINDOW(g_hash_table_lookup(self->priv->windows, nth));
+        VirtViewerDisplay *display = virt_viewer_window_get_display(vwin);
         GtkWidget *item;
-        gboolean visible;
+        gboolean visible, sensitive = FALSE;
         gchar *label;
 
         label = g_strdup_printf(_("Display %d"), *nth + 1);
         item = gtk_check_menu_item_new_with_label(label);
         g_free(label);
+
         visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
         gtk_check_menu_item_set_active(GTK_CHECK_MENU_ITEM(item), visible);
+
+        if (display) {
+            guint hint = virt_viewer_display_get_show_hint(display);
+
+            if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY)
+                sensitive = TRUE;
+
+            if ((hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED) &&
+                virt_viewer_display_get_selectable(display))
+                sensitive = TRUE;
+        }
+
+        gtk_widget_set_sensitive(item, sensitive);
+
         g_signal_connect(G_OBJECT(item),
                          "toggled", G_CALLBACK(menu_display_visible_toggled_cb), vwin);
         gtk_menu_shell_append(submenu, item);
diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index ca74c1a..a91d73e 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -47,6 +47,7 @@ static void virt_viewer_display_spice_send_keys(VirtViewerDisplay *display,
 static GdkPixbuf *virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display);
 static void virt_viewer_display_spice_release_cursor(VirtViewerDisplay *display);
 static void virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED);
+static gboolean virt_viewer_display_spice_selectable(VirtViewerDisplay *display);
 
 static void
 virt_viewer_display_spice_finalize(GObject *obj)
@@ -70,23 +71,36 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
     dclass->get_pixbuf = virt_viewer_display_spice_get_pixbuf;
     dclass->release_cursor = virt_viewer_display_spice_release_cursor;
     dclass->close = virt_viewer_display_spice_close;
+    dclass->selectable = virt_viewer_display_spice_selectable;
 
     g_type_class_add_private(klass, sizeof(VirtViewerDisplaySpicePrivate));
 }
 
+static SpiceMainChannel*
+get_main(VirtViewerDisplay *self)
+{
+    VirtViewerSessionSpice *session;
+
+    session = VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self));
+
+    return virt_viewer_session_spice_get_main_channel(session);
+}
+
 static void
 show_hint_changed(VirtViewerDisplay *self)
 {
-    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
-        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self)));
+    SpiceMainChannel *cmain = get_main(self);
     guint enabled = TRUE;
     guint nth;
 
+    if (!cmain)
+        return;
+
     g_object_get(self, "nth-display", &nth, NULL);
     if (virt_viewer_display_get_show_hint(self) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
         enabled = FALSE;
 
-    spice_main_set_display_enabled(main_channel, nth, enabled);
+    spice_main_set_display_enabled(cmain, nth, enabled);
 }
 
 static void
@@ -189,9 +203,8 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
 
     g_object_get(self, "nth-display", &nth, NULL);
 
-    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
-        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
-    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
+    spice_main_set_display(get_main(VIRT_VIEWER_DISPLAY(self)),
+                           nth, 0, 0, dw, dh);
 }
 
 static void
@@ -270,12 +283,25 @@ virt_viewer_display_spice_release_cursor(VirtViewerDisplay *display)
     spice_display_mouse_ungrab(self->priv->display);
 }
 
-
 static void
 virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED)
 {
 }
 
+static gboolean
+virt_viewer_display_spice_selectable(VirtViewerDisplay *self)
+{
+    gboolean agent_connected;
+    SpiceMainChannel *mainc;
+
+    mainc = get_main(self);
+    g_object_get(mainc,
+                 "agent-connected", &agent_connected,
+                 NULL);
+
+    return agent_connected;
+}
+
 
 /*
  * Local variables:
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 004f027..4612970 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -80,6 +80,7 @@ enum {
     PROP_ZOOM_LEVEL,
     PROP_SHOW_HINT,
     PROP_SESSION,
+    PROP_SELECTABLE,
 };
 
 static void
@@ -167,6 +168,13 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
                                                         G_PARAM_READWRITE |
                                                         G_PARAM_CONSTRUCT_ONLY));
 
+    g_object_class_install_property(object_class,
+                                    PROP_SELECTABLE,
+                                    g_param_spec_boolean("selectable",
+                                                         "Selectable",
+                                                         "Selectable",
+                                                         FALSE,
+                                                         G_PARAM_READABLE));
 
     g_signal_new("display-pointer-grab",
                  G_OBJECT_CLASS_TYPE(object_class),
@@ -302,6 +310,9 @@ virt_viewer_display_get_property(GObject *object,
     case PROP_SESSION:
         g_value_set_object(value, virt_viewer_display_get_session(display));
         break;
+    case PROP_SELECTABLE:
+        g_value_set_boolean(value, virt_viewer_display_get_selectable(display));
+        break;
 
     default:
         G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
@@ -601,6 +612,18 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *self)
     klass->release_cursor(self);
 }
 
+gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *self)
+{
+    VirtViewerDisplayClass *klass;
+
+    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), FALSE);
+
+    klass = VIRT_VIEWER_DISPLAY_GET_CLASS(self);
+    if (klass->selectable)
+        return klass->selectable(self);
+
+    return FALSE;
+}
 
 void virt_viewer_display_close(VirtViewerDisplay *self)
 {
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index 89b4817..4247570 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -77,6 +77,7 @@ struct _VirtViewerDisplayClass {
     void (*release_cursor)(VirtViewerDisplay *display);
 
     void (*close)(VirtViewerDisplay *display);
+    gboolean (*selectable)(VirtViewerDisplay *display);
 
     /* signals */
     void (*display_pointer_grab)(VirtViewerDisplay *display);
@@ -118,6 +119,7 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
 
 void virt_viewer_display_close(VirtViewerDisplay *display);
 void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
+gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *display);
 
 G_END_DECLS
 
diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
index 0df11ef..6e9c03d 100644
--- a/src/virt-viewer-session-spice.c
+++ b/src/virt-viewer-session-spice.c
@@ -405,12 +405,21 @@ virt_viewer_session_spice_usb_device_selection(VirtViewerSession *session,
 }
 
 static void
-agent_connected_changed(SpiceChannel *cmain,
+agent_connected_changed(SpiceChannel *cmain G_GNUC_UNUSED,
                         GParamSpec *pspec G_GNUC_UNUSED,
                         VirtViewerSessionSpice *self)
 {
+    // this will force refresh of application menu
+    g_signal_emit_by_name(self, "session-display-updated");
+}
+
+static void
+agent_connected_fullscreen_auto_conf(SpiceChannel *cmain,
+                                     GParamSpec *pspec G_GNUC_UNUSED,
+                                     VirtViewerSessionSpice *self)
+{
     if (virt_viewer_session_spice_fullscreen_auto_conf(self))
-        g_signal_handlers_disconnect_by_func(cmain, agent_connected_changed, self);
+        g_signal_handlers_disconnect_by_func(cmain, agent_connected_fullscreen_auto_conf, self);
 }
 
 static void
@@ -513,8 +522,9 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
                          G_CALLBACK(virt_viewer_session_spice_main_channel_event), self);
         self->priv->main_channel = SPICE_MAIN_CHANNEL(channel);
 
-        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
-        agent_connected_changed(channel, NULL, self);
+        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed), self);
+        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_fullscreen_auto_conf), self);
+        agent_connected_fullscreen_auto_conf(channel, NULL, self);
 
         g_signal_emit_by_name(session, "session-connected");
     }
diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
index 9249a1f..a1d96c2 100644
--- a/src/virt-viewer-session.c
+++ b/src/virt-viewer-session.c
@@ -233,6 +233,16 @@ virt_viewer_session_class_init(VirtViewerSessionClass *class)
                  1,
                  VIRT_VIEWER_TYPE_DISPLAY);
 
+    g_signal_new("session-display-updated",
+                 G_OBJECT_CLASS_TYPE(object_class),
+                 G_SIGNAL_RUN_LAST | G_SIGNAL_NO_HOOKS,
+                 G_STRUCT_OFFSET(VirtViewerSessionClass, session_display_updated),
+                 NULL,
+                 NULL,
+                 g_cclosure_marshal_VOID__VOID,
+                 G_TYPE_NONE,
+                 0);
+
     g_signal_new("session-cut-text",
                  G_OBJECT_CLASS_TYPE(object_class),
                  G_SIGNAL_RUN_LAST | G_SIGNAL_NO_HOOKS,
diff --git a/src/virt-viewer-session.h b/src/virt-viewer-session.h
index 44e4674..38ed988 100644
--- a/src/virt-viewer-session.h
+++ b/src/virt-viewer-session.h
@@ -88,6 +88,7 @@ struct _VirtViewerSessionClass {
                                   VirtViewerDisplay *display);
     void (*session_display_removed)(VirtViewerSession *session,
                                     VirtViewerDisplay *display);
+    void (*session_display_updated)(VirtViewerSession *session);
 
     void (*session_cut_text)(VirtViewerSession *session, const gchar *str);
     void (*session_bell)(VirtViewerSession *session);
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 46e44be..4bc427a 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -112,7 +112,8 @@ static void
 virt_viewer_window_get_property (GObject *object, guint property_id,
                                  GValue *value, GParamSpec *pspec)
 {
-    VirtViewerWindowPrivate *priv = VIRT_VIEWER_WINDOW(object)->priv;
+    VirtViewerWindow *self = VIRT_VIEWER_WINDOW(object);
+    VirtViewerWindowPrivate *priv = self->priv;
 
     switch (property_id) {
     case PROP_SUBTITLE:
@@ -124,7 +125,7 @@ virt_viewer_window_get_property (GObject *object, guint property_id,
         break;
 
     case PROP_DISPLAY:
-        g_value_set_object(value, priv->display);
+        g_value_set_object(value, virt_viewer_window_get_display(self));
         break;
 
     case PROP_CONTAINER:
@@ -1117,6 +1118,14 @@ virt_viewer_window_get_builder(VirtViewerWindow *self)
     return self->priv->builder;
 }
 
+VirtViewerDisplay*
+virt_viewer_window_get_display(VirtViewerWindow *self)
+{
+    g_return_val_if_fail(VIRT_VIEWER_WINDOW(self), FALSE);
+
+    return self->priv->display;
+}
+
 /*
  * Local variables:
  *  c-indent-level: 4
diff --git a/src/virt-viewer-window.h b/src/virt-viewer-window.h
index f80c4c7..44db585 100644
--- a/src/virt-viewer-window.h
+++ b/src/virt-viewer-window.h
@@ -63,6 +63,7 @@ GType virt_viewer_window_get_type (void);
 GtkWindow* virt_viewer_window_get_window (VirtViewerWindow* window);
 VirtViewerNotebook* virt_viewer_window_get_notebook (VirtViewerWindow* window);
 void virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *display);
+VirtViewerDisplay* virt_viewer_window_get_display(VirtViewerWindow *self);
 void virt_viewer_window_set_usb_options_sensitive(VirtViewerWindow *self, gboolean sensitive);
 void virt_viewer_window_update_title(VirtViewerWindow *self);
 void virt_viewer_window_show(VirtViewerWindow *self);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:37 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:37 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/19] spice: improve
	fullscreen=auto-conf
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-3-git-send-email-marcandre.lureau@gmail.com>

Do keep client monitor position, do not align monitors automatically.
The align property is only available since v0.12.101.
---
 configure.ac                    |    2 +-
 src/virt-viewer-session-spice.c |    4 ++++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/configure.ac b/configure.ac
index 494cd13..a2471f4 100644
--- a/configure.ac
+++ b/configure.ac
@@ -19,7 +19,7 @@ GTK2_REQUIRED="2.18.0"
 GTK3_REQUIRED="3.0"
 GTK_VNC1_REQUIRED="0.3.8"
 GTK_VNC2_REQUIRED="0.4.0"
-SPICE_GTK_REQUIRED="0.11"
+SPICE_GTK_REQUIRED="0.12.101"
 SPICE_PROTOCOL_REQUIRED="0.10.1"
 
 AC_MSG_CHECKING([for native Win32])
diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
index 7b47735..6577237 100644
--- a/src/virt-viewer-session-spice.c
+++ b/src/virt-viewer-session-spice.c
@@ -499,6 +499,10 @@ virt_viewer_session_spice_fullscreen_auto_conf(VirtViewerSessionSpice *self)
 
     DEBUG_LOG("Performing full screen auto-conf, %d host monitors",
               gdk_screen_get_n_monitors(screen));
+    g_object_set(G_OBJECT(cmain),
+                 "disable-display-position", FALSE,
+                 "disable-display-align", TRUE,
+                 NULL);
     spice_main_set_display_enabled(cmain, -1, FALSE);
     for (i = 0; i < gdk_screen_get_n_monitors(screen); i++) {
         gdk_screen_get_monitor_geometry(screen, i, &dest);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:51 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:51 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 16/19] Wait until idle handler
	has run to mark dirty = FALSE
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-17-git-send-email-marcandre.lureau@gmail.com>

This can avoid having multiple idler added
---
 src/virt-viewer-display.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 18fed19..004f027 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -326,6 +326,7 @@ virt_viewer_display_idle(gpointer opaque)
     VirtViewerDisplayPrivate *priv = display->priv;
     if (!priv->dirty)
         gtk_widget_queue_resize_no_redraw(GTK_WIDGET(display));
+    priv->dirty = FALSE;
     return FALSE;
 }
 
@@ -437,7 +438,6 @@ end:
      */
     if (priv->dirty) {
         g_idle_add(virt_viewer_display_idle, widget);
-        priv->dirty = FALSE;
     }
 }
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:35 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:35 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 00/19] Add new Spice monitors
	handling support
Message-ID: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>

Hi,

Spice now supports multiple arbitrary-sized monitors per display
channel. The client has to learn how to handle the new events in order
to create, to show, and to manage the displays appropriately.

The feature is currently being merged in the various projects, to test
it you'll need the following projects update with monitor config
support: spice-server, qemu, spice-gtk, qxl driver and vdagent.

cheers

Marc-Andr? Lureau (19):
  build: allow building with newer glibc-headers and -O0
  spice: improve fullscreen=auto-conf
  spice: disconnect signal handlers when either object is destroyed
  build: generate enums type boilerplate
  Destroy dialog immediately after run
  Turn display:show-hint into flags type
  Run-time check values before doing bad computation
  Mark a menu string as translatable
  Rebuild display menu when a window is added or removed
  Use virt_viewer_connect_object() for display
  Add a DISABLED display hint
  Prevent from adding the same display several time in the session
  Number display starting from 1
  Use SpiceDisplay:ready property instead of channel mark
  Hook up handling of Monitors
  Wait until idle handler has run to mark dirty = FALSE
  Make display submenu insensitive if they can't be modified
  Simplify display flag handling
  Show status page by default if display not ready

 acinclude.m4                      |    9 ++-
 configure.ac                      |    5 +-
 src/Makefile.am                   |   23 ++++++-
 src/virt-viewer-app.c             |   53 ++++++++++++----
 src/virt-viewer-display-spice.c   |  120 +++++++++++++++++++++++--------------
 src/virt-viewer-display-spice.h   |    2 +-
 src/virt-viewer-display.c         |   72 ++++++++++++++++++----
 src/virt-viewer-display.h         |   15 +++--
 src/virt-viewer-enums.c.etemplate |   60 +++++++++++++++++++
 src/virt-viewer-enums.h.etemplate |   41 +++++++++++++
 src/virt-viewer-notebook.c        |    1 +
 src/virt-viewer-session-spice.c   |  120 ++++++++++++++++++++++++++++++++-----
 src/virt-viewer-session-vnc.c     |    8 +--
 src/virt-viewer-session.c         |   13 ++++
 src/virt-viewer-session.h         |    1 +
 src/virt-viewer-util.c            |  117 ++++++++++++++++++++++++++++++++++++
 src/virt-viewer-util.h            |    6 ++
 src/virt-viewer-window.c          |   67 +++++++++++++++------
 src/virt-viewer-window.h          |    1 +
 19 files changed, 620 insertions(+), 114 deletions(-)
 create mode 100644 src/virt-viewer-enums.c.etemplate
 create mode 100644 src/virt-viewer-enums.h.etemplate

-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:42 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:42 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 07/19] Run-time check values
	before doing bad computation
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-8-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-display.c |    6 +++++-
 src/virt-viewer-window.c  |    5 +++++
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 43ec45e..a954671 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -403,6 +403,10 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
     DEBUG_LOG("Allocated %dx%d", allocation->width, allocation->height);
     gtk_widget_set_allocation(widget, allocation);
 
+    if (priv->desktopWidth == 0 ||
+        priv->desktopHeight == 0)
+        goto end;
+
     desktopAspect = (double)priv->desktopWidth / (double)priv->desktopHeight;
 
     if (child && gtk_widget_get_visible(child)) {
@@ -427,7 +431,7 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
         gtk_widget_size_allocate(child, &child_allocation);
     }
 
-
+end:
     /* This unsets the size request, so that the user can
      * manually resize the window smaller again
      */
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index c873ebc..4887e28 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -433,6 +433,11 @@ virt_viewer_window_resize(VirtViewerWindow *self)
                                     (screen, gtk_widget_get_window(priv->window)),
                                     &fullscreen);
 
+    g_return_if_fail(fullscreen.height > 128);
+    g_return_if_fail(fullscreen.width > 128);
+    g_return_if_fail(desktopWidth > 0);
+    g_return_if_fail(desktopHeight > 0);
+
     desktopAspect = (double)desktopWidth / (double)desktopHeight;
     screenAspect = (double)(fullscreen.width - 128) / (double)(fullscreen.height - 128);
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:43 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:43 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 08/19] Mark a menu string as
	translatable
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-9-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index b93a978..03cf4b4 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -1633,7 +1633,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
         gboolean visible;
         gchar *label;
 
-        label = g_strdup_printf("Display %d", *nth);
+        label = g_strdup_printf(_("Display %d"), *nth);
         item = gtk_check_menu_item_new_with_label(label);
         g_free(label);
         visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:48 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:48 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/19] Number display starting
	from 1
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index f581531..f590641 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -1638,7 +1638,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
         gboolean visible;
         gchar *label;
 
-        label = g_strdup_printf(_("Display %d"), *nth);
+        label = g_strdup_printf(_("Display %d"), *nth + 1);
         item = gtk_check_menu_item_new_with_label(label);
         g_free(label);
         visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:44 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:44 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 09/19] Rebuild display menu
	when a window is added or removed
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-10-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 03cf4b4..758cfe9 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -520,6 +520,7 @@ virt_viewer_app_remove_nth_window(VirtViewerApp *self, gint nth)
     DEBUG_LOG("Remove window %d %p", nth, win);
     removed = g_hash_table_steal(self->priv->windows, &nth);
     g_warn_if_fail(removed);
+    virt_viewer_app_update_menu_displays(self);
 
     if (removed)
         g_signal_emit(self, signals[SIGNAL_WINDOW_REMOVED], 0, win);
@@ -540,6 +541,7 @@ virt_viewer_app_set_nth_window(VirtViewerApp *self, gint nth, VirtViewerWindow *
     DEBUG_LOG("Insert window %d %p", nth, win);
     g_hash_table_insert(self->priv->windows, key, win);
     virt_viewer_app_set_window_subtitle(self, win, nth);
+    virt_viewer_app_update_menu_displays(self);
 
     g_signal_emit(self, signals[SIGNAL_WINDOW_ADDED], 0, win);
 }
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:50 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:50 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>

Rely on spice-gtk display channel monitors property to manage
displays. The same display channel may now provide several monitors,
the SpiceDisplay widget must be told which monitor to display
---
 src/virt-viewer-display-spice.c |   17 +++----
 src/virt-viewer-display-spice.h |    2 +-
 src/virt-viewer-session-spice.c |   98 +++++++++++++++++++++++++++++++++++----
 3 files changed, 98 insertions(+), 19 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 53430dd..ca74c1a 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -172,7 +172,7 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
 {
     gdouble dw = allocation->width, dh = allocation->height;
     guint zoom = 100;
-    guint channelid;
+    guint nth;
 
     if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
         return;
@@ -187,13 +187,11 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
         dh /= ((double)zoom / 100.0);
     }
 
-    g_object_get(self->priv->channel, "channel-id", &channelid, NULL);
+    g_object_get(self, "nth-display", &nth, NULL);
 
     SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
         VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
-    spice_main_set_display(main_channel,
-                           channelid,
-                           0, 0, dw, dh);
+    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
 }
 
 static void
@@ -212,7 +210,8 @@ enable_accel_changed(VirtViewerApp *app,
 
 GtkWidget *
 virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
-                              SpiceChannel *channel)
+                              SpiceChannel *channel,
+                              gint monitorid)
 {
     VirtViewerDisplaySpice *self;
     VirtViewerApp *app;
@@ -222,15 +221,17 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
     g_return_val_if_fail(SPICE_IS_DISPLAY_CHANNEL(channel), NULL);
 
     g_object_get(channel, "channel-id", &channelid, NULL);
+    // We don't allow monitorid != 0 && channelid != 0
+    g_return_val_if_fail(channelid == 0 || monitorid == 0, NULL);
 
     self = g_object_new(VIRT_VIEWER_TYPE_DISPLAY_SPICE,
                         "session", session,
-                        "nth-display", channelid,
+                        "nth-display", channelid + monitorid,
                         NULL);
     self->priv->channel = channel;
 
     g_object_get(session, "spice-session", &s, NULL);
-    self->priv->display = spice_display_new(s, channelid);
+    self->priv->display = spice_display_new_with_monitor(s, channelid, monitorid);
     g_object_unref(s);
 
     virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
diff --git a/src/virt-viewer-display-spice.h b/src/virt-viewer-display-spice.h
index 701ed85..c2013ec 100644
--- a/src/virt-viewer-display-spice.h
+++ b/src/virt-viewer-display-spice.h
@@ -66,7 +66,7 @@ struct _VirtViewerDisplaySpiceClass {
 
 GType virt_viewer_display_spice_get_type(void);
 
-GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel);
+GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel, gint monitorid);
 
 G_END_DECLS
 
diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
index 6577237..0df11ef 100644
--- a/src/virt-viewer-session-spice.c
+++ b/src/virt-viewer-session-spice.c
@@ -414,6 +414,80 @@ agent_connected_changed(SpiceChannel *cmain,
 }
 
 static void
+destroy_display(gpointer data)
+{
+    VirtViewerDisplay *display = VIRT_VIEWER_DISPLAY(data);
+    VirtViewerSession *session = virt_viewer_display_get_session(display);
+
+    DEBUG_LOG("Destroying spice display %p", display);
+    virt_viewer_session_remove_display(session, display);
+    g_object_unref(display);
+}
+
+static void
+virt_viewer_session_spice_display_monitors(SpiceChannel *channel,
+                                           GParamSpec *pspec G_GNUC_UNUSED,
+                                           VirtViewerSessionSpice *self)
+{
+    GArray *monitors = NULL;
+    GPtrArray *displays = NULL;
+    GtkWidget *display;
+    guint i, monitors_max;
+
+    g_object_get(channel,
+                 "monitors", &monitors,
+                 "monitors-max", &monitors_max,
+                 NULL);
+    g_return_if_fail(monitors != NULL);
+    g_return_if_fail(monitors->len <= monitors_max);
+
+    displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
+    if (displays == NULL) {
+        displays = g_ptr_array_new();
+#if GLIB_CHECK_VERSION(2, 22, 0)
+        g_ptr_array_set_free_func(displays, destroy_display);
+        g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
+                               displays, (GDestroyNotify)g_ptr_array_unref);
+#endif
+    }
+
+    g_ptr_array_set_size(displays, monitors_max);
+
+    for (i = 0; i < monitors_max; i++) {
+        display = g_ptr_array_index(displays, i);
+        if (display == NULL) {
+            display = virt_viewer_display_spice_new(self, channel, i);
+            DEBUG_LOG("creating spice display (#:%d)", i);
+            g_ptr_array_index(displays, i) = g_object_ref(display);
+        }
+
+        g_object_freeze_notify(G_OBJECT(display));
+        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), FALSE);
+        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(self),
+                                        VIRT_VIEWER_DISPLAY(display));
+    }
+
+    for (i = 0; i < monitors->len; i++) {
+        SpiceDisplayMonitorConfig *c = &g_array_index(monitors, SpiceDisplayMonitorConfig, i);
+        display = g_ptr_array_index(displays, c->id);
+        g_return_if_fail(display != NULL);
+
+        if (c->width == 0 || c->width == 0)
+            continue;
+
+        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), TRUE);
+        virt_viewer_display_set_desktop_size(VIRT_VIEWER_DISPLAY(display),
+                                             c->width, c->height);
+    }
+
+    for (i = 0; i < monitors_max; i++)
+        g_object_thaw_notify(g_ptr_array_index(displays, i));
+
+    g_clear_pointer(&monitors, g_array_unref);
+
+}
+
+static void
 virt_viewer_session_spice_channel_new(SpiceSession *s,
                                       SpiceChannel *channel,
                                       VirtViewerSession *session)
@@ -441,20 +515,17 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
 
         g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
         agent_connected_changed(channel, NULL, self);
+
+        g_signal_emit_by_name(session, "session-connected");
     }
 
     if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
-        GtkWidget *display;
-
-        g_signal_emit_by_name(session, "session-connected");
+        g_signal_emit_by_name(session, "session-initialized");
 
-        DEBUG_LOG("new display channel (#%d)", id);
-        display = virt_viewer_display_spice_new(self, channel);
-        g_object_set_data(G_OBJECT(channel), "virt-viewer-display", display);
-        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
-                                        VIRT_VIEWER_DISPLAY(display));
+        g_signal_connect(channel, "notify::monitors",
+                         G_CALLBACK(virt_viewer_session_spice_display_monitors), self);
 
-        g_signal_emit_by_name(session, "session-initialized");
+        spice_channel_connect(channel);
     }
 
     if (SPICE_IS_INPUTS_CHANNEL(channel)) {
@@ -538,7 +609,14 @@ virt_viewer_session_spice_channel_destroy(G_GNUC_UNUSED SpiceSession *s,
     if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
         VirtViewerDisplay *display = g_object_get_data(G_OBJECT(channel), "virt-viewer-display");
         DEBUG_LOG("zap display channel (#%d, %p)", id, display);
-        virt_viewer_session_remove_display(session, display);
+#if !GLIB_CHECK_VERSION(2, 22, 0)
+        GPtrArray *displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
+        if (displays) {
+            g_ptr_array_foreach(displays, (GFunc)destroy_display, NULL);
+            g_ptr_array_free(displays, TRUE);
+            g_object_set_data(channel, "virt-viewer-displays", NULL);
+        }
+#endif
     }
 
     if (SPICE_IS_PLAYBACK_CHANNEL(channel) && self->priv->audio) {
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:38 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:38 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 03/19] spice: disconnect
	signal handlers when either object is destroyed
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-4-git-send-email-marcandre.lureau@gmail.com>

Use virt_viewer_signal_connect_object(), a copy of telepathy
utility function tp_g_signal_connect_object(). This function
will take care of removing signal handler if any of emitter or
attached object are destroyed.
---
 src/virt-viewer-display-spice.c |   26 ++++-----
 src/virt-viewer-util.c          |  117 +++++++++++++++++++++++++++++++++++++++
 src/virt-viewer-util.h          |    6 ++
 3 files changed, 135 insertions(+), 14 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 0b6949b..7bf9a9a 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -219,10 +219,10 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
     self->priv->display = spice_display_new(s, channelid);
     g_object_unref(s);
 
-    g_signal_connect(channel, "display-primary-create",
-                     G_CALLBACK(primary_create), self);
-    g_signal_connect(channel, "display-mark",
-                     G_CALLBACK(display_mark), self);
+    virt_viewer_signal_connect_object(channel, "display-primary-create",
+                                      G_CALLBACK(primary_create), self, 0);
+    virt_viewer_signal_connect_object(channel, "display-mark",
+                                      G_CALLBACK(display_mark), self, 0);
 
     gtk_container_add(GTK_CONTAINER(self), g_object_ref(self->priv->display));
     gtk_widget_show(GTK_WIDGET(self->priv->display));
@@ -233,19 +233,17 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
                  "scaling", TRUE,
                  NULL);
 
-    g_signal_connect(self->priv->display,
-                     "keyboard-grab",
-                     G_CALLBACK(virt_viewer_display_spice_keyboard_grab), self);
-    g_signal_connect(self->priv->display,
-                     "mouse-grab",
-                     G_CALLBACK(virt_viewer_display_spice_mouse_grab), self);
-    g_signal_connect(self,
-                     "size-allocate",
-                     G_CALLBACK(virt_viewer_display_spice_size_allocate), self);
+    virt_viewer_signal_connect_object(self->priv->display, "keyboard-grab",
+                                      G_CALLBACK(virt_viewer_display_spice_keyboard_grab), self, 0);
+    virt_viewer_signal_connect_object(self->priv->display, "mouse-grab",
+                                      G_CALLBACK(virt_viewer_display_spice_mouse_grab), self, 0);
+    virt_viewer_signal_connect_object(self, "size-allocate",
+                                      G_CALLBACK(virt_viewer_display_spice_size_allocate), self, 0);
 
 
     app = virt_viewer_session_get_app(VIRT_VIEWER_SESSION(session));
-    g_signal_connect(app, "notify::enable-accel", G_CALLBACK(enable_accel_changed), self);
+    virt_viewer_signal_connect_object(app, "notify::enable-accel",
+                                      G_CALLBACK(enable_accel_changed), self, 0);
     enable_accel_changed(app, NULL, self);
 
     return GTK_WIDGET(self);
diff --git a/src/virt-viewer-util.c b/src/virt-viewer-util.c
index c1182af..9a163fa 100644
--- a/src/virt-viewer-util.c
+++ b/src/virt-viewer-util.c
@@ -136,6 +136,123 @@ virt_viewer_util_extract_host(const char *uristr,
     return 0;
 }
 
+typedef struct {
+    GObject *instance;
+    GObject *observer;
+    GClosure *closure;
+    gulong handler_id;
+} WeakHandlerCtx;
+
+static WeakHandlerCtx *
+whc_new(GObject *instance,
+        GObject *observer)
+{
+    WeakHandlerCtx *ctx = g_slice_new0(WeakHandlerCtx);
+
+    ctx->instance = instance;
+    ctx->observer = observer;
+
+    return ctx;
+}
+
+static void
+whc_free(WeakHandlerCtx *ctx)
+{
+    g_slice_free(WeakHandlerCtx, ctx);
+}
+
+static void observer_destroyed_cb(gpointer, GObject *);
+static void closure_invalidated_cb(gpointer, GClosure *);
+
+/*
+ * If signal handlers are removed before the object is destroyed, this
+ * callback will never get triggered.
+ */
+static void
+instance_destroyed_cb(gpointer ctx_,
+                      GObject *where_the_instance_was G_GNUC_UNUSED)
+{
+    WeakHandlerCtx *ctx = ctx_;
+
+    /* No need to disconnect the signal here, the instance has gone away. */
+    g_object_weak_unref(ctx->observer, observer_destroyed_cb, ctx);
+    g_closure_remove_invalidate_notifier(ctx->closure, ctx,
+                                         closure_invalidated_cb);
+    whc_free(ctx);
+}
+
+/* Triggered when the observer is destroyed. */
+static void
+observer_destroyed_cb(gpointer ctx_,
+                      GObject *where_the_observer_was G_GNUC_UNUSED)
+{
+    WeakHandlerCtx *ctx = ctx_;
+
+    g_closure_remove_invalidate_notifier(ctx->closure, ctx,
+                                         closure_invalidated_cb);
+    g_signal_handler_disconnect(ctx->instance, ctx->handler_id);
+    g_object_weak_unref(ctx->instance, instance_destroyed_cb, ctx);
+    whc_free(ctx);
+}
+
+/* Triggered when either object is destroyed or the handler is disconnected. */
+static void
+closure_invalidated_cb(gpointer ctx_,
+                       GClosure *where_the_closure_was G_GNUC_UNUSED)
+{
+    WeakHandlerCtx *ctx = ctx_;
+
+    g_object_weak_unref(ctx->instance, instance_destroyed_cb, ctx);
+    g_object_weak_unref(ctx->observer, observer_destroyed_cb, ctx);
+    whc_free(ctx);
+}
+
+/* Copied from tp_g_signal_connect_object. */
+/**
+  * virt_viewer_signal_connect_object: (skip)
+  * @instance: the instance to connect to.
+  * @detailed_signal: a string of the form "signal-name::detail".
+  * @c_handler: the #GCallback to connect.
+  * @gobject: the object to pass as data to @c_handler.
+  * @connect_flags: a combination of #GConnectFlags.
+  *
+  * Similar to g_signal_connect_object() but will delete connection
+  * when any of the objects is destroyed.
+  *
+  * Returns: the handler id.
+  */
+gulong virt_viewer_signal_connect_object(gpointer instance,
+                                         const gchar *detailed_signal,
+                                         GCallback c_handler,
+                                         gpointer gobject,
+                                         GConnectFlags connect_flags)
+{
+    GObject *instance_obj = G_OBJECT(instance);
+    WeakHandlerCtx *ctx = whc_new(instance_obj, gobject);
+
+    g_return_val_if_fail(G_TYPE_CHECK_INSTANCE (instance), 0);
+    g_return_val_if_fail(detailed_signal != NULL, 0);
+    g_return_val_if_fail(c_handler != NULL, 0);
+    g_return_val_if_fail(G_IS_OBJECT (gobject), 0);
+    g_return_val_if_fail((connect_flags & ~(G_CONNECT_AFTER|G_CONNECT_SWAPPED)) == 0, 0);
+
+    if (connect_flags & G_CONNECT_SWAPPED)
+        ctx->closure = g_cclosure_new_object_swap(c_handler, gobject);
+    else
+        ctx->closure = g_cclosure_new_object(c_handler, gobject);
+
+    ctx->handler_id = g_signal_connect_closure(instance, detailed_signal,
+                                               ctx->closure, (connect_flags & G_CONNECT_AFTER) ? TRUE : FALSE);
+
+    g_object_weak_ref(instance_obj, instance_destroyed_cb, ctx);
+    g_object_weak_ref(gobject, observer_destroyed_cb, ctx);
+    g_closure_add_invalidate_notifier(ctx->closure, ctx,
+                                      closure_invalidated_cb);
+
+    return ctx->handler_id;
+}
+
+
 /*
  * Local variables:
  *  c-indent-level: 4
diff --git a/src/virt-viewer-util.h b/src/virt-viewer-util.h
index 3a40651..38c8078 100644
--- a/src/virt-viewer-util.h
+++ b/src/virt-viewer-util.h
@@ -40,6 +40,12 @@ int virt_viewer_util_extract_host(const char *uristr,
                                   char **user,
                                   int *port);
 
+gulong virt_viewer_signal_connect_object(gpointer instance,
+                                         const gchar *detailed_signal,
+                                         GCallback c_handler,
+                                         gpointer gobject,
+                                         GConnectFlags connect_flags);
+
 #endif
 
 /*
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:36 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:36 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 01/19] build: allow building
	with newer glibc-headers and -O0
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-2-git-send-email-marcandre.lureau@gmail.com>

    Fix copied from libvirt, commit by Eric Blake.

    glibc 2.15 (on Fedora 17) coupled with explicit disabling of
    optimization during development dies a painful death:

    /usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires compiling with optimization (-O) [-Werror=cpp]

    Work around this by only conditionally defining _FORTIFY_SOURCE,
    in the case where glibc can actually use it.  The trick is using
    AH_VERBATIM instead of AC_DEFINE.
---
 acinclude.m4 |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/acinclude.m4 b/acinclude.m4
index 73b5a7b..0c308bc 100644
--- a/acinclude.m4
+++ b/acinclude.m4
@@ -19,7 +19,7 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
 
     warnCFLAGS=
 
-    try_compiler_flags="-Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
+    try_compiler_flags="-fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
 
     case "$enable_compile_warnings" in
     no)
@@ -44,6 +44,13 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
 	;;
     esac
 
+    AH_VERBATIM([FORTIFY_SOURCE],
+                [/* Enable compile-time and run-time bounds-checking, and some warnings. */
+                 #if defined __OPTIMIZE__ && __OPTIMIZE__
+                 # define _FORTIFY_SOURCE 2
+                 #endif
+                ])
+
     compiler_flags=
     for option in $try_compiler_flags; do
 	SAVE_CFLAGS="$CFLAGS"
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:46 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:46 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 11/19] Add a DISABLED display
	hint
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-12-git-send-email-marcandre.lureau@gmail.com>

This flag will help to track wether the display has been
removed/closed and whether it have a really a valid display.

Ready in contrast, is used to "hide" temporarily the display (when
starting or redrawing the display, to avoid artifacts)
---
 src/virt-viewer-app.c           |   15 +++++++++------
 src/virt-viewer-display-spice.c |   20 ++++++++++++++++++++
 src/virt-viewer-display.c       |   15 +++++++++++++++
 src/virt-viewer-display.h       |    2 ++
 src/virt-viewer-window.c        |   12 ++++++++++++
 5 files changed, 58 insertions(+), 6 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index e5c674a..f581531 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -611,7 +611,8 @@ display_show_hint(VirtViewerDisplay *display,
 {
     VirtViewerApp *self;
     VirtViewerNotebook *nb = virt_viewer_window_get_notebook(win);
-    gint nth, hint;
+    gint nth;
+    guint hint;
 
     g_object_get(win,
                  "app", &self,
@@ -621,15 +622,17 @@ display_show_hint(VirtViewerDisplay *display,
                  "show-hint", &hint,
                  NULL);
 
-    if (hint == 0) {
+    if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED) {
+        virt_viewer_window_hide(win);
+    } else if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY) {
+        virt_viewer_notebook_show_display(nb);
+        virt_viewer_window_show(win);
+        gtk_window_present(virt_viewer_window_get_window(win));
+    } else {
         if (win != self->priv->main_window &&
             g_getenv("VIRT_VIEWER_HIDE"))
             virt_viewer_window_hide(win);
         virt_viewer_notebook_show_status(nb, _("Waiting for display %d..."), nth + 1);
-    } else {
-        virt_viewer_notebook_show_display(nb);
-        virt_viewer_window_show(win);
-        gtk_window_present(virt_viewer_window_get_window(win));
     }
 
     g_object_unref(self);
diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 7bf9a9a..101abc5 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -75,9 +75,26 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
 }
 
 static void
+show_hint_changed(VirtViewerDisplay *self)
+{
+    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
+        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self)));
+    guint enabled = TRUE;
+    guint nth;
+
+    g_object_get(self, "nth-display", &nth, NULL);
+    if (virt_viewer_display_get_show_hint(self) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
+        enabled = FALSE;
+
+    spice_main_set_display_enabled(main_channel, nth, enabled);
+}
+
+static void
 virt_viewer_display_spice_init(VirtViewerDisplaySpice *self G_GNUC_UNUSED)
 {
     self->priv = VIRT_VIEWER_DISPLAY_SPICE_GET_PRIVATE(self);
+
+    g_signal_connect(self, "notify::show-hint", G_CALLBACK(show_hint_changed), NULL);
 }
 
 static void
@@ -166,6 +183,9 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
     if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
         return;
 
+    if (virt_viewer_display_get_show_hint(VIRT_VIEWER_DISPLAY(self)) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
+        return;
+
     if (virt_viewer_display_get_zoom(VIRT_VIEWER_DISPLAY(self))) {
         zoom = virt_viewer_display_get_zoom_level(VIRT_VIEWER_DISPLAY(self));
 
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index a954671..18fed19 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -553,6 +553,21 @@ void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
     g_object_notify(G_OBJECT(self), "show-hint");
 }
 
+void virt_viewer_display_set_enabled(VirtViewerDisplay *self, gboolean enabled)
+{
+    guint hint;
+
+    g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
+
+    hint = virt_viewer_display_get_show_hint(self);
+    if (enabled)
+        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
+    else
+        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
+
+    virt_viewer_display_set_show_hint(self, hint);
+}
+
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *self)
 {
     g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), NULL);
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index 225080d..89b4817 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -57,6 +57,7 @@ typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
 
 typedef enum {
     VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
+    VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED         = 1 << 1,
 } VirtViewerDisplayShowHintFlags;
 
 /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
@@ -116,6 +117,7 @@ gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
 void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
 
 void virt_viewer_display_close(VirtViewerDisplay *display);
+void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
 
 G_END_DECLS
 
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 3f7cbbe..46e44be 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1062,6 +1062,9 @@ virt_viewer_window_show(VirtViewerWindow *self)
 {
     gtk_widget_show(self->priv->window);
 
+    if (self->priv->display)
+        virt_viewer_display_set_enabled(self->priv->display, TRUE);
+
     if (self->priv->desktop_resize_pending) {
         virt_viewer_window_resize(self);
         self->priv->desktop_resize_pending = FALSE;
@@ -1072,6 +1075,15 @@ void
 virt_viewer_window_hide(VirtViewerWindow *self)
 {
     gtk_widget_hide(self->priv->window);
+
+    if (self->priv->display) {
+        VirtViewerDisplay *display = self->priv->display;
+        guint nth;
+
+        g_object_get(display, "nth-display", &nth, NULL);
+        if (nth != 0)
+            virt_viewer_display_set_enabled(display, FALSE);
+    }
 }
 
 void
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Mon Jul 16 16:57:53 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 16 Jul 2012 18:57:53 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 18/19] Simplify display flag
	handling
In-Reply-To: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342457874-15369-19-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-display-spice.c |    9 +--------
 src/virt-viewer-display.c       |   20 ++++++++++----------
 src/virt-viewer-display.h       |    2 +-
 src/virt-viewer-session-vnc.c   |    8 ++++----
 4 files changed, 16 insertions(+), 23 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index a91d73e..bbc1773 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -140,19 +140,12 @@ display_ready(GObject *display,
               GParamSpec *pspec G_GNUC_UNUSED,
               VirtViewerDisplay *self)
 {
-    guint hint;
     gboolean ready;
 
     g_object_get(display, "ready", &ready, NULL);
     DEBUG_LOG("display %p ready:%d", self, ready);
 
-    hint = virt_viewer_display_get_show_hint(self);
-    if (ready)
-        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
-    else
-        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
-
-    virt_viewer_display_set_show_hint(self, hint);
+    virt_viewer_display_set_show_hint(self, VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, ready);
 }
 
 static void
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 4612970..b8e6d67 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -551,12 +551,20 @@ guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
     return self->priv->show_hint;
 }
 
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint mask, gboolean enabled)
 {
     VirtViewerDisplayPrivate *priv;
+    guint hint;
     g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
 
     priv = self->priv;
+    hint = priv->show_hint;
+
+    if (enabled)
+        hint |= mask;
+    else
+        hint &= ~mask;
+
     if (priv->show_hint == hint)
         return;
 
@@ -566,17 +574,9 @@ void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
 
 void virt_viewer_display_set_enabled(VirtViewerDisplay *self, gboolean enabled)
 {
-    guint hint;
-
     g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
 
-    hint = virt_viewer_display_get_show_hint(self);
-    if (enabled)
-        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
-    else
-        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
-
-    virt_viewer_display_set_show_hint(self, hint);
+    virt_viewer_display_set_show_hint(self, VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED, !enabled);
 }
 
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *self)
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index 4247570..d838c07 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -110,7 +110,7 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
 void virt_viewer_display_send_keys(VirtViewerDisplay *display,
                                    const guint *keyvals, int nkeyvals);
 GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint mask, gboolean enabled);
 guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
 void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
diff --git a/src/virt-viewer-session-vnc.c b/src/virt-viewer-session-vnc.c
index 238c865..52baf5c 100644
--- a/src/virt-viewer-session-vnc.c
+++ b/src/virt-viewer-session-vnc.c
@@ -95,7 +95,7 @@ virt_viewer_session_vnc_connected(VncDisplay *vnc G_GNUC_UNUSED,
     GtkWidget *display = virt_viewer_display_vnc_new(session->priv->vnc);
     g_signal_emit_by_name(session, "session-connected");
     virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(display),
-                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
+                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, TRUE);
     virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
                                     VIRT_VIEWER_DISPLAY(display));
 }
@@ -104,11 +104,11 @@ static void
 virt_viewer_session_vnc_disconnected(VncDisplay *vnc G_GNUC_UNUSED,
                                      VirtViewerSessionVnc *session)
 {
+    GtkWidget *display = virt_viewer_display_vnc_new(session->priv->vnc);
     DEBUG_LOG("Disconnected");
     g_signal_emit_by_name(session, "session-disconnected");
-    /* TODO perhaps? */
-    /* virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(session->priv->vnc), */
-    /*                                   VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE); */
+    virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(display),
+                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, FALSE);
 }
 
 static void
-- 
1.7.10.4



From cfergeau at redhat.com  Tue Jul 17 08:28:40 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:28:40 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/19] spice: improve
 fullscreen=auto-conf
In-Reply-To: <1342457874-15369-3-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-3-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717082840.GL2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:37PM +0200, Marc-Andr? Lureau wrote:
> Do keep client monitor position, do not align monitors automatically.
> The align property is only available since v0.12.101.

There are no spice-gtk release tarball yet with these changes, right?
Looks good otherwise

Christophe

> ---
>  configure.ac                    |    2 +-
>  src/virt-viewer-session-spice.c |    4 ++++
>  2 files changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/configure.ac b/configure.ac
> index 494cd13..a2471f4 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -19,7 +19,7 @@ GTK2_REQUIRED="2.18.0"
>  GTK3_REQUIRED="3.0"
>  GTK_VNC1_REQUIRED="0.3.8"
>  GTK_VNC2_REQUIRED="0.4.0"
> -SPICE_GTK_REQUIRED="0.11"
> +SPICE_GTK_REQUIRED="0.12.101"
>  SPICE_PROTOCOL_REQUIRED="0.10.1"
>  
>  AC_MSG_CHECKING([for native Win32])
> diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
> index 7b47735..6577237 100644
> --- a/src/virt-viewer-session-spice.c
> +++ b/src/virt-viewer-session-spice.c
> @@ -499,6 +499,10 @@ virt_viewer_session_spice_fullscreen_auto_conf(VirtViewerSessionSpice *self)
>  
>      DEBUG_LOG("Performing full screen auto-conf, %d host monitors",
>                gdk_screen_get_n_monitors(screen));
> +    g_object_set(G_OBJECT(cmain),
> +                 "disable-display-position", FALSE,
> +                 "disable-display-align", TRUE,
> +                 NULL);
>      spice_main_set_display_enabled(cmain, -1, FALSE);
>      for (i = 0; i < gdk_screen_get_n_monitors(screen); i++) {
>          gdk_screen_get_monitor_geometry(screen, i, &dest);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/1d2d7ce8/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 08:32:37 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:32:37 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 01/19] build: allow
 building with newer glibc-headers and -O0
In-Reply-To: <1342457874-15369-2-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-2-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717083236.GM2162@teriyaki.cdg.redhat.com>

Looks good, ACK

Christophe

On Mon, Jul 16, 2012 at 06:57:36PM +0200, Marc-Andr? Lureau wrote:
>     Fix copied from libvirt, commit by Eric Blake.
> 
>     glibc 2.15 (on Fedora 17) coupled with explicit disabling of
>     optimization during development dies a painful death:
> 
>     /usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires compiling with optimization (-O) [-Werror=cpp]
> 
>     Work around this by only conditionally defining _FORTIFY_SOURCE,
>     in the case where glibc can actually use it.  The trick is using
>     AH_VERBATIM instead of AC_DEFINE.
> ---
>  acinclude.m4 |    9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/acinclude.m4 b/acinclude.m4
> index 73b5a7b..0c308bc 100644
> --- a/acinclude.m4
> +++ b/acinclude.m4
> @@ -19,7 +19,7 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
>  
>      warnCFLAGS=
>  
> -    try_compiler_flags="-Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
> +    try_compiler_flags="-fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
>  
>      case "$enable_compile_warnings" in
>      no)
> @@ -44,6 +44,13 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
>  	;;
>      esac
>  
> +    AH_VERBATIM([FORTIFY_SOURCE],
> +                [/* Enable compile-time and run-time bounds-checking, and some warnings. */
> +                 #if defined __OPTIMIZE__ && __OPTIMIZE__
> +                 # define _FORTIFY_SOURCE 2
> +                 #endif
> +                ])
> +
>      compiler_flags=
>      for option in $try_compiler_flags; do
>  	SAVE_CFLAGS="$CFLAGS"
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/f5dd4c40/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 08:37:23 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:37:23 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 03/19] spice: disconnect
 signal handlers when either object is destroyed
In-Reply-To: <1342457874-15369-4-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-4-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717083723.GN2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:38PM +0200, Marc-Andr? Lureau wrote:
> Use virt_viewer_signal_connect_object(), a copy of telepathy
> utility function tp_g_signal_connect_object(). This function
> will take care of removing signal handler if any of emitter or
> attached object are destroyed.

Why do you need to do this? Are you seeing concrete issues? Or will it be
needed by subsequent changes?

Christophe


> ---
>  src/virt-viewer-display-spice.c |   26 ++++-----
>  src/virt-viewer-util.c          |  117 +++++++++++++++++++++++++++++++++++++++
>  src/virt-viewer-util.h          |    6 ++
>  3 files changed, 135 insertions(+), 14 deletions(-)
> 
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index 0b6949b..7bf9a9a 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -219,10 +219,10 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>      self->priv->display = spice_display_new(s, channelid);
>      g_object_unref(s);
>  
> -    g_signal_connect(channel, "display-primary-create",
> -                     G_CALLBACK(primary_create), self);
> -    g_signal_connect(channel, "display-mark",
> -                     G_CALLBACK(display_mark), self);
> +    virt_viewer_signal_connect_object(channel, "display-primary-create",
> +                                      G_CALLBACK(primary_create), self, 0);
> +    virt_viewer_signal_connect_object(channel, "display-mark",
> +                                      G_CALLBACK(display_mark), self, 0);
>  
>      gtk_container_add(GTK_CONTAINER(self), g_object_ref(self->priv->display));
>      gtk_widget_show(GTK_WIDGET(self->priv->display));
> @@ -233,19 +233,17 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>                   "scaling", TRUE,
>                   NULL);
>  
> -    g_signal_connect(self->priv->display,
> -                     "keyboard-grab",
> -                     G_CALLBACK(virt_viewer_display_spice_keyboard_grab), self);
> -    g_signal_connect(self->priv->display,
> -                     "mouse-grab",
> -                     G_CALLBACK(virt_viewer_display_spice_mouse_grab), self);
> -    g_signal_connect(self,
> -                     "size-allocate",
> -                     G_CALLBACK(virt_viewer_display_spice_size_allocate), self);
> +    virt_viewer_signal_connect_object(self->priv->display, "keyboard-grab",
> +                                      G_CALLBACK(virt_viewer_display_spice_keyboard_grab), self, 0);
> +    virt_viewer_signal_connect_object(self->priv->display, "mouse-grab",
> +                                      G_CALLBACK(virt_viewer_display_spice_mouse_grab), self, 0);
> +    virt_viewer_signal_connect_object(self, "size-allocate",
> +                                      G_CALLBACK(virt_viewer_display_spice_size_allocate), self, 0);
>  
>  
>      app = virt_viewer_session_get_app(VIRT_VIEWER_SESSION(session));
> -    g_signal_connect(app, "notify::enable-accel", G_CALLBACK(enable_accel_changed), self);
> +    virt_viewer_signal_connect_object(app, "notify::enable-accel",
> +                                      G_CALLBACK(enable_accel_changed), self, 0);
>      enable_accel_changed(app, NULL, self);
>  
>      return GTK_WIDGET(self);
> diff --git a/src/virt-viewer-util.c b/src/virt-viewer-util.c
> index c1182af..9a163fa 100644
> --- a/src/virt-viewer-util.c
> +++ b/src/virt-viewer-util.c
> @@ -136,6 +136,123 @@ virt_viewer_util_extract_host(const char *uristr,
>      return 0;
>  }
>  
> +typedef struct {
> +    GObject *instance;
> +    GObject *observer;
> +    GClosure *closure;
> +    gulong handler_id;
> +} WeakHandlerCtx;
> +
> +static WeakHandlerCtx *
> +whc_new(GObject *instance,
> +        GObject *observer)
> +{
> +    WeakHandlerCtx *ctx = g_slice_new0(WeakHandlerCtx);
> +
> +    ctx->instance = instance;
> +    ctx->observer = observer;
> +
> +    return ctx;
> +}
> +
> +static void
> +whc_free(WeakHandlerCtx *ctx)
> +{
> +    g_slice_free(WeakHandlerCtx, ctx);
> +}
> +
> +static void observer_destroyed_cb(gpointer, GObject *);
> +static void closure_invalidated_cb(gpointer, GClosure *);
> +
> +/*
> + * If signal handlers are removed before the object is destroyed, this
> + * callback will never get triggered.
> + */
> +static void
> +instance_destroyed_cb(gpointer ctx_,
> +                      GObject *where_the_instance_was G_GNUC_UNUSED)
> +{
> +    WeakHandlerCtx *ctx = ctx_;
> +
> +    /* No need to disconnect the signal here, the instance has gone away. */
> +    g_object_weak_unref(ctx->observer, observer_destroyed_cb, ctx);
> +    g_closure_remove_invalidate_notifier(ctx->closure, ctx,
> +                                         closure_invalidated_cb);
> +    whc_free(ctx);
> +}
> +
> +/* Triggered when the observer is destroyed. */
> +static void
> +observer_destroyed_cb(gpointer ctx_,
> +                      GObject *where_the_observer_was G_GNUC_UNUSED)
> +{
> +    WeakHandlerCtx *ctx = ctx_;
> +
> +    g_closure_remove_invalidate_notifier(ctx->closure, ctx,
> +                                         closure_invalidated_cb);
> +    g_signal_handler_disconnect(ctx->instance, ctx->handler_id);
> +    g_object_weak_unref(ctx->instance, instance_destroyed_cb, ctx);
> +    whc_free(ctx);
> +}
> +
> +/* Triggered when either object is destroyed or the handler is disconnected. */
> +static void
> +closure_invalidated_cb(gpointer ctx_,
> +                       GClosure *where_the_closure_was G_GNUC_UNUSED)
> +{
> +    WeakHandlerCtx *ctx = ctx_;
> +
> +    g_object_weak_unref(ctx->instance, instance_destroyed_cb, ctx);
> +    g_object_weak_unref(ctx->observer, observer_destroyed_cb, ctx);
> +    whc_free(ctx);
> +}
> +
> +/* Copied from tp_g_signal_connect_object. */
> +/**
> +  * virt_viewer_signal_connect_object: (skip)
> +  * @instance: the instance to connect to.
> +  * @detailed_signal: a string of the form "signal-name::detail".
> +  * @c_handler: the #GCallback to connect.
> +  * @gobject: the object to pass as data to @c_handler.
> +  * @connect_flags: a combination of #GConnectFlags.
> +  *
> +  * Similar to g_signal_connect_object() but will delete connection
> +  * when any of the objects is destroyed.
> +  *
> +  * Returns: the handler id.
> +  */
> +gulong virt_viewer_signal_connect_object(gpointer instance,
> +                                         const gchar *detailed_signal,
> +                                         GCallback c_handler,
> +                                         gpointer gobject,
> +                                         GConnectFlags connect_flags)
> +{
> +    GObject *instance_obj = G_OBJECT(instance);
> +    WeakHandlerCtx *ctx = whc_new(instance_obj, gobject);
> +
> +    g_return_val_if_fail(G_TYPE_CHECK_INSTANCE (instance), 0);
> +    g_return_val_if_fail(detailed_signal != NULL, 0);
> +    g_return_val_if_fail(c_handler != NULL, 0);
> +    g_return_val_if_fail(G_IS_OBJECT (gobject), 0);
> +    g_return_val_if_fail((connect_flags & ~(G_CONNECT_AFTER|G_CONNECT_SWAPPED)) == 0, 0);
> +
> +    if (connect_flags & G_CONNECT_SWAPPED)
> +        ctx->closure = g_cclosure_new_object_swap(c_handler, gobject);
> +    else
> +        ctx->closure = g_cclosure_new_object(c_handler, gobject);
> +
> +    ctx->handler_id = g_signal_connect_closure(instance, detailed_signal,
> +                                               ctx->closure, (connect_flags & G_CONNECT_AFTER) ? TRUE : FALSE);
> +
> +    g_object_weak_ref(instance_obj, instance_destroyed_cb, ctx);
> +    g_object_weak_ref(gobject, observer_destroyed_cb, ctx);
> +    g_closure_add_invalidate_notifier(ctx->closure, ctx,
> +                                      closure_invalidated_cb);
> +
> +    return ctx->handler_id;
> +}
> +
> +
>  /*
>   * Local variables:
>   *  c-indent-level: 4
> diff --git a/src/virt-viewer-util.h b/src/virt-viewer-util.h
> index 3a40651..38c8078 100644
> --- a/src/virt-viewer-util.h
> +++ b/src/virt-viewer-util.h
> @@ -40,6 +40,12 @@ int virt_viewer_util_extract_host(const char *uristr,
>                                    char **user,
>                                    int *port);
>  
> +gulong virt_viewer_signal_connect_object(gpointer instance,
> +                                         const gchar *detailed_signal,
> +                                         GCallback c_handler,
> +                                         gpointer gobject,
> +                                         GConnectFlags connect_flags);
> +
>  #endif
>  
>  /*
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/65d4cb47/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 08:41:09 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:41:09 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 04/19] build: generate
 enums type boilerplate
In-Reply-To: <1342457874-15369-5-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-5-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717084109.GO2162@teriyaki.cdg.redhat.com>

ACK

On Mon, Jul 16, 2012 at 06:57:39PM +0200, Marc-Andr? Lureau wrote:
> ---
>  configure.ac                      |    3 ++
>  src/Makefile.am                   |   23 ++++++++++++--
>  src/virt-viewer-enums.c.etemplate |   60 +++++++++++++++++++++++++++++++++++++
>  src/virt-viewer-enums.h.etemplate |   41 +++++++++++++++++++++++++
>  4 files changed, 124 insertions(+), 3 deletions(-)
>  create mode 100644 src/virt-viewer-enums.c.etemplate
>  create mode 100644 src/virt-viewer-enums.h.etemplate
> 
> diff --git a/configure.ac b/configure.ac
> index a2471f4..548b945 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -72,6 +72,9 @@ AC_DEFINE_UNQUOTED([GETTEXT_PACKAGE],"$GETTEXT_PACKAGE", [GETTEXT package name])
>  AM_GLIB_GNU_GETTEXT
>  IT_PROG_INTLTOOL([0.35.0])
>  
> +PKG_PROG_PKG_CONFIG
> +GLIB_MKENUMS=`$PKG_CONFIG --variable=glib_mkenums glib-2.0`
> +AC_SUBST(GLIB_MKENUMS)
>  
>  PKG_CHECK_MODULES(GLIB2, glib-2.0 >= $GLIB2_REQUIRED gthread-2.0 gmodule-export-2.0)
>  PKG_CHECK_MODULES(LIBXML2, libxml-2.0 >= $LIBXML2_REQUIRED)
> diff --git a/src/Makefile.am b/src/Makefile.am
> index d99b043..5daa98a 100644
> --- a/src/Makefile.am
> +++ b/src/Makefile.am
> @@ -10,11 +10,28 @@ builderxml_DATA =				\
>  	virt-viewer-auth.xml			\
>  	$(NULL)
>  
> -EXTRA_DIST = $(builderxml_DATA) \
> -	gbinding.c \
> -	gbinding.h
> +EXTRA_DIST =					\
> +	$(builderxml_DATA)			\
> +	virt-viewer-enums.c.etemplate		\
> +	virt-viewer-enums.h.etemplate		\
> +	gbinding.c				\
> +	gbinding.h				\
> +	$(NULL)
> +
> +ENUMS_FILES =					\
> +	virt-viewer-display.h			\
> +	$(NULL)
> +
> +BUILT_SOURCES =					\
> +	virt-viewer-enums.h			\
> +	virt-viewer-enums.c			\
> +	$(NULL)
> +
> +$(BUILT_SOURCES): %: %.etemplate $(ENUMS_FILES)
> +	$(AM_V_GEN)$(GLIB_MKENUMS) --template $^ | sed -e 's/VIRT_TYPE_VIEWER/VIRT_VIEWER_TYPE/' > $@
>  
>  COMMON_SOURCES =					\
> +	$(BUILT_SOURCES)				\
>  	virt-gtk-compat.h				\
>  	virt-viewer-util.h virt-viewer-util.c		\
>  	virt-viewer-auth.h virt-viewer-auth.c		\
> diff --git a/src/virt-viewer-enums.c.etemplate b/src/virt-viewer-enums.c.etemplate
> new file mode 100644
> index 0000000..6f9925f
> --- /dev/null
> +++ b/src/virt-viewer-enums.c.etemplate
> @@ -0,0 +1,60 @@
> +/*** BEGIN file-header ***/
> +/*
> + * Virt Viewer: A virtual machine console viewer
> + *
> + * Copyright (C) 2007-2012 Red Hat, Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
> + *
> + * Author: Marc-Andr? Lureau <marcandre.lureau at redhat.com>
> + */
> +
> +#include "virt-viewer-enums.h"
> +
> +/*** END file-header ***/
> +
> +/*** BEGIN file-production ***/
> +#include "@filename@"
> +/*** END file-production ***/
> +
> +
> +/*** BEGIN value-header ***/
> +
> +GType
> + at enum_name@_get_type (void)
> +{
> +  static volatile gsize g_define_type_id__volatile = 0;
> +
> +  if (g_once_init_enter (&g_define_type_id__volatile))
> +    {
> +      static const G at Type@Value values[] = {
> +/*** END value-header ***/
> +
> +/*** BEGIN value-production ***/
> +        { @VALUENAME@, "@VALUENAME@", "@valuenick@" },
> +/*** END value-production ***/
> +
> +/*** BEGIN value-tail ***/
> +        { 0, NULL, NULL }
> +      };
> +      GType g_define_type_id =
> +        g_ at type@_register_static (g_intern_static_string ("@EnumName@"), values);
> +      g_once_init_leave (&g_define_type_id__volatile, g_define_type_id);
> +    }
> +
> +  return g_define_type_id__volatile;
> +}
> +
> +/*** END value-tail ***/
> diff --git a/src/virt-viewer-enums.h.etemplate b/src/virt-viewer-enums.h.etemplate
> new file mode 100644
> index 0000000..f33158d
> --- /dev/null
> +++ b/src/virt-viewer-enums.h.etemplate
> @@ -0,0 +1,41 @@
> +/*** BEGIN file-header ***/
> +/*
> + * Virt Viewer: A virtual machine console viewer
> + *
> + * Copyright (C) 2007-2012 Red Hat, Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
> + *
> + * Author: Marc-Andr? Lureau <marcandre.lureau at redhat.com>
> + */
> +
> +#ifndef VIRT_VIEWER_ENUMS_H
> +#define VIRT_VIEWER_ENUMS_H
> +
> +#include <glib-object.h>
> +
> +G_BEGIN_DECLS
> +/*** END file-header ***/
> +
> +/*** BEGIN value-header ***/
> +GType @enum_name at _get_type (void) G_GNUC_CONST;
> +#define @ENUMPREFIX at _TYPE_@ENUMSHORT@ (@enum_name at _get_type ())
> +/*** END value-header ***/
> +
> +/*** BEGIN file-tail ***/
> +G_END_DECLS
> +
> +#endif
> +/*** END file-tail ***/
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/5634df78/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 08:42:37 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:42:37 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 05/19] Destroy dialog
 immediately after run
In-Reply-To: <1342457874-15369-6-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-6-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717084237.GP2162@teriyaki.cdg.redhat.com>

Makes sense, ACK

On Mon, Jul 16, 2012 at 06:57:40PM +0200, Marc-Andr? Lureau wrote:
> If the parent is already destroyed, it looks like the dialog is
> destroyed too. This avoids a crash when calling app_quit().
> ---
>  src/virt-viewer-app.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 40d1ff7..1447ce7 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -283,6 +283,7 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
>                                          GTK_BUTTONS_OK_CANCEL,
>                                          _("This is the last visible display. Do you want to quit?"));
>              gint result = gtk_dialog_run (GTK_DIALOG (dialog));
> +            gtk_widget_destroy(dialog);
>              switch (result) {
>              case GTK_RESPONSE_OK:
>                  virt_viewer_app_quit(self);
> @@ -290,7 +291,6 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
>              default:
>                  break;
>              }
> -            gtk_widget_destroy(dialog);
>              return FALSE;
>          } else {
>              virt_viewer_app_quit(self);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/9e8154f9/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 08:57:40 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:57:40 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn
 display:show-hint into flags type
In-Reply-To: <1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717085740.GQ2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:41PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c     |    2 +-
>  src/virt-viewer-display.c |   26 ++++++++++++++++----------
>  src/virt-viewer-display.h |   11 ++++++-----
>  src/virt-viewer-window.c  |    5 ++++-
>  4 files changed, 27 insertions(+), 17 deletions(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 1447ce7..b93a978 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -619,7 +619,7 @@ display_show_hint(VirtViewerDisplay *display,
>                   "show-hint", &hint,
>                   NULL);
>  
> -    if (hint == VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE) {
> +    if (hint == 0) {
>          if (win != self->priv->main_window &&
>              g_getenv("VIRT_VIEWER_HIDE"))
>              virt_viewer_window_hide(win);
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 6d6052b..43ec45e 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -41,7 +41,7 @@ struct _VirtViewerDisplayPrivate
>      guint zoom_level;
>      gboolean zoom;
>      gint nth_display;
> -    gint show_hint;
> +    guint show_hint;
>      VirtViewerSession *session;
>      gboolean auto_resize;
>  };
> @@ -151,13 +151,12 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
>  
>      g_object_class_install_property(object_class,
>                                      PROP_SHOW_HINT,
> -                                    g_param_spec_int("show-hint",
> -                                                     "Show hint",
> -                                                     "Show state hint",
> -                                                     0,
> -                                                     G_MAXINT32,
> -                                                     0,
> -                                                     G_PARAM_READABLE));
> +                                    g_param_spec_flags("show-hint",
> +                                                       "Show hint",
> +                                                       "Show state hint",
> +                                                       VIRT_VIEWER_TYPE_DISPLAY_SHOW_HINT_FLAGS,
> +                                                       0,
> +                                                       G_PARAM_READABLE));
>  
>      g_object_class_install_property(object_class,
>                                      PROP_SESSION,
> @@ -298,7 +297,7 @@ virt_viewer_display_get_property(GObject *object,
>          g_value_set_int(value, priv->nth_display);
>          break;
>      case PROP_SHOW_HINT:
> -        g_value_set_int(value, priv->show_hint);
> +        g_value_set_flags(value, priv->show_hint);
>          break;
>      case PROP_SESSION:
>          g_value_set_object(value, virt_viewer_display_get_session(display));
> @@ -530,7 +529,14 @@ GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display)
>      return VIRT_VIEWER_DISPLAY_GET_CLASS(display)->get_pixbuf(display);
>  }
>  
> -void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, gint hint)
> +guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
> +{
> +    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), 0);
> +
> +    return self->priv->show_hint;
> +}
> +
> +void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
>  {
>      VirtViewerDisplayPrivate *priv;
>      g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
> diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
> index ffbaf0e..225080d 100644
> --- a/src/virt-viewer-display.h
> +++ b/src/virt-viewer-display.h
> @@ -25,6 +25,7 @@
>  #define _VIRT_VIEWER_DISPLAY_H
>  
>  #include <gtk/gtk.h>
> +#include "virt-viewer-enums.h"
>  
>  G_BEGIN_DECLS
>  
> @@ -54,10 +55,9 @@ typedef struct _VirtViewerDisplayPrivate VirtViewerDisplayPrivate;
>  
>  typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
>  
> -enum {
> -    VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE = 0,
> -    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY,
> -};
> +typedef enum {
> +    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
> +} VirtViewerDisplayShowHintFlags;

I would have kept HINT_HIDE as the flag name as it's more
expressive than _READY imo. Looks good either way.

Christophe

>  
>  /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
>  struct _VirtViewerDisplay {
> @@ -108,7 +108,8 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
>  void virt_viewer_display_send_keys(VirtViewerDisplay *display,
>                                     const guint *keyvals, int nkeyvals);
>  GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
> -void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, gint hint);
> +void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
> +guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
>  VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
>  void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
>  gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 7b582a0..c873ebc 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -1002,9 +1002,12 @@ display_show_hint(VirtViewerDisplay *display,
>                    GParamSpec *pspec G_GNUC_UNUSED,
>                    VirtViewerWindow *self)
>  {
> -    gboolean hint;
> +    guint hint;
>  
>      g_object_get(display, "show-hint", &hint, NULL);
> +
> +    hint = (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
> +
>      gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-send")), hint);
>      gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-file-screenshot")), hint);
>      gtk_widget_set_sensitive(self->priv->toolbar_send_key, hint);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/7e1cabdf/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 08:59:12 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 10:59:12 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn
 display:show-hint into flags type
In-Reply-To: <1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717085912.GR2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:41PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c     |    2 +-
>  src/virt-viewer-display.c |   26 ++++++++++++++++----------
>  src/virt-viewer-display.h |   11 ++++++-----
>  src/virt-viewer-window.c  |    5 ++++-
>  4 files changed, 27 insertions(+), 17 deletions(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 1447ce7..b93a978 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -619,7 +619,7 @@ display_show_hint(VirtViewerDisplay *display,
>                   "show-hint", &hint,
>                   NULL);
>  
> -    if (hint == VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE) {
> +    if (hint == 0) {

On second thought, shouldn't this be if (!(hint & _HINT_READY)) { } ? I'm
not sure which flags you are going to add.

Christophe

>          if (win != self->priv->main_window &&
>              g_getenv("VIRT_VIEWER_HIDE"))
>              virt_viewer_window_hide(win);
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 6d6052b..43ec45e 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -41,7 +41,7 @@ struct _VirtViewerDisplayPrivate
>      guint zoom_level;
>      gboolean zoom;
>      gint nth_display;
> -    gint show_hint;
> +    guint show_hint;
>      VirtViewerSession *session;
>      gboolean auto_resize;
>  };
> @@ -151,13 +151,12 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
>  
>      g_object_class_install_property(object_class,
>                                      PROP_SHOW_HINT,
> -                                    g_param_spec_int("show-hint",
> -                                                     "Show hint",
> -                                                     "Show state hint",
> -                                                     0,
> -                                                     G_MAXINT32,
> -                                                     0,
> -                                                     G_PARAM_READABLE));
> +                                    g_param_spec_flags("show-hint",
> +                                                       "Show hint",
> +                                                       "Show state hint",
> +                                                       VIRT_VIEWER_TYPE_DISPLAY_SHOW_HINT_FLAGS,
> +                                                       0,
> +                                                       G_PARAM_READABLE));
>  
>      g_object_class_install_property(object_class,
>                                      PROP_SESSION,
> @@ -298,7 +297,7 @@ virt_viewer_display_get_property(GObject *object,
>          g_value_set_int(value, priv->nth_display);
>          break;
>      case PROP_SHOW_HINT:
> -        g_value_set_int(value, priv->show_hint);
> +        g_value_set_flags(value, priv->show_hint);
>          break;
>      case PROP_SESSION:
>          g_value_set_object(value, virt_viewer_display_get_session(display));
> @@ -530,7 +529,14 @@ GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display)
>      return VIRT_VIEWER_DISPLAY_GET_CLASS(display)->get_pixbuf(display);
>  }
>  
> -void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, gint hint)
> +guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
> +{
> +    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), 0);
> +
> +    return self->priv->show_hint;
> +}
> +
> +void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
>  {
>      VirtViewerDisplayPrivate *priv;
>      g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
> diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
> index ffbaf0e..225080d 100644
> --- a/src/virt-viewer-display.h
> +++ b/src/virt-viewer-display.h
> @@ -25,6 +25,7 @@
>  #define _VIRT_VIEWER_DISPLAY_H
>  
>  #include <gtk/gtk.h>
> +#include "virt-viewer-enums.h"
>  
>  G_BEGIN_DECLS
>  
> @@ -54,10 +55,9 @@ typedef struct _VirtViewerDisplayPrivate VirtViewerDisplayPrivate;
>  
>  typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
>  
> -enum {
> -    VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE = 0,
> -    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY,
> -};
> +typedef enum {
> +    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
> +} VirtViewerDisplayShowHintFlags;
>  
>  /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
>  struct _VirtViewerDisplay {
> @@ -108,7 +108,8 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
>  void virt_viewer_display_send_keys(VirtViewerDisplay *display,
>                                     const guint *keyvals, int nkeyvals);
>  GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
> -void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, gint hint);
> +void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
> +guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
>  VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
>  void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
>  gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 7b582a0..c873ebc 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -1002,9 +1002,12 @@ display_show_hint(VirtViewerDisplay *display,
>                    GParamSpec *pspec G_GNUC_UNUSED,
>                    VirtViewerWindow *self)
>  {
> -    gboolean hint;
> +    guint hint;
>  
>      g_object_get(display, "show-hint", &hint, NULL);
> +
> +    hint = (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
> +
>      gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-send")), hint);
>      gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-file-screenshot")), hint);
>      gtk_widget_set_sensitive(self->priv->toolbar_send_key, hint);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/6ebdf88d/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:10:09 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:10:09 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 07/19] Run-time check
 values before doing bad computation
In-Reply-To: <1342457874-15369-8-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-8-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717091008.GS2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:42PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-display.c |    6 +++++-
>  src/virt-viewer-window.c  |    5 +++++
>  2 files changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 43ec45e..a954671 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -403,6 +403,10 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
>      DEBUG_LOG("Allocated %dx%d", allocation->width, allocation->height);
>      gtk_widget_set_allocation(widget, allocation);
>  
> +    if (priv->desktopWidth == 0 ||
> +        priv->desktopHeight == 0)
> +        goto end;

Why can this happen?

> +
>      desktopAspect = (double)priv->desktopWidth / (double)priv->desktopHeight;
>  
>      if (child && gtk_widget_get_visible(child)) {
> @@ -427,7 +431,7 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
>          gtk_widget_size_allocate(child, &child_allocation);
>      }
>  
> -
> +end:
>      /* This unsets the size request, so that the user can
>       * manually resize the window smaller again
>       */
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index c873ebc..4887e28 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -433,6 +433,11 @@ virt_viewer_window_resize(VirtViewerWindow *self)
>                                      (screen, gtk_widget_get_window(priv->window)),
>                                      &fullscreen);
>  
> +    g_return_if_fail(fullscreen.height > 128);
> +    g_return_if_fail(fullscreen.width > 128);
> +    g_return_if_fail(desktopWidth > 0);
> +    g_return_if_fail(desktopHeight > 0);

In size_allocate 0 width/height are silently ignored, here you return with
a warning, is it on purpose?

Christophe

> +
>      desktopAspect = (double)desktopWidth / (double)desktopHeight;
>      screenAspect = (double)(fullscreen.width - 128) / (double)(fullscreen.height - 128);
>  
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/65a8e02e/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:11:03 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:11:03 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 08/19] Mark a menu string
 as translatable
In-Reply-To: <1342457874-15369-9-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-9-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717091103.GT2162@teriyaki.cdg.redhat.com>

ACK. Maybe add a translator comment explaining what 'display' means in this
context?

Christophe

On Mon, Jul 16, 2012 at 06:57:43PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index b93a978..03cf4b4 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -1633,7 +1633,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
>          gboolean visible;
>          gchar *label;
>  
> -        label = g_strdup_printf("Display %d", *nth);
> +        label = g_strdup_printf(_("Display %d"), *nth);
>          item = gtk_check_menu_item_new_with_label(label);
>          g_free(label);
>          visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/9b412e9d/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:24:54 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:24:54 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 09/19] Rebuild display
 menu when a window is added or removed
In-Reply-To: <1342457874-15369-10-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-10-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717092454.GU2162@teriyaki.cdg.redhat.com>

Makes sense, ACK

However, I'm worried by the way VirtViewerApp uses hash tables,
virt_viewer_app_update_menu_displays calls g_list_sort and then g_list_free on
the list returned by g_hash_table_get_keys while this list should not be
modified (I'm assuming g_list_sort does the sorting in place and does not
create a new list). Keys to the 'windows' hash table should also be using
GUINT_TO_POINTER. I can send patches for this after this series gets in.

Christophe

On Mon, Jul 16, 2012 at 06:57:44PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |    2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 03cf4b4..758cfe9 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -520,6 +520,7 @@ virt_viewer_app_remove_nth_window(VirtViewerApp *self, gint nth)
>      DEBUG_LOG("Remove window %d %p", nth, win);
>      removed = g_hash_table_steal(self->priv->windows, &nth);
>      g_warn_if_fail(removed);
> +    virt_viewer_app_update_menu_displays(self);
>  
>      if (removed)
>          g_signal_emit(self, signals[SIGNAL_WINDOW_REMOVED], 0, win);
> @@ -540,6 +541,7 @@ virt_viewer_app_set_nth_window(VirtViewerApp *self, gint nth, VirtViewerWindow *
>      DEBUG_LOG("Insert window %d %p", nth, win);
>      g_hash_table_insert(self->priv->windows, key, win);
>      virt_viewer_app_set_window_subtitle(self, win, nth);
> +    virt_viewer_app_update_menu_displays(self);
>  
>      g_signal_emit(self, signals[SIGNAL_WINDOW_ADDED], 0, win);
>  }
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/8876abf8/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:26:19 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:26:19 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 10/19] Use
 virt_viewer_connect_object() for display
In-Reply-To: <1342457874-15369-11-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-11-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717092619.GV2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:45PM +0200, Marc-Andr? Lureau wrote:
> When display is released, detach signal automatically

Same question as the similar patch before, why? I'm assuming that it's
because we are going to handle displays dynamically, and if we don't do
that, we'll sometimes end up with signals using dead objects?

> ---
>  src/virt-viewer-app.c    |    4 ++--
>  src/virt-viewer-window.c |   24 ++++++++++++------------
>  2 files changed, 14 insertions(+), 14 deletions(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 758cfe9..e5c674a 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -658,8 +658,8 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
>      }
>  
>      virt_viewer_window_set_display(window, display);
> -    g_signal_connect(display, "notify::show-hint",
> -                     G_CALLBACK(display_show_hint), window);
> +    virt_viewer_signal_connect_object(display, "notify::show-hint",
> +                                      G_CALLBACK(display_show_hint), window, 0);
>      g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */
>  }
>  
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 4887e28..3f7cbbe 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -1042,18 +1042,18 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
>          gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
>          gtk_widget_show_all(GTK_WIDGET(display));
>  
> -        g_signal_connect(display, "display-pointer-grab",
> -                         G_CALLBACK(virt_viewer_window_pointer_grab), self);
> -        g_signal_connect(display, "display-pointer-ungrab",
> -                         G_CALLBACK(virt_viewer_window_pointer_ungrab), self);
> -        g_signal_connect(display, "display-keyboard-grab",
> -                         G_CALLBACK(virt_viewer_window_keyboard_grab), self);
> -        g_signal_connect(display, "display-keyboard-ungrab",
> -                         G_CALLBACK(virt_viewer_window_keyboard_ungrab), self);
> -        g_signal_connect(display, "display-desktop-resize",
> -                         G_CALLBACK(virt_viewer_window_desktop_resize), self);
> -        g_signal_connect(display, "notify::show-hint",
> -                         G_CALLBACK(display_show_hint), self);
> +        virt_viewer_signal_connect_object(display, "display-pointer-grab",
> +                                          G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);
> +        virt_viewer_signal_connect_object(display, "display-pointer-ungrab",
> +                                          G_CALLBACK(virt_viewer_window_pointer_ungrab), self, 0);
> +        virt_viewer_signal_connect_object(display, "display-keyboard-grab",
> +                                          G_CALLBACK(virt_viewer_window_keyboard_grab), self, 0);
> +        virt_viewer_signal_connect_object(display, "display-keyboard-ungrab",
> +                                          G_CALLBACK(virt_viewer_window_keyboard_ungrab), self, 0);
> +        virt_viewer_signal_connect_object(display, "display-desktop-resize",
> +                                          G_CALLBACK(virt_viewer_window_desktop_resize), self, 0);
> +        virt_viewer_signal_connect_object(display, "notify::show-hint",
> +                                          G_CALLBACK(display_show_hint), self, 0);
>      }
>  }
>  
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/8719688e/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:42:31 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:42:31 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 11/19] Add a DISABLED
 display hint
In-Reply-To: <1342457874-15369-12-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-12-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717094231.GW2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:46PM +0200, Marc-Andr? Lureau wrote:
> This flag will help to track wether the display has been

whether

> removed/closed and whether it have a really a valid display.

the part starting at "whether" needs fixing. "Whether it really has a valid
display" ?

> 
> Ready in contrast, is used to "hide" temporarily the display (when
> starting or redrawing the display, to avoid artifacts)
> ---
>  src/virt-viewer-app.c           |   15 +++++++++------
>  src/virt-viewer-display-spice.c |   20 ++++++++++++++++++++
>  src/virt-viewer-display.c       |   15 +++++++++++++++
>  src/virt-viewer-display.h       |    2 ++
>  src/virt-viewer-window.c        |   12 ++++++++++++
>  5 files changed, 58 insertions(+), 6 deletions(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index e5c674a..f581531 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -611,7 +611,8 @@ display_show_hint(VirtViewerDisplay *display,
>  {
>      VirtViewerApp *self;
>      VirtViewerNotebook *nb = virt_viewer_window_get_notebook(win);
> -    gint nth, hint;
> +    gint nth;
> +    guint hint;

Doesn't this change belong to the patch turning 'hint' into a bitfield?

Patch looks good otherwise.

Christophe

>  
>      g_object_get(win,
>                   "app", &self,
> @@ -621,15 +622,17 @@ display_show_hint(VirtViewerDisplay *display,
>                   "show-hint", &hint,
>                   NULL);
>  
> -    if (hint == 0) {
> +    if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED) {
> +        virt_viewer_window_hide(win);
> +    } else if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY) {
> +        virt_viewer_notebook_show_display(nb);
> +        virt_viewer_window_show(win);
> +        gtk_window_present(virt_viewer_window_get_window(win));
> +    } else {
>          if (win != self->priv->main_window &&
>              g_getenv("VIRT_VIEWER_HIDE"))
>              virt_viewer_window_hide(win);
>          virt_viewer_notebook_show_status(nb, _("Waiting for display %d..."), nth + 1);
> -    } else {
> -        virt_viewer_notebook_show_display(nb);
> -        virt_viewer_window_show(win);
> -        gtk_window_present(virt_viewer_window_get_window(win));
>      }
>  
>      g_object_unref(self);
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index 7bf9a9a..101abc5 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -75,9 +75,26 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
>  }
>  
>  static void
> +show_hint_changed(VirtViewerDisplay *self)
> +{
> +    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
> +        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self)));
> +    guint enabled = TRUE;
> +    guint nth;
> +
> +    g_object_get(self, "nth-display", &nth, NULL);
> +    if (virt_viewer_display_get_show_hint(self) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
> +        enabled = FALSE;
> +
> +    spice_main_set_display_enabled(main_channel, nth, enabled);
> +}
> +
> +static void
>  virt_viewer_display_spice_init(VirtViewerDisplaySpice *self G_GNUC_UNUSED)
>  {
>      self->priv = VIRT_VIEWER_DISPLAY_SPICE_GET_PRIVATE(self);
> +
> +    g_signal_connect(self, "notify::show-hint", G_CALLBACK(show_hint_changed), NULL);
>  }
>  
>  static void
> @@ -166,6 +183,9 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>      if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
>          return;
>  
> +    if (virt_viewer_display_get_show_hint(VIRT_VIEWER_DISPLAY(self)) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
> +        return;
> +
>      if (virt_viewer_display_get_zoom(VIRT_VIEWER_DISPLAY(self))) {
>          zoom = virt_viewer_display_get_zoom_level(VIRT_VIEWER_DISPLAY(self));
>  
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index a954671..18fed19 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -553,6 +553,21 @@ void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
>      g_object_notify(G_OBJECT(self), "show-hint");
>  }
>  
> +void virt_viewer_display_set_enabled(VirtViewerDisplay *self, gboolean enabled)
> +{
> +    guint hint;
> +
> +    g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
> +
> +    hint = virt_viewer_display_get_show_hint(self);
> +    if (enabled)
> +        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
> +    else
> +        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
> +
> +    virt_viewer_display_set_show_hint(self, hint);
> +}
> +
>  VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *self)
>  {
>      g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), NULL);
> diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
> index 225080d..89b4817 100644
> --- a/src/virt-viewer-display.h
> +++ b/src/virt-viewer-display.h
> @@ -57,6 +57,7 @@ typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
>  
>  typedef enum {
>      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
> +    VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED         = 1 << 1,
>  } VirtViewerDisplayShowHintFlags;
>  
>  /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
> @@ -116,6 +117,7 @@ gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
>  void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
>  
>  void virt_viewer_display_close(VirtViewerDisplay *display);
> +void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
>  
>  G_END_DECLS
>  
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 3f7cbbe..46e44be 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -1062,6 +1062,9 @@ virt_viewer_window_show(VirtViewerWindow *self)
>  {
>      gtk_widget_show(self->priv->window);
>  
> +    if (self->priv->display)
> +        virt_viewer_display_set_enabled(self->priv->display, TRUE);
> +
>      if (self->priv->desktop_resize_pending) {
>          virt_viewer_window_resize(self);
>          self->priv->desktop_resize_pending = FALSE;
> @@ -1072,6 +1075,15 @@ void
>  virt_viewer_window_hide(VirtViewerWindow *self)
>  {
>      gtk_widget_hide(self->priv->window);
> +
> +    if (self->priv->display) {
> +        VirtViewerDisplay *display = self->priv->display;
> +        guint nth;
> +
> +        g_object_get(display, "nth-display", &nth, NULL);
> +        if (nth != 0)
> +            virt_viewer_display_set_enabled(display, FALSE);
> +    }
>  }
>  
>  void
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/f8238115/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:43:19 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:43:19 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 12/19] Prevent from adding
 the same display several time in the session
In-Reply-To: <1342457874-15369-13-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-13-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717094319.GX2162@teriyaki.cdg.redhat.com>

How can this happen?

Christophe

On Mon, Jul 16, 2012 at 06:57:47PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-session.c |    3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
> index 18b6922..9249a1f 100644
> --- a/src/virt-viewer-session.c
> +++ b/src/virt-viewer-session.c
> @@ -282,6 +282,9 @@ virt_viewer_session_new(void)
>  void virt_viewer_session_add_display(VirtViewerSession *session,
>                                       VirtViewerDisplay *display)
>  {
> +    if (g_list_find(session->priv->displays, display))
> +        return;
> +
>      session->priv->displays = g_list_append(session->priv->displays, display);
>      g_object_ref(display);
>      g_signal_emit_by_name(session, "session-display-added", display);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/75f071a3/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 09:46:04 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 11:46:04 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/19] Number display
 starting from 1
In-Reply-To: <1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717094604.GY2162@teriyaki.cdg.redhat.com>

Why? To match the window subtitle? Please add this to the commit log.
Id add a +1 as well to
DEBUG_LOG("Remove window %d %p", nth, win);
and
DEBUG_LOG("Insert window %d %p", nth, win);

Christophe

On Mon, Jul 16, 2012 at 06:57:48PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index f581531..f590641 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -1638,7 +1638,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
>          gboolean visible;
>          gchar *label;
>  
> -        label = g_strdup_printf(_("Display %d"), *nth);
> +        label = g_strdup_printf(_("Display %d"), *nth + 1);
>          item = gtk_check_menu_item_new_with_label(label);
>          g_free(label);
>          visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/001da276/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 10:50:36 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 12:50:36 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 03/19] spice: disconnect
 signal handlers when either object is destroyed
In-Reply-To: <20120717083723.GN2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-4-git-send-email-marcandre.lureau@gmail.com>
	<20120717083723.GN2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1C+OsagLgTB=xmczGT+1O3Zec2ePQuRPTXP4VmhEShPGDA@mail.gmail.com>

On Tue, Jul 17, 2012 at 10:37 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> Why do you need to do this? Are you seeing concrete issues? Or will it be
> needed by subsequent changes?

"This function
will take care of removing signal handler if any of emitter or
attached object are destroyed."

The following patches will have this condition met, since there is no
longer 1-1 relation between channel and display. The channels can
continue to be around when some of the display are removed.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 10:57:24 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 12:57:24 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn
 display:show-hint into flags type
In-Reply-To: <20120717085740.GQ2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
	<20120717085740.GQ2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CKoMkzvtL-n=UMzcQ0rFNf5vMia4NX9wjwoBm6vpSg0DQ@mail.gmail.com>

On Tue, Jul 17, 2012 at 10:57 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> I would have kept HINT_HIDE as the flag name as it's more
> expressive than _READY imo. Looks good either way.

"HIDE" isn't necessarily 0 anymore, there will be a DISABLED flag in a
following patch. Now that you say it, it could have been "ENABLED"
flag instead, so 0 would still map to !READY && !ENABLED, but I prefer
to DISABLED, since this is the exception, and we might had more flags
in the future, making 0 hard to keep meaningful perhaps.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 10:59:22 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 12:59:22 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn
 display:show-hint into flags type
In-Reply-To: <20120717085912.GR2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
	<20120717085912.GR2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJSFbUbAdVAq_NBhOJKbQTqko7Wkr=_i4cdvXX-Yr+42Q@mail.gmail.com>

On Tue, Jul 17, 2012 at 10:59 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> On second thought, shouldn't this be if (!(hint & _HINT_READY)) { } ? I'm
> not sure which flags you are going to add.

VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE used to be an enum of value 0, now
there is only a READY flag. The following patch will make this easier
to read.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 11:03:57 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 13:03:57 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 07/19] Run-time check
 values before doing bad computation
In-Reply-To: <20120717091008.GS2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-8-git-send-email-marcandre.lureau@gmail.com>
	<20120717091008.GS2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1C+60OCuircobDcSgixL+Q0xHhgo8WGT6UnmBiPzttHJfA@mail.gmail.com>

Hi

On Tue, Jul 17, 2012 at 11:10 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
>> +    if (priv->desktopWidth == 0 ||
>> +        priv->desktopHeight == 0)
>> +        goto end;
>
> Why can this happen?

It the spice display doesn't know, for example, we don't yet have
monitor geometry.

>> +    g_return_if_fail(fullscreen.height > 128);
>> +    g_return_if_fail(fullscreen.width > 128);
>> +    g_return_if_fail(desktopWidth > 0);
>> +    g_return_if_fail(desktopHeight > 0);
>
> In size_allocate 0 width/height are silently ignored, here you return with
> a warning, is it on purpose?

This is the following code:

    desktopAspect = (double)desktopWidth / (double)desktopHeight;
    screenAspect = (double)(fullscreen.width - 128) /
(double)(fullscreen.height - 128);

    if ((desktopWidth > (fullscreen.width - 128)) ||
        (desktopHeight > (fullscreen.height - 128))) {

I wanted a bit of run-time check, in practice we don't seem to hit
these condition, but you can see it would be nice to have a warning if
it happens.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 11:08:30 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 13:08:30 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 09/19] Rebuild display
 menu when a window is added or removed
In-Reply-To: <20120717092454.GU2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-10-git-send-email-marcandre.lureau@gmail.com>
	<20120717092454.GU2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CKc6q_QDSQkQ45puQ-eZf=trJohK-BckQodMeV3baqrgQ@mail.gmail.com>

On Tue, Jul 17, 2012 at 11:24 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> Makes sense, ACK
>
> However, I'm worried by the way VirtViewerApp uses hash tables,
> virt_viewer_app_update_menu_displays calls g_list_sort and then g_list_free on
> the list returned by g_hash_table_get_keys while this list should not be
> modified (I'm assuming g_list_sort does the sorting in place and does not
> create a new list). Keys to the 'windows' hash table should also be using
> GUINT_TO_POINTER. I can send patches for this after this series gets in.

Agree, I had the same thought. Either a sorted-list or a GTree would
do better I think.

I would be happy if you clean that up.

Another thing that is not so nice about the menu rebuild, is that it
may now happen quite often, at startup time. It doesn't seem to be a
problem atm, but I wonder if an "on-demand" approach wouldn't make
more sense. Anyway, just something to keep in mind.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 11:11:50 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 13:11:50 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 10/19] Use
 virt_viewer_connect_object() for display
In-Reply-To: <20120717092619.GV2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-11-git-send-email-marcandre.lureau@gmail.com>
	<20120717092619.GV2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1C+=Wr5ApDTK5oS9qV_R7z27hrp=+LHH4dHDvbLJa5vFbA@mail.gmail.com>

On Tue, Jul 17, 2012 at 11:26 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> On Mon, Jul 16, 2012 at 06:57:45PM +0200, Marc-Andr? Lureau wrote:
>> When display is released, detach signal automatically
>
> Same question as the similar patch before, why? I'm assuming that it's
> because we are going to handle displays dynamically, and if we don't do
> that, we'll sometimes end up with signals using dead objects?

Yeah, I don't have a backtrace, but you can get into various crash
related to not cleaning up signal handlers properly. In general, this
signal_connect helper is really awesome, as you don't have to worry
about either emitter or receiver being dead, but some time you may
prefer to actually crash if something goes wrong. So a systematic
replace isn't what I would recommend. But for Spice channel & display,
it is needed to cleanup if either of the two is destroyed.


-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 11:14:27 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 13:14:27 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 12/19] Prevent from adding
 the same display several time in the session
In-Reply-To: <20120717094319.GX2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-13-git-send-email-marcandre.lureau@gmail.com>
	<20120717094319.GX2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1C++rRPDB0a4FEjk4UeDmprG=AL-FzeqzsY8Vju39nd=vw@mail.gmail.com>

On Tue, Jul 17, 2012 at 11:43 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> How can this happen?

In virt_viewer_session_spice_display_monitors(), we (re)add the
display unconditionnaly every time we receive a new MonitorConfig.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 11:16:19 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 13:16:19 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/19] Number display
 starting from 1
In-Reply-To: <20120717094604.GY2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>
	<20120717094604.GY2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJmMVoBp_bmFBBunkumqxq=hE=NyDD8v8GJvCmH=7oSnQ@mail.gmail.com>

On Tue, Jul 17, 2012 at 11:46 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> Why? To match the window subtitle? Please add this to the commit log.

It just is more user friendly, no? I think human start counting from 1
in general.

> Id add a +1 as well to
> DEBUG_LOG("Remove window %d %p", nth, win);
> and
> DEBUG_LOG("Insert window %d %p", nth, win);

However, this is not meant to be user friendly, but rather developper
friendly, so we should just print the value of these variable, imho


-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 11:26:34 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 13:26:34 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 11/19] Add a DISABLED
 display hint
In-Reply-To: <20120717094231.GW2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-12-git-send-email-marcandre.lureau@gmail.com>
	<20120717094231.GW2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CLLfSFLSo=O-CLg5_pK-4cP1=M0Ur_217=LsX+iq3zCBg@mail.gmail.com>

On Tue, Jul 17, 2012 at 11:42 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> On Mon, Jul 16, 2012 at 06:57:46PM +0200, Marc-Andr? Lureau wrote:
>> This flag will help to track wether the display has been
>
> whether
>
>> removed/closed and whether it have a really a valid display.
>
> the part starting at "whether" needs fixing. "Whether it really has a valid
> display" ?
>
>>
>> Ready in contrast, is used to "hide" temporarily the display (when
>> starting or redrawing the display, to avoid artifacts)
>> ---
>>  src/virt-viewer-app.c           |   15 +++++++++------
>>  src/virt-viewer-display-spice.c |   20 ++++++++++++++++++++
>>  src/virt-viewer-display.c       |   15 +++++++++++++++
>>  src/virt-viewer-display.h       |    2 ++
>>  src/virt-viewer-window.c        |   12 ++++++++++++
>>  5 files changed, 58 insertions(+), 6 deletions(-)
>>
>> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
>> index e5c674a..f581531 100644
>> --- a/src/virt-viewer-app.c
>> +++ b/src/virt-viewer-app.c
>> @@ -611,7 +611,8 @@ display_show_hint(VirtViewerDisplay *display,
>>  {
>>      VirtViewerApp *self;
>>      VirtViewerNotebook *nb = virt_viewer_window_get_notebook(win);
>> -    gint nth, hint;
>> +    gint nth;
>> +    guint hint;
>
> Doesn't this change belong to the patch turning 'hint' into a bitfield?
>

You are right, rebased with the spelling fixes.

thanks

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 12:15:08 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:15:08 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717121506.GZ2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:50PM +0200, Marc-Andr? Lureau wrote:
> Rely on spice-gtk display channel monitors property to manage
> displays. The same display channel may now provide several monitors,
> the SpiceDisplay widget must be told which monitor to display
> ---
>  src/virt-viewer-display-spice.c |   17 +++----
>  src/virt-viewer-display-spice.h |    2 +-
>  src/virt-viewer-session-spice.c |   98 +++++++++++++++++++++++++++++++++++----
>  3 files changed, 98 insertions(+), 19 deletions(-)
> 
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index 53430dd..ca74c1a 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -172,7 +172,7 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>  {
>      gdouble dw = allocation->width, dh = allocation->height;
>      guint zoom = 100;
> -    guint channelid;
> +    guint nth;
>  
>      if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
>          return;
> @@ -187,13 +187,11 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>          dh /= ((double)zoom / 100.0);
>      }
>  
> -    g_object_get(self->priv->channel, "channel-id", &channelid, NULL);
> +    g_object_get(self, "nth-display", &nth, NULL);
>  
>      SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
>          VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
> -    spice_main_set_display(main_channel,
> -                           channelid,
> -                           0, 0, dw, dh);
> +    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
>  }
>  
>  static void
> @@ -212,7 +210,8 @@ enable_accel_changed(VirtViewerApp *app,
>  
>  GtkWidget *
>  virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
> -                              SpiceChannel *channel)
> +                              SpiceChannel *channel,
> +                              gint monitorid)
>  {
>      VirtViewerDisplaySpice *self;
>      VirtViewerApp *app;
> @@ -222,15 +221,17 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>      g_return_val_if_fail(SPICE_IS_DISPLAY_CHANNEL(channel), NULL);
>  
>      g_object_get(channel, "channel-id", &channelid, NULL);
> +    // We don't allow monitorid != 0 && channelid != 0
> +    g_return_val_if_fail(channelid == 0 || monitorid == 0, NULL);
>  
>      self = g_object_new(VIRT_VIEWER_TYPE_DISPLAY_SPICE,
>                          "session", session,
> -                        "nth-display", channelid,
> +                        "nth-display", channelid + monitorid,

This looks like it's not necessarily unique with (channel id, monitor id)
being (0, 2) and (2, 0), can it an issue? Or are we guaranteed that this
situation cannot occur?
Update: Alon told me on IRC that the 2 situations are mutually exclusive,
can you add that to the comment?

>                          NULL);
>      self->priv->channel = channel;
>  
>      g_object_get(session, "spice-session", &s, NULL);
> -    self->priv->display = spice_display_new(s, channelid);
> +    self->priv->display = spice_display_new_with_monitor(s, channelid, monitorid);
>      g_object_unref(s);
>  
>      virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
> diff --git a/src/virt-viewer-display-spice.h b/src/virt-viewer-display-spice.h
> index 701ed85..c2013ec 100644
> --- a/src/virt-viewer-display-spice.h
> +++ b/src/virt-viewer-display-spice.h
> @@ -66,7 +66,7 @@ struct _VirtViewerDisplaySpiceClass {
>  
>  GType virt_viewer_display_spice_get_type(void);
>  
> -GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel);
> +GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel, gint monitorid);
>  
>  G_END_DECLS
>  
> diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
> index 6577237..0df11ef 100644
> --- a/src/virt-viewer-session-spice.c
> +++ b/src/virt-viewer-session-spice.c
> @@ -414,6 +414,80 @@ agent_connected_changed(SpiceChannel *cmain,
>  }
>  
>  static void
> +destroy_display(gpointer data)
> +{
> +    VirtViewerDisplay *display = VIRT_VIEWER_DISPLAY(data);
> +    VirtViewerSession *session = virt_viewer_display_get_session(display);
> +
> +    DEBUG_LOG("Destroying spice display %p", display);
> +    virt_viewer_session_remove_display(session, display);
> +    g_object_unref(display);
> +}
> +
> +static void
> +virt_viewer_session_spice_display_monitors(SpiceChannel *channel,
> +                                           GParamSpec *pspec G_GNUC_UNUSED,
> +                                           VirtViewerSessionSpice *self)
> +{
> +    GArray *monitors = NULL;
> +    GPtrArray *displays = NULL;
> +    GtkWidget *display;
> +    guint i, monitors_max;
> +
> +    g_object_get(channel,
> +                 "monitors", &monitors,
> +                 "monitors-max", &monitors_max,
> +                 NULL);
> +    g_return_if_fail(monitors != NULL);
> +    g_return_if_fail(monitors->len <= monitors_max);

Is monitors-max coming from the network? If so, should we make sure it's
not very huge to avoid allocating too much memory?

> +
> +    displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
> +    if (displays == NULL) {
> +        displays = g_ptr_array_new();
> +#if GLIB_CHECK_VERSION(2, 22, 0)
> +        g_ptr_array_set_free_func(displays, destroy_display);
> +        g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
> +                               displays, (GDestroyNotify)g_ptr_array_unref);
> +#endif

missing
#else
        g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
                               displays, NULL);
#endif
or something similar

VirtViewerSession already has its list of displays, do we really need to
maintain an additional list of displays and associate it with the
SpiceChannel object?

> +    }
> +
> +    g_ptr_array_set_size(displays, monitors_max);
> +
> +    for (i = 0; i < monitors_max; i++) {
> +        display = g_ptr_array_index(displays, i);
> +        if (display == NULL) {
> +            display = virt_viewer_display_spice_new(self, channel, i);
> +            DEBUG_LOG("creating spice display (#:%d)", i);
> +            g_ptr_array_index(displays, i) = g_object_ref(display);

The g_object_ref is not needed, this is leaking the initial ref from
_display_spice_new

> +        }
> +
> +        g_object_freeze_notify(G_OBJECT(display));
> +        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), FALSE);
> +        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(self),
> +                                        VIRT_VIEWER_DISPLAY(display));


> +    }
> +
> +    for (i = 0; i < monitors->len; i++) {
> +        SpiceDisplayMonitorConfig *c = &g_array_index(monitors, SpiceDisplayMonitorConfig, i);

Naming this variable 'monitor' would make things more readable imo.

> +        display = g_ptr_array_index(displays, c->id);
> +        g_return_if_fail(display != NULL);
> +
> +        if (c->width == 0 || c->width == 0)
> +            continue;
> +
> +        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), TRUE);
> +        virt_viewer_display_set_desktop_size(VIRT_VIEWER_DISPLAY(display),
> +                                             c->width, c->height);
> +    }
> +
> +    for (i = 0; i < monitors_max; i++)
> +        g_object_thaw_notify(g_ptr_array_index(displays, i));
> +
> +    g_clear_pointer(&monitors, g_array_unref);

This is very new in glib, do we have a fallback definition for this symbol?

> +
> +}
> +
> +static void
>  virt_viewer_session_spice_channel_new(SpiceSession *s,
>                                        SpiceChannel *channel,
>                                        VirtViewerSession *session)
> @@ -441,20 +515,17 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
>  
>          g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
>          agent_connected_changed(channel, NULL, self);
> +
> +        g_signal_emit_by_name(session, "session-connected");
>      }
>  
>      if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
> -        GtkWidget *display;
> -
> -        g_signal_emit_by_name(session, "session-connected");
> +        g_signal_emit_by_name(session, "session-initialized");
>  
> -        DEBUG_LOG("new display channel (#%d)", id);
> -        display = virt_viewer_display_spice_new(self, channel);
> -        g_object_set_data(G_OBJECT(channel), "virt-viewer-display", display);
> -        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
> -                                        VIRT_VIEWER_DISPLAY(display));
> +        g_signal_connect(channel, "notify::monitors",
> +                         G_CALLBACK(virt_viewer_session_spice_display_monitors), self);
>  
> -        g_signal_emit_by_name(session, "session-initialized");
> +        spice_channel_connect(channel);
>      }
>  
>      if (SPICE_IS_INPUTS_CHANNEL(channel)) {
> @@ -538,7 +609,14 @@ virt_viewer_session_spice_channel_destroy(G_GNUC_UNUSED SpiceSession *s,
>      if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
>          VirtViewerDisplay *display = g_object_get_data(G_OBJECT(channel), "virt-viewer-display");
>          DEBUG_LOG("zap display channel (#%d, %p)", id, display);
> -        virt_viewer_session_remove_display(session, display);
> +#if !GLIB_CHECK_VERSION(2, 22, 0)
> +        GPtrArray *displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
> +        if (displays) {
> +            g_ptr_array_foreach(displays, (GFunc)destroy_display, NULL);
> +            g_ptr_array_free(displays, TRUE);
> +            g_object_set_data(channel, "virt-viewer-displays", NULL);
> +        }
> +#endif
>      }
>  
>      if (SPICE_IS_PLAYBACK_CHANNEL(channel) && self->priv->audio) {
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/8928b485/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:22:58 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:22:58 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 12/19] Prevent from adding
 the same display several time in the session
In-Reply-To: <CAJ+F1C++rRPDB0a4FEjk4UeDmprG=AL-FzeqzsY8Vju39nd=vw@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-13-git-send-email-marcandre.lureau@gmail.com>
	<20120717094319.GX2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C++rRPDB0a4FEjk4UeDmprG=AL-FzeqzsY8Vju39nd=vw@mail.gmail.com>
Message-ID: <20120717122258.GA2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 01:14:27PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 11:43 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > How can this happen?
> 
> In virt_viewer_session_spice_display_monitors(), we (re)add the
> display unconditionnaly every time we receive a new MonitorConfig.

Yep, saw that after reading the next patches, just add that to the commit
log (most of my 'why?' questions after the commit log just mean that a more
detailed changelog would be welcome).

Christophe

> 
> -- 
> Marc-Andr? Lureau
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/a03dfda3/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:24:25 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:24:25 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 03/19] spice: disconnect
 signal handlers when either object is destroyed
In-Reply-To: <CAJ+F1C+OsagLgTB=xmczGT+1O3Zec2ePQuRPTXP4VmhEShPGDA@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-4-git-send-email-marcandre.lureau@gmail.com>
	<20120717083723.GN2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C+OsagLgTB=xmczGT+1O3Zec2ePQuRPTXP4VmhEShPGDA@mail.gmail.com>
Message-ID: <20120717122425.GB2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 12:50:36PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 10:37 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > Why do you need to do this? Are you seeing concrete issues? Or will it be
> > needed by subsequent changes?
> 
> "This function
> will take care of removing signal handler if any of emitter or
> attached object are destroyed."

Yes, I understand what it does, I asked you _why_ it's needed.

> 
> The following patches will have this condition met, since there is no
> longer 1-1 relation between channel and display. The channels can
> continue to be around when some of the display are removed.

add this explanation to the commit log then ;)

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/1dc5aeb0/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:26:53 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:26:53 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/19] Number display
 starting from 1
In-Reply-To: <CAJ+F1CJmMVoBp_bmFBBunkumqxq=hE=NyDD8v8GJvCmH=7oSnQ@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>
	<20120717094604.GY2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJmMVoBp_bmFBBunkumqxq=hE=NyDD8v8GJvCmH=7oSnQ@mail.gmail.com>
Message-ID: <20120717122653.GC2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 01:16:19PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 11:46 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > Why? To match the window subtitle? Please add this to the commit log.
> 
> It just is more user friendly, no? I think human start counting from 1
> in general.

I don't disagree with the change, just curious as to why it's needed since
the commit log doesn't mention it.

> 
> > Id add a +1 as well to
> > DEBUG_LOG("Remove window %d %p", nth, win);
> > and
> > DEBUG_LOG("Insert window %d %p", nth, win);
> 
> However, this is not meant to be user friendly, but rather developper
> friendly, so we should just print the value of these variable, imho

This means that when the GUI talks about display 1 and 2, and the debug log
talks about window 1, you have to know it's really about display 2, imo
this is not developer friendly.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/6eb49d8b/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 12:38:15 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 14:38:15 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717121506.GZ2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>

Hi

On Tue, Jul 17, 2012 at 2:15 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> On Mon, Jul 16, 2012 at 06:57:50PM +0200, Marc-Andr? Lureau wrote:
>> Rely on spice-gtk display channel monitors property to manage
>> displays. The same display channel may now provide several monitors,
>> the SpiceDisplay widget must be told which monitor to display
>> ---
>>  src/virt-viewer-display-spice.c |   17 +++----
>>  src/virt-viewer-display-spice.h |    2 +-
>>  src/virt-viewer-session-spice.c |   98 +++++++++++++++++++++++++++++++++++----
>>  3 files changed, 98 insertions(+), 19 deletions(-)
>>
>> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
>> index 53430dd..ca74c1a 100644
>> --- a/src/virt-viewer-display-spice.c
>> +++ b/src/virt-viewer-display-spice.c
>> @@ -172,7 +172,7 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>>  {
>>      gdouble dw = allocation->width, dh = allocation->height;
>>      guint zoom = 100;
>> -    guint channelid;
>> +    guint nth;
>>
>>      if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
>>          return;
>> @@ -187,13 +187,11 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>>          dh /= ((double)zoom / 100.0);
>>      }
>>
>> -    g_object_get(self->priv->channel, "channel-id", &channelid, NULL);
>> +    g_object_get(self, "nth-display", &nth, NULL);
>>
>>      SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
>>          VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
>> -    spice_main_set_display(main_channel,
>> -                           channelid,
>> -                           0, 0, dw, dh);
>> +    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
>>  }
>>
>>  static void
>> @@ -212,7 +210,8 @@ enable_accel_changed(VirtViewerApp *app,
>>
>>  GtkWidget *
>>  virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>> -                              SpiceChannel *channel)
>> +                              SpiceChannel *channel,
>> +                              gint monitorid)
>>  {
>>      VirtViewerDisplaySpice *self;
>>      VirtViewerApp *app;
>> @@ -222,15 +221,17 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>>      g_return_val_if_fail(SPICE_IS_DISPLAY_CHANNEL(channel), NULL);
>>
>>      g_object_get(channel, "channel-id", &channelid, NULL);
>> +    // We don't allow monitorid != 0 && channelid != 0
>> +    g_return_val_if_fail(channelid == 0 || monitorid == 0, NULL);
>>
>>      self = g_object_new(VIRT_VIEWER_TYPE_DISPLAY_SPICE,
>>                          "session", session,
>> -                        "nth-display", channelid,
>> +                        "nth-display", channelid + monitorid,
>
> This looks like it's not necessarily unique with (channel id, monitor id)
> being (0, 2) and (2, 0), can it an issue? Or are we guaranteed that this
> situation cannot occur?
> Update: Alon told me on IRC that the 2 situations are mutually exclusive,
> can you add that to the comment?

added a comment

>>                          NULL);
>>      self->priv->channel = channel;
>>
>>      g_object_get(session, "spice-session", &s, NULL);
>> -    self->priv->display = spice_display_new(s, channelid);
>> +    self->priv->display = spice_display_new_with_monitor(s, channelid, monitorid);
>>      g_object_unref(s);
>>
>>      virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
>> diff --git a/src/virt-viewer-display-spice.h b/src/virt-viewer-display-spice.h
>> index 701ed85..c2013ec 100644
>> --- a/src/virt-viewer-display-spice.h
>> +++ b/src/virt-viewer-display-spice.h
>> @@ -66,7 +66,7 @@ struct _VirtViewerDisplaySpiceClass {
>>
>>  GType virt_viewer_display_spice_get_type(void);
>>
>> -GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel);
>> +GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel, gint monitorid);
>>
>>  G_END_DECLS
>>
>> diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
>> index 6577237..0df11ef 100644
>> --- a/src/virt-viewer-session-spice.c
>> +++ b/src/virt-viewer-session-spice.c
>> @@ -414,6 +414,80 @@ agent_connected_changed(SpiceChannel *cmain,
>>  }
>>
>>  static void
>> +destroy_display(gpointer data)
>> +{
>> +    VirtViewerDisplay *display = VIRT_VIEWER_DISPLAY(data);
>> +    VirtViewerSession *session = virt_viewer_display_get_session(display);
>> +
>> +    DEBUG_LOG("Destroying spice display %p", display);
>> +    virt_viewer_session_remove_display(session, display);
>> +    g_object_unref(display);
>> +}
>> +
>> +static void
>> +virt_viewer_session_spice_display_monitors(SpiceChannel *channel,
>> +                                           GParamSpec *pspec G_GNUC_UNUSED,
>> +                                           VirtViewerSessionSpice *self)
>> +{
>> +    GArray *monitors = NULL;
>> +    GPtrArray *displays = NULL;
>> +    GtkWidget *display;
>> +    guint i, monitors_max;
>> +
>> +    g_object_get(channel,
>> +                 "monitors", &monitors,
>> +                 "monitors-max", &monitors_max,
>> +                 NULL);
>> +    g_return_if_fail(monitors != NULL);
>> +    g_return_if_fail(monitors->len <= monitors_max);
>
> Is monitors-max coming from the network? If so, should we make sure it's
> not very huge to avoid allocating too much memory?

I don't know what would be very huge, probably >16 starts to be
*really* a lot already, but if it's what's the server has..

>> +
>> +    displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
>> +    if (displays == NULL) {
>> +        displays = g_ptr_array_new();
>> +#if GLIB_CHECK_VERSION(2, 22, 0)
>> +        g_ptr_array_set_free_func(displays, destroy_display);
>> +        g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
>> +                               displays, (GDestroyNotify)g_ptr_array_unref);
>> +#endif
>
> missing
> #else
>         g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
>                                displays, NULL);
> #endif
> or something similar

Yes, corrected. I wonder if it wouldn't be simpler to have only the
fallback version.. What do you think?

>
> VirtViewerSession already has its list of displays, do we really need to
> maintain an additional list of displays and associate it with the
> SpiceChannel object?

They are not necessarily correlated. We need a list of display per
channel, not per virt viewer session. The session could have fewer
displays, or perhaps more in the future (if it later supports both
monitorconfig and multiple display channels)

>> +    }
>> +
>> +    g_ptr_array_set_size(displays, monitors_max);
>> +
>> +    for (i = 0; i < monitors_max; i++) {
>> +        display = g_ptr_array_index(displays, i);
>> +        if (display == NULL) {
>> +            display = virt_viewer_display_spice_new(self, channel, i);
>> +            DEBUG_LOG("creating spice display (#:%d)", i);
>> +            g_ptr_array_index(displays, i) = g_object_ref(display);
>
> The g_object_ref is not needed, this is leaking the initial ref from
> _display_spice_new

The display widget is GInitiallyUnowned, and will get sink ref by the
parent widget. We need an additional ref for the book-keeping here.
>> +    for (i = 0; i < monitors->len; i++) {
>> +        SpiceDisplayMonitorConfig *c = &g_array_index(monitors, SpiceDisplayMonitorConfig, i);
>
> Naming this variable 'monitor' would make things more readable imo.

ack
>> +    g_clear_pointer(&monitors, g_array_unref);
>
> This is very new in glib, do we have a fallback definition for this symbol?

No, adding it to glib-compat.h.


-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 12:39:06 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:39:06 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn
 display:show-hint into flags type
In-Reply-To: <CAJ+F1CJSFbUbAdVAq_NBhOJKbQTqko7Wkr=_i4cdvXX-Yr+42Q@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
	<20120717085912.GR2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJSFbUbAdVAq_NBhOJKbQTqko7Wkr=_i4cdvXX-Yr+42Q@mail.gmail.com>
Message-ID: <20120717123906.GD2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 12:59:22PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 10:59 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > On second thought, shouldn't this be if (!(hint & _HINT_READY)) { } ? I'm
> > not sure which flags you are going to add.
> 
> VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE used to be an enum of value 0, now
> there is only a READY flag. The following patch will make this easier
> to read.

I've assumed that _READY/_HIDE were just 2 named constant for the same
concept, one name for TRUE, another one for FALSE, hence my question. If
they are 2 different things, then it would be less confusing to have a HIDE
flag as well. Though it's not really important as it's indeed changed in
one of the next patches.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/191c5b1c/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:41:09 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:41:09 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/19] Turn
 display:show-hint into flags type
In-Reply-To: <CAJ+F1CKoMkzvtL-n=UMzcQ0rFNf5vMia4NX9wjwoBm6vpSg0DQ@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-7-git-send-email-marcandre.lureau@gmail.com>
	<20120717085740.GQ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKoMkzvtL-n=UMzcQ0rFNf5vMia4NX9wjwoBm6vpSg0DQ@mail.gmail.com>
Message-ID: <20120717124109.GE2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 12:57:24PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 10:57 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > I would have kept HINT_HIDE as the flag name as it's more
> > expressive than _READY imo. Looks good either way.
> 
> "HIDE" isn't necessarily 0 anymore, there will be a DISABLED flag in a
> following patch. Now that you say it, it could have been "ENABLED"
> flag instead, so 0 would still map to !READY && !ENABLED, but I prefer
> to DISABLED, since this is the exception, and we might had more flags
> in the future, making 0 hard to keep meaningful perhaps.

Yes, that's what I meant. At this point in the series, there is no DISABLED
flag, so I just assume that other flags will have totally different
meanings that are unrelated to showing/hiding the window, so they could be
set while we want to hide the window, or things like that.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/4fc3041c/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:45:27 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:45:27 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 10/19] Use
 virt_viewer_connect_object() for display
In-Reply-To: <CAJ+F1C+=Wr5ApDTK5oS9qV_R7z27hrp=+LHH4dHDvbLJa5vFbA@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-11-git-send-email-marcandre.lureau@gmail.com>
	<20120717092619.GV2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C+=Wr5ApDTK5oS9qV_R7z27hrp=+LHH4dHDvbLJa5vFbA@mail.gmail.com>
Message-ID: <20120717124527.GF2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 01:11:50PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 11:26 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > On Mon, Jul 16, 2012 at 06:57:45PM +0200, Marc-Andr? Lureau wrote:
> >> When display is released, detach signal automatically
> >
> > Same question as the similar patch before, why? I'm assuming that it's
> > because we are going to handle displays dynamically, and if we don't do
> > that, we'll sometimes end up with signals using dead objects?
> 
> Yeah, I don't have a backtrace, but you can get into various crash
> related to not cleaning up signal handlers properly. 

Just add this to the commit log.

> In general, this
> signal_connect helper is really awesome, as you don't have to worry
> about either emitter or receiver being dead, but some time you may
> prefer to actually crash if something goes wrong. So a systematic
> replace isn't what I would recommend. But for Spice channel & display,
> it is needed to cleanup if either of the two is destroyed.

I agree using it makes sense, but it's useful for the review, or for future
bisections to know if it fixes a bug, or if it's in preparation for future
changes, ...
As far as I know, signal handlers will get cleaned up when the emitter goes
away, so the main advantage of this function is of being a working
g_signal_connect_object implementation which disconnects the signal handler
when user_data goes away.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/171fa59b/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:52:31 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:52:31 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
Message-ID: <20120717125231.GG2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 02:38:15PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 2:15 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > On Mon, Jul 16, 2012 at 06:57:50PM +0200, Marc-Andr? Lureau wrote:
> >> +    g_object_get(channel,
> >> +                 "monitors", &monitors,
> >> +                 "monitors-max", &monitors_max,
> >> +                 NULL);
> >> +    g_return_if_fail(monitors != NULL);
> >> +    g_return_if_fail(monitors->len <= monitors_max);
> >
> > Is monitors-max coming from the network? If so, should we make sure it's
> > not very huge to avoid allocating too much memory?
> 
> I don't know what would be very huge, probably >16 starts to be
> *really* a lot already, but if it's what's the server has..

I'm concerned about malicious payload putting a huge number there for a
nasty purpose. Moreover, is this value coming from the server, or is it
coming from the qxl driver in the guest?


> Yes, corrected. I wonder if it wouldn't be simpler to have only the
> fallback version.. What do you think?

This would get rid of various #ifdef so this sounds good to me.

> >> +    g_clear_pointer(&monitors, g_array_unref);
> >
> > This is very new in glib, do we have a fallback definition for this symbol?
> 
> No, adding it to glib-compat.h.

I just realized that g_array_unref is glib 2.22+

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/dfd5d843/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 12:54:35 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 14:54:35 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 07/19] Run-time check
 values before doing bad computation
In-Reply-To: <CAJ+F1C+60OCuircobDcSgixL+Q0xHhgo8WGT6UnmBiPzttHJfA@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-8-git-send-email-marcandre.lureau@gmail.com>
	<20120717091008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C+60OCuircobDcSgixL+Q0xHhgo8WGT6UnmBiPzttHJfA@mail.gmail.com>
Message-ID: <20120717125435.GH2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 01:03:57PM +0200, Marc-Andr? Lureau wrote:
> Hi
> 
> On Tue, Jul 17, 2012 at 11:10 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> >> +    if (priv->desktopWidth == 0 ||
> >> +        priv->desktopHeight == 0)
> >> +        goto end;
> >
> > Why can this happen?
> 
> It the spice display doesn't know, for example, we don't yet have
> monitor geometry.
> 
> >> +    g_return_if_fail(fullscreen.height > 128);
> >> +    g_return_if_fail(fullscreen.width > 128);
> >> +    g_return_if_fail(desktopWidth > 0);
> >> +    g_return_if_fail(desktopHeight > 0);
> >
> > In size_allocate 0 width/height are silently ignored, here you return with
> > a warning, is it on purpose?
> 
> This is the following code:
> 
>     desktopAspect = (double)desktopWidth / (double)desktopHeight;
>     screenAspect = (double)(fullscreen.width - 128) /
> (double)(fullscreen.height - 128);
> 
>     if ((desktopWidth > (fullscreen.width - 128)) ||
>         (desktopHeight > (fullscreen.height - 128))) {
> 
> I wanted a bit of run-time check, in practice we don't seem to hit
> these condition, but you can see it would be nice to have a warning if
> it happens.

Yes I've seen why you wanted these checks, I was just curious about the
difference between the first hunk and this one.

> 
> -- 
> Marc-Andr? Lureau
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/530c6e64/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 12:59:44 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 14:59:44 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717125231.GG2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJeOgx4e_oPSC53njg6+oAGR4P-g1sjBt-sdgP2qJ_DnA@mail.gmail.com>

On Tue, Jul 17, 2012 at 2:52 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> I just realized that g_array_unref is glib 2.22+

Glib 2.22 is in rhel6, so it's probably okay, although we need to bump
configure.ac version check

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 13:01:26 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 15:01:26 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 16/19] Wait until idle
 handler has run to mark dirty = FALSE
In-Reply-To: <1342457874-15369-17-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-17-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717130126.GI2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:51PM +0200, Marc-Andr? Lureau wrote:
> This can avoid having multiple idler added
> ---
>  src/virt-viewer-display.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 18fed19..004f027 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -326,6 +326,7 @@ virt_viewer_display_idle(gpointer opaque)
>      VirtViewerDisplayPrivate *priv = display->priv;
>      if (!priv->dirty)
>          gtk_widget_queue_resize_no_redraw(GTK_WIDGET(display));
> +    priv->dirty = FALSE;

priv->dirty is only set to FALSE in the place you changed, this means that
after your change the
if (!priv->dirty)
    gtk_widget_queue_resize_no_redraw(GTK_WIDGET(display));
just above will never run.
I think the intent of this code is this:
'queue an idle to do the resize, but use dirty to know about size changes
happening between the time the idle was queued and the time the idle runs'

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/8d542ddc/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 13:02:46 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 15:02:46 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717125231.GG2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>

On Tue, Jul 17, 2012 at 2:52 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> I'm concerned about malicious payload putting a huge number there for a
> nasty purpose. Moreover, is this value coming from the server, or is it
> coming from the qxl driver in the guest?


I don't think this is a concern here. You might worry about a lot of
other parts of spice then.. In general, it can be very hard to verify
integrity, and I guess we rely on lower level of the stack to do that
for us.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 13:08:55 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 15:08:55 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/19] Number display
 starting from 1
In-Reply-To: <20120717122653.GC2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>
	<20120717094604.GY2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJmMVoBp_bmFBBunkumqxq=hE=NyDD8v8GJvCmH=7oSnQ@mail.gmail.com>
	<20120717122653.GC2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CL2goPnYiMmhWxRDsvTfgmnjGp7_FDK_F4v4=FuBYWfKA@mail.gmail.com>

On Tue, Jul 17, 2012 at 2:26 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> This means that when the GUI talks about display 1 and 2, and the debug log
> talks about window 1, you have to know it's really about display 2, imo
> this is not developer friendly.

We have to disagree on that, my point is that a developper is more
interested by true variable values rather than something
user-friendly.

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 13:34:42 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 15:34:42 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 17/19] Make display
 submenu insensitive if they can't be modified
In-Reply-To: <1342457874-15369-18-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-18-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717133442.GJ2162@teriyaki.cdg.redhat.com>

This patch is quite long, but it could have easily been split in trivial
standalone patches...


- 'they' in the subject should be 'displays'

On Mon, Jul 16, 2012 at 06:57:52PM +0200, Marc-Andr? Lureau wrote:
> Only when the agent is running may a display be enabled/disabled.
> ---
>  src/virt-viewer-app.c           |   28 ++++++++++++++++++++++++++-
>  src/virt-viewer-display-spice.c |   40 ++++++++++++++++++++++++++++++++-------
>  src/virt-viewer-display.c       |   23 ++++++++++++++++++++++
>  src/virt-viewer-display.h       |    2 ++
>  src/virt-viewer-session-spice.c |   18 ++++++++++++++----
>  src/virt-viewer-session.c       |   10 ++++++++++
>  src/virt-viewer-session.h       |    1 +
>  src/virt-viewer-window.c        |   13 +++++++++++--
>  src/virt-viewer-window.h        |    1 +
>  9 files changed, 122 insertions(+), 14 deletions(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index f590641..def52c5 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -661,6 +661,7 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
>      }
>  
>      virt_viewer_window_set_display(window, display);
> +    virt_viewer_app_update_menu_displays(self);

This bit could go in "Rebuild display menu when a window is added or removed" I think

>      virt_viewer_signal_connect_object(display, "notify::show-hint",
>                                        G_CALLBACK(display_show_hint), window, 0);
>      g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */
> @@ -684,6 +685,13 @@ virt_viewer_app_display_removed(VirtViewerSession *session G_GNUC_UNUSED,
>          virt_viewer_app_remove_nth_window(self, nth);
>  }
>  
> +static void
> +virt_viewer_app_display_updated(VirtViewerSession *session G_GNUC_UNUSED,
> +                                VirtViewerApp *self)
> +{
> +    virt_viewer_app_update_menu_displays(self);
> +}
> +
>  int
>  virt_viewer_app_create_session(VirtViewerApp *self, const gchar *type)
>  {
> @@ -734,6 +742,8 @@ virt_viewer_app_create_session(VirtViewerApp *self, const gchar *type)
>                       G_CALLBACK(virt_viewer_app_display_added), self);
>      g_signal_connect(priv->session, "session-display-removed",
>                       G_CALLBACK(virt_viewer_app_display_removed), self);
> +    g_signal_connect(priv->session, "session-display-updated",
> +                     G_CALLBACK(virt_viewer_app_display_updated), self);

The addition of this signal + the updating of the displays menu is not
strictly related to setting the sensitivity of the menu either.

>  
>      g_signal_connect(priv->session, "session-cut-text",
>                       G_CALLBACK(virt_viewer_app_server_cut_text), self);
> @@ -1634,15 +1644,31 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
>      while (tmp) {
>          int *nth = tmp->data;
>          VirtViewerWindow *vwin = VIRT_VIEWER_WINDOW(g_hash_table_lookup(self->priv->windows, nth));
> +        VirtViewerDisplay *display = virt_viewer_window_get_display(vwin);
>          GtkWidget *item;
> -        gboolean visible;
> +        gboolean visible, sensitive = FALSE;
>          gchar *label;
>  
>          label = g_strdup_printf(_("Display %d"), *nth + 1);
>          item = gtk_check_menu_item_new_with_label(label);
>          g_free(label);
> +
>          visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
>          gtk_check_menu_item_set_active(GTK_CHECK_MENU_ITEM(item), visible);
> +
> +        if (display) {
> +            guint hint = virt_viewer_display_get_show_hint(display);
> +
> +            if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY)
> +                sensitive = TRUE;
> +
> +            if ((hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED) &&
> +                virt_viewer_display_get_selectable(display))
> +                sensitive = TRUE;
> +        }
> +
> +        gtk_widget_set_sensitive(item, sensitive);
> +

This is the gist of this change, and with it, disabled screens will show in
the display menu, and they are clickable in order to enable them?



>          g_signal_connect(G_OBJECT(item),
>                           "toggled", G_CALLBACK(menu_display_visible_toggled_cb), vwin);
>          gtk_menu_shell_append(submenu, item);
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index ca74c1a..a91d73e 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -47,6 +47,7 @@ static void virt_viewer_display_spice_send_keys(VirtViewerDisplay *display,
>  static GdkPixbuf *virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display);
>  static void virt_viewer_display_spice_release_cursor(VirtViewerDisplay *display);
>  static void virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED);
> +static gboolean virt_viewer_display_spice_selectable(VirtViewerDisplay *display);
>  
>  static void
>  virt_viewer_display_spice_finalize(GObject *obj)
> @@ -70,23 +71,36 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
>      dclass->get_pixbuf = virt_viewer_display_spice_get_pixbuf;
>      dclass->release_cursor = virt_viewer_display_spice_release_cursor;
>      dclass->close = virt_viewer_display_spice_close;
> +    dclass->selectable = virt_viewer_display_spice_selectable;
>  
>      g_type_class_add_private(klass, sizeof(VirtViewerDisplaySpicePrivate));
>  }
>  
> +static SpiceMainChannel*
> +get_main(VirtViewerDisplay *self)
> +{
> +    VirtViewerSessionSpice *session;
> +
> +    session = VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self));
> +
> +    return virt_viewer_session_spice_get_main_channel(session);
> +}
> +
>  static void
>  show_hint_changed(VirtViewerDisplay *self)
>  {
> -    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
> -        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self)));
> +    SpiceMainChannel *cmain = get_main(self);
>      guint enabled = TRUE;
>      guint nth;
>  
> +    if (!cmain)
> +        return;
> +
>      g_object_get(self, "nth-display", &nth, NULL);
>      if (virt_viewer_display_get_show_hint(self) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
>          enabled = FALSE;
>  
> -    spice_main_set_display_enabled(main_channel, nth, enabled);
> +    spice_main_set_display_enabled(cmain, nth, enabled);
>  }

>  
>  static void
> @@ -189,9 +203,8 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>  
>      g_object_get(self, "nth-display", &nth, NULL);
>  
> -    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
> -        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
> -    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
> +    spice_main_set_display(get_main(VIRT_VIEWER_DISPLAY(self)),
> +                           nth, 0, 0, dw, dh);
>  }

These get_main() changes do not really belong in this patch, they could be
preparatory cleanups (this patch is big enough as it is). And main_channel
is a better name than mainc.


>  
>  static void
> @@ -270,12 +283,25 @@ virt_viewer_display_spice_release_cursor(VirtViewerDisplay *display)
>      spice_display_mouse_ungrab(self->priv->display);
>  }
>  
> -
>  static void
>  virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED)
>  {
>  }
>  
> +static gboolean
> +virt_viewer_display_spice_selectable(VirtViewerDisplay *self)
> +{
> +    gboolean agent_connected;
> +    SpiceMainChannel *mainc;
> +
> +    mainc = get_main(self);
> +    g_object_get(mainc,
> +                 "agent-connected", &agent_connected,
> +                 NULL);
> +
> +    return agent_connected;
> +}
> +
>  
>  /*
>   * Local variables:
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 004f027..4612970 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -80,6 +80,7 @@ enum {
>      PROP_ZOOM_LEVEL,
>      PROP_SHOW_HINT,
>      PROP_SESSION,
> +    PROP_SELECTABLE,
>  };
>  
>  static void
> @@ -167,6 +168,13 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
>                                                          G_PARAM_READWRITE |
>                                                          G_PARAM_CONSTRUCT_ONLY));
>  
> +    g_object_class_install_property(object_class,
> +                                    PROP_SELECTABLE,
> +                                    g_param_spec_boolean("selectable",
> +                                                         "Selectable",
> +                                                         "Selectable",
> +                                                         FALSE,
> +                                                         G_PARAM_READABLE));
>  
>      g_signal_new("display-pointer-grab",
>                   G_OBJECT_CLASS_TYPE(object_class),
> @@ -302,6 +310,9 @@ virt_viewer_display_get_property(GObject *object,
>      case PROP_SESSION:
>          g_value_set_object(value, virt_viewer_display_get_session(display));
>          break;
> +    case PROP_SELECTABLE:
> +        g_value_set_boolean(value, virt_viewer_display_get_selectable(display));
> +        break;
>  
>      default:
>          G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
> @@ -601,6 +612,18 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *self)
>      klass->release_cursor(self);
>  }
>  
> +gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *self)
> +{
> +    VirtViewerDisplayClass *klass;
> +
> +    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), FALSE);
> +
> +    klass = VIRT_VIEWER_DISPLAY_GET_CLASS(self);
> +    if (klass->selectable)
> +        return klass->selectable(self);
> +
> +    return FALSE;
> +}
>  
>  void virt_viewer_display_close(VirtViewerDisplay *self)
>  {
> diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
> index 89b4817..4247570 100644
> --- a/src/virt-viewer-display.h
> +++ b/src/virt-viewer-display.h
> @@ -77,6 +77,7 @@ struct _VirtViewerDisplayClass {
>      void (*release_cursor)(VirtViewerDisplay *display);
>  
>      void (*close)(VirtViewerDisplay *display);
> +    gboolean (*selectable)(VirtViewerDisplay *display);
>  
>      /* signals */
>      void (*display_pointer_grab)(VirtViewerDisplay *display);
> @@ -118,6 +119,7 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
>  
>  void virt_viewer_display_close(VirtViewerDisplay *display);
>  void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
> +gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *display);

The addition of this property could be split as well.

>  
>  G_END_DECLS
>  
> diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
> index 0df11ef..6e9c03d 100644
> --- a/src/virt-viewer-session-spice.c
> +++ b/src/virt-viewer-session-spice.c
> @@ -405,12 +405,21 @@ virt_viewer_session_spice_usb_device_selection(VirtViewerSession *session,
>  }
>  
>  static void
> -agent_connected_changed(SpiceChannel *cmain,
> +agent_connected_changed(SpiceChannel *cmain G_GNUC_UNUSED,
>                          GParamSpec *pspec G_GNUC_UNUSED,
>                          VirtViewerSessionSpice *self)
>  {
> +    // this will force refresh of application menu
> +    g_signal_emit_by_name(self, "session-display-updated");
> +}
> +
> +static void
> +agent_connected_fullscreen_auto_conf(SpiceChannel *cmain,
> +                                     GParamSpec *pspec G_GNUC_UNUSED,
> +                                     VirtViewerSessionSpice *self)
> +{
>      if (virt_viewer_session_spice_fullscreen_auto_conf(self))
> -        g_signal_handlers_disconnect_by_func(cmain, agent_connected_changed, self);
> +        g_signal_handlers_disconnect_by_func(cmain, agent_connected_fullscreen_auto_conf, self);
>  }
>  
>  static void
> @@ -513,8 +522,9 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
>                           G_CALLBACK(virt_viewer_session_spice_main_channel_event), self);
>          self->priv->main_channel = SPICE_MAIN_CHANNEL(channel);
>  
> -        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
> -        agent_connected_changed(channel, NULL, self);
> +        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed), self);
> +        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_fullscreen_auto_conf), self);
> +        agent_connected_fullscreen_auto_conf(channel, NULL, self);
>  
>          g_signal_emit_by_name(session, "session-connected");
>      }
 diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
> index 9249a1f..a1d96c2 100644
> --- a/src/virt-viewer-session.c
> +++ b/src/virt-viewer-session.c
> @@ -233,6 +233,16 @@ virt_viewer_session_class_init(VirtViewerSessionClass *class)
>                   1,
>                   VIRT_VIEWER_TYPE_DISPLAY);
>  
> +    g_signal_new("session-display-updated",
> +                 G_OBJECT_CLASS_TYPE(object_class),
> +                 G_SIGNAL_RUN_LAST | G_SIGNAL_NO_HOOKS,
> +                 G_STRUCT_OFFSET(VirtViewerSessionClass, session_display_updated),
> +                 NULL,
> +                 NULL,
> +                 g_cclosure_marshal_VOID__VOID,
> +                 G_TYPE_NONE,
> +                 0);
> +
>      g_signal_new("session-cut-text",
>                   G_OBJECT_CLASS_TYPE(object_class),
>                   G_SIGNAL_RUN_LAST | G_SIGNAL_NO_HOOKS,
> diff --git a/src/virt-viewer-session.h b/src/virt-viewer-session.h
> index 44e4674..38ed988 100644
> --- a/src/virt-viewer-session.h
> +++ b/src/virt-viewer-session.h
> @@ -88,6 +88,7 @@ struct _VirtViewerSessionClass {
>                                    VirtViewerDisplay *display);
>      void (*session_display_removed)(VirtViewerSession *session,
>                                      VirtViewerDisplay *display);
> +    void (*session_display_updated)(VirtViewerSession *session);
>  
>      void (*session_cut_text)(VirtViewerSession *session, const gchar *str);
>      void (*session_bell)(VirtViewerSession *session);
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 46e44be..4bc427a 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -112,7 +112,8 @@ static void
>  virt_viewer_window_get_property (GObject *object, guint property_id,
>                                   GValue *value, GParamSpec *pspec)
>  {
> -    VirtViewerWindowPrivate *priv = VIRT_VIEWER_WINDOW(object)->priv;
> +    VirtViewerWindow *self = VIRT_VIEWER_WINDOW(object);
> +    VirtViewerWindowPrivate *priv = self->priv;
>  
>      switch (property_id) {
>      case PROP_SUBTITLE:
> @@ -124,7 +125,7 @@ virt_viewer_window_get_property (GObject *object, guint property_id,
>          break;
>  
>      case PROP_DISPLAY:
> -        g_value_set_object(value, priv->display);
> +        g_value_set_object(value, virt_viewer_window_get_display(self));
>          break;
>  
>      case PROP_CONTAINER:
> @@ -1117,6 +1118,14 @@ virt_viewer_window_get_builder(VirtViewerWindow *self)
>      return self->priv->builder;
>  }
>  
> +VirtViewerDisplay*
> +virt_viewer_window_get_display(VirtViewerWindow *self)
> +{
> +    g_return_val_if_fail(VIRT_VIEWER_WINDOW(self), FALSE);
> +
> +    return self->priv->display;
> +}
> +

Another change that could stand on its own.

>  /*
>   * Local variables:
>   *  c-indent-level: 4
> diff --git a/src/virt-viewer-window.h b/src/virt-viewer-window.h
> index f80c4c7..44db585 100644
> --- a/src/virt-viewer-window.h
> +++ b/src/virt-viewer-window.h
> @@ -63,6 +63,7 @@ GType virt_viewer_window_get_type (void);
>  GtkWindow* virt_viewer_window_get_window (VirtViewerWindow* window);
>  VirtViewerNotebook* virt_viewer_window_get_notebook (VirtViewerWindow* window);
>  void virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *display);
> +VirtViewerDisplay* virt_viewer_window_get_display(VirtViewerWindow *self);
>  void virt_viewer_window_set_usb_options_sensitive(VirtViewerWindow *self, gboolean sensitive);
>  void virt_viewer_window_update_title(VirtViewerWindow *self);
>  void virt_viewer_window_show(VirtViewerWindow *self);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/e088ae86/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 13:41:46 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 15:41:46 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 18/19] Simplify display
 flag handling
In-Reply-To: <1342457874-15369-19-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-19-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717134146.GK2162@teriyaki.cdg.redhat.com>

Looks good, I'd have put this much earlier in the series, but no
big deal.
ACK, a few comments below.

On Mon, Jul 16, 2012 at 06:57:53PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-display-spice.c |    9 +--------
>  src/virt-viewer-display.c       |   20 ++++++++++----------
>  src/virt-viewer-display.h       |    2 +-
>  src/virt-viewer-session-vnc.c   |    8 ++++----
>  4 files changed, 16 insertions(+), 23 deletions(-)
> 
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index a91d73e..bbc1773 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -140,19 +140,12 @@ display_ready(GObject *display,
>                GParamSpec *pspec G_GNUC_UNUSED,
>                VirtViewerDisplay *self)
>  {
> -    guint hint;
>      gboolean ready;
>  
>      g_object_get(display, "ready", &ready, NULL);
>      DEBUG_LOG("display %p ready:%d", self, ready);
>  
> -    hint = virt_viewer_display_get_show_hint(self);
> -    if (ready)
> -        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
> -    else
> -        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
> -
> -    virt_viewer_display_set_show_hint(self, hint);
> +    virt_viewer_display_set_show_hint(self, VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, ready);
>  }
>  
>  static void
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 4612970..b8e6d67 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -551,12 +551,20 @@ guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
>      return self->priv->show_hint;
>  }
>  
> -void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
> +void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint mask, gboolean enabled)

maybe s/enabled/enable since the flag is not enabled yet.

>  {
>      VirtViewerDisplayPrivate *priv;
> +    guint hint;
>      g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
>  
>      priv = self->priv;
> +    hint = priv->show_hint;
> +
> +    if (enabled)
> +        hint |= mask;
> +    else
> +        hint &= ~mask;
> +
>      if (priv->show_hint == hint)
>          return;
>  
> @@ -566,17 +574,9 @@ void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
>  
>  void virt_viewer_display_set_enabled(VirtViewerDisplay *self, gboolean enabled)
>  {
> -    guint hint;
> -
>      g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
>  
> -    hint = virt_viewer_display_get_show_hint(self);
> -    if (enabled)
> -        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
> -    else
> -        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
> -
> -    virt_viewer_display_set_show_hint(self, hint);
> +    virt_viewer_display_set_show_hint(self, VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED, !enabled);
>  }
>  
>  VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *self)
> diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
> index 4247570..d838c07 100644
> --- a/src/virt-viewer-display.h
> +++ b/src/virt-viewer-display.h
> @@ -110,7 +110,7 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
>  void virt_viewer_display_send_keys(VirtViewerDisplay *display,
>                                     const guint *keyvals, int nkeyvals);
>  GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
> -void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
> +void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint mask, gboolean enabled);
>  guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
>  VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
>  void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
> diff --git a/src/virt-viewer-session-vnc.c b/src/virt-viewer-session-vnc.c
> index 238c865..52baf5c 100644
> --- a/src/virt-viewer-session-vnc.c
> +++ b/src/virt-viewer-session-vnc.c
> @@ -95,7 +95,7 @@ virt_viewer_session_vnc_connected(VncDisplay *vnc G_GNUC_UNUSED,
>      GtkWidget *display = virt_viewer_display_vnc_new(session->priv->vnc);
>      g_signal_emit_by_name(session, "session-connected");
>      virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(display),
> -                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
> +                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, TRUE);
>      virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
>                                      VIRT_VIEWER_DISPLAY(display));
>  }
> @@ -104,11 +104,11 @@ static void
>  virt_viewer_session_vnc_disconnected(VncDisplay *vnc G_GNUC_UNUSED,
>                                       VirtViewerSessionVnc *session)
>  {
> +    GtkWidget *display = virt_viewer_display_vnc_new(session->priv->vnc);
>      DEBUG_LOG("Disconnected");
>      g_signal_emit_by_name(session, "session-disconnected");
> -    /* TODO perhaps? */
> -    /* virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(session->priv->vnc), */
> -    /*                                   VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE); */
> +    virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(display),
> +                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, FALSE);

This is a slight change in behaviour, maybe this patch should change the
call but keep it commented out, and then another patch would uncomment it?

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/e3c61adb/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 13:44:24 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 15:44:24 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 19/19] Show status page by
 default if display not ready
In-Reply-To: <1342457874-15369-20-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-20-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717134424.GL2162@teriyaki.cdg.redhat.com>

On Mon, Jul 16, 2012 at 06:57:54PM +0200, Marc-Andr? Lureau wrote:
> GtkNotebook will use the currently visible widget as default page.
> If we don't show status widget before we append the display, the
> current page will be on display. Quoting Gtk+ documentation:
> 
> "Note that due to historical reasons, GtkNotebook refuses to switch to
> a page unless the child widget is visible. Therefore, it is
> recommended to show child widgets before adding them to a notebook."

ACK for the first hunk, the second hunk seems to be unrelated to the change
described here.

Christophe

> ---
>  src/virt-viewer-notebook.c |    1 +
>  src/virt-viewer-window.c   |    6 +++++-
>  2 files changed, 6 insertions(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-notebook.c b/src/virt-viewer-notebook.c
> index 09e8ec4..cc7136b 100644
> --- a/src/virt-viewer-notebook.c
> +++ b/src/virt-viewer-notebook.c
> @@ -86,6 +86,7 @@ virt_viewer_notebook_init (VirtViewerNotebook *self)
>      priv->status = gtk_label_new("");
>      gtk_notebook_set_show_tabs(GTK_NOTEBOOK(self), FALSE);
>      gtk_notebook_set_show_border(GTK_NOTEBOOK(self), FALSE);
> +    gtk_widget_show_all(priv->status);
>      gtk_notebook_append_page(GTK_NOTEBOOK(self), priv->status, NULL);
>      gdk_color_parse("white", &color);
>      gtk_widget_modify_fg(priv->status, GTK_STATE_NORMAL, &color);
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 4bc427a..7c6e41f 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -1040,8 +1040,12 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
>          virt_viewer_display_set_zoom_level(VIRT_VIEWER_DISPLAY(priv->display), priv->zoomlevel);
>          virt_viewer_display_set_auto_resize(VIRT_VIEWER_DISPLAY(priv->display), priv->auto_resize);
>  
> -        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
>          gtk_widget_show_all(GTK_WIDGET(display));
> +        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
> +        /* switch back to non-display if not ready */
> +        if (!(virt_viewer_display_get_show_hint(display) &
> +              VIRT_VIEWER_DISPLAY_SHOW_HINT_READY))
> +            gtk_notebook_set_current_page(GTK_NOTEBOOK(priv->notebook), 0);
>  
>          virt_viewer_signal_connect_object(display, "display-pointer-grab",
>                                            G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/cf9d232e/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 13:53:10 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 15:53:10 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
Message-ID: <20120717135310.GM2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 03:02:46PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 2:52 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > I'm concerned about malicious payload putting a huge number there for a
> > nasty purpose. Moreover, is this value coming from the server, or is it
> > coming from the qxl driver in the guest?
> 
> 
> I don't think this is a concern here. You might worry about a lot of
> other parts of spice then.. In general, it can be very hard to verify
> integrity,

I'm indeed worried about the day when someone starts actively fuzzing the
spice protocol...


> and I guess we rely on lower level of the stack to do that for us.

Except I'm not sure any part of the stack is doing this for us, is there
such a part? In this specific case, the protocol can handle an arbitrary
number of monitors as I understand it, it's the client code that cannot
handle too many monitors, so limiting the number of monitors here would
make sense.
It's an issue I wanted to raise, I'm not saying this must be fixed in this
patch.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/2083b6cb/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 13:57:22 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 15:57:22 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/19] Number display
 starting from 1
In-Reply-To: <CAJ+F1CL2goPnYiMmhWxRDsvTfgmnjGp7_FDK_F4v4=FuBYWfKA@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-14-git-send-email-marcandre.lureau@gmail.com>
	<20120717094604.GY2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJmMVoBp_bmFBBunkumqxq=hE=NyDD8v8GJvCmH=7oSnQ@mail.gmail.com>
	<20120717122653.GC2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CL2goPnYiMmhWxRDsvTfgmnjGp7_FDK_F4v4=FuBYWfKA@mail.gmail.com>
Message-ID: <20120717135722.GN2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 03:08:55PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 2:26 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > This means that when the GUI talks about display 1 and 2, and the debug log
> > talks about window 1, you have to know it's really about display 2, imo
> > this is not developer friendly.
> 
> We have to disagree on that, my point is that a developper is more
> interested by true variable values rather than something
> user-friendly.

They are interested in predictible behaviour, having a debug message refer to
something subtly different than what they can see on screen is just
misleading, it's not helpful at all.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/1436714d/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 13:59:20 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 15:59:20 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717135310.GM2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>

On Tue, Jul 17, 2012 at 3:53 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
>> and I guess we rely on lower level of the stack to do that for us.
>
> Except I'm not sure any part of the stack is doing this for us, is there
> such a part? In this specific case, the protocol can handle an arbitrary

That would be tcp & ssl in our case.

> number of monitors as I understand it, it's the client code that cannot
> handle too many monitors, so limiting the number of monitors here would
> make sense.
> It's an issue I wanted to raise, I'm not saying this must be fixed in this
> patch.

I can't think of a reasonable way of handling that, except having
limitation of max monitors == number of monitors on client side. This
would be a logical change, except that it will not allow, for
instance, to work with multi-display on a single monitor (like I do
now, but perhaps this is only interesting for developpers)

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 14:00:39 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 16:00:39 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CJeOgx4e_oPSC53njg6+oAGR4P-g1sjBt-sdgP2qJ_DnA@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJeOgx4e_oPSC53njg6+oAGR4P-g1sjBt-sdgP2qJ_DnA@mail.gmail.com>
Message-ID: <20120717140039.GO2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 02:59:44PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 2:52 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > I just realized that g_array_unref is glib 2.22+
> 
> Glib 2.22 is in rhel6, so it's probably okay, although we need to bump
> configure.ac version check

This would remove the need for the #if GLIB_VERSION() in the rest of the
patch as well, so this would indeed be a good thing.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/06b3b8d4/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 14:06:45 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 16:06:45 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
Message-ID: <20120717140645.GP2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 03:59:20PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 3:53 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> >> and I guess we rely on lower level of the stack to do that for us.
> >
> > Except I'm not sure any part of the stack is doing this for us, is there
> > such a part? In this specific case, the protocol can handle an arbitrary
> 
> That would be tcp & ssl in our case.
> 
> > number of monitors as I understand it, it's the client code that cannot
> > handle too many monitors, so limiting the number of monitors here would
> > make sense.
> > It's an issue I wanted to raise, I'm not saying this must be fixed in this
> > patch.
> 
> I can't think of a reasonable way of handling that, except having
> limitation of max monitors == number of monitors on client side. This
> would be a logical change, except that it will not allow, for
> instance, to work with multi-display on a single monitor (like I do
> now, but perhaps this is only interesting for developpers)

Oh, I was mostly thinking of checking max_monitors for an arbitrary max
value (4, 16 or 256) to avoid allocating arbitrary amount of memory by
trusting a network value.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/597b7e4c/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 14:18:00 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 16:18:00 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717140645.GP2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>

On Tue, Jul 17, 2012 at 4:06 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> Oh, I was mostly thinking of checking max_monitors for an arbitrary max
> value (4, 16 or 256) to avoid allocating arbitrary amount of memory by
> trusting a network value.

There is no arbitrary value that would be really appropriate.

Furthermore, as I tried to say, there are many places where we trust
server values. For instance size of a message. It doesn't really makes
sense to check only this value here.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 14:22:15 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 16:22:15 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 18/19] Simplify display
	flag handling
In-Reply-To: <20120717134146.GK2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-19-git-send-email-marcandre.lureau@gmail.com>
	<20120717134146.GK2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CLsQTEpa+csNsN6=fymj0yJ=SerQWBpx_WdxzHa=7ex_A@mail.gmail.com>

On Tue, Jul 17, 2012 at 3:41 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> This is a slight change in behaviour, maybe this patch should change the
> call but keep it commented out, and then another patch would uncomment it?
>

Yeah, makes sense. done

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 14:30:08 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 16:30:08 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
Message-ID: <20120717143008.GS2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 04:18:00PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 4:06 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > Oh, I was mostly thinking of checking max_monitors for an arbitrary max
> > value (4, 16 or 256) to avoid allocating arbitrary amount of memory by
> > trusting a network value.
> 
> There is no arbitrary value that would be really appropriate.
> 
> Furthermore, as I tried to say, there are many places where we trust
> server values. For instance size of a message. It doesn't really makes
> sense to check only this value here.

Yeah I know there are many worrying places, for new code and new protocol
additions, it would be nice to start thinking about this...
I'm not seeing this as a blocking issue, but this is getting more and more
scary nonetheless...

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/246d4c74/attachment.sig>

From cfergeau at redhat.com  Tue Jul 17 14:37:00 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 16:37:00 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 14/19] Use
 SpiceDisplay:ready property instead of channel mark
In-Reply-To: <1342457874-15369-15-git-send-email-marcandre.lureau@gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-15-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120717143700.GT2162@teriyaki.cdg.redhat.com>

ACK
On Mon, Jul 16, 2012 at 06:57:49PM +0200, Marc-Andr? Lureau wrote:
> The display can now check several conditions before the display can be
> shown, use that instead of display mark, which was not high-level
> enough.
> ---
>  src/virt-viewer-display-spice.c |   38 +++++++++++++++-----------------------
>  1 file changed, 15 insertions(+), 23 deletions(-)
> 
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index 101abc5..53430dd 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -122,31 +122,25 @@ virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display)
>  }
>  
>  static void
> -display_mark(SpiceChannel *channel G_GNUC_UNUSED,
> -             gint mark,
> -             VirtViewerDisplay *display)
> +display_ready(GObject *display,
> +              GParamSpec *pspec G_GNUC_UNUSED,
> +              VirtViewerDisplay *self)
>  {
> -    DEBUG_LOG("Toggle monitor visibility %p %d", channel, mark);
> +    guint hint;
> +    gboolean ready;
>  
> -    virt_viewer_display_set_show_hint(display, mark);
> -}
> +    g_object_get(display, "ready", &ready, NULL);
> +    DEBUG_LOG("display %p ready:%d", self, ready);
>  
> -static void
> -primary_create(SpiceChannel *channel G_GNUC_UNUSED,
> -               gint format G_GNUC_UNUSED,
> -               gint width,
> -               gint height,
> -               gint stride G_GNUC_UNUSED,
> -               gint shmid G_GNUC_UNUSED,
> -               gpointer imgdata G_GNUC_UNUSED,
> -               VirtViewerDisplay *display)
> -{
> -    DEBUG_LOG("spice desktop resize %dx%d", width, height);
> +    hint = virt_viewer_display_get_show_hint(self);
> +    if (ready)
> +        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
> +    else
> +        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
>  
> -    virt_viewer_display_set_desktop_size(display, width, height);
> +    virt_viewer_display_set_show_hint(self, hint);
>  }
>  
> -
>  static void
>  virt_viewer_display_spice_keyboard_grab(SpiceDisplay *display G_GNUC_UNUSED,
>                                          int grabbed,
> @@ -239,10 +233,8 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>      self->priv->display = spice_display_new(s, channelid);
>      g_object_unref(s);
>  
> -    virt_viewer_signal_connect_object(channel, "display-primary-create",
> -                                      G_CALLBACK(primary_create), self, 0);
> -    virt_viewer_signal_connect_object(channel, "display-mark",
> -                                      G_CALLBACK(display_mark), self, 0);
> +    virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
> +                                      G_CALLBACK(display_ready), self, 0);
>  
>      gtk_container_add(GTK_CONTAINER(self), g_object_ref(self->priv->display));
>      gtk_widget_show(GTK_WIDGET(self->priv->display));
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/6cc0d53c/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 14:48:47 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 16:48:47 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717143008.GS2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-16-git-send-email-marcandre.lureau@gmail.com>
	<20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>

Hi

On Tue, Jul 17, 2012 at 4:30 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> Yeah I know there are many worrying places, for new code and new protocol
> additions, it would be nice to start thinking about this...
> I'm not seeing this as a blocking issue, but this is getting more and more
> scary nonetheless...

I don't think you have reasons to be worried here. What will a browser
do if it receives a message or say an image with a gigantic size? it
will probably keep reading an decoding it, until you run out of RAM,
no? That's the same for Spice. Checking server sizes doesn't make
sense. Checking out-of-bounds of memory / array lead by a decoding
logic (no matter how deep in the code) is what we should be careful
about (think about dictionnaries, or cache etc).

There is nothing wrong in OOM if the server tells us we have 1024
maximum monitors on the guest for example, although this is very
unlikely, and that's why the code allocated the widget ahead of time,
becausein general, you won't have > 4 anyway. If really we get into
this situation, we can allocate monitors widget on-demand. But I
really don't think it's necessary here. And it's easy enough for the
guest to enable all of them at once, so that wouldn't protect against
that for example.


-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 15:09:59 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 17:09:59 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
References: <20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
Message-ID: <20120717150959.GU2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 04:48:47PM +0200, Marc-Andr? Lureau wrote:
> Hi
> 
> On Tue, Jul 17, 2012 at 4:30 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > Yeah I know there are many worrying places, for new code and new protocol
> > additions, it would be nice to start thinking about this...
> > I'm not seeing this as a blocking issue, but this is getting more and more
> > scary nonetheless...
> 
> I don't think you have reasons to be worried here. What will a browser
> do if it receives a message or say an image with a gigantic size? it
> will probably keep reading an decoding it, until you run out of RAM,
> no? That's the same for Spice. Checking server sizes doesn't make
> sense. Checking out-of-bounds of memory / array lead by a decoding
> logic (no matter how deep in the code) is what we should be careful
> about (think about dictionnaries, or cache etc).

Ok, so monitors_max is a guint, the second argument to g_ptr_array_set_size
is a gint, and the GPtrArray code doesn't seem to handle. A quick reading
of GPtrArray code didn't make me feel confident that it would do the right
thing when size > G_INT_MAX / sizeof(void *), which could cause a buffer
overflow when very huge values are used (I haven't carefully checked that).


> There is nothing wrong in OOM if the server tells us we have 1024
> maximum monitors on the guest for example,

Is this max value coming from the server, or is it the guest qxl driver
telling the server about how many monitors it supports?

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/488bb342/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 15:24:02 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 17:24:02 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717150959.GU2162@teriyaki.cdg.redhat.com>
References: <20120717121506.GZ2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLKeS+yv4_HgH86rzKyg7KPOk5UtLDfjcTuVPNb=wXqyA@mail.gmail.com>
	<20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
	<20120717150959.GU2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>

Hi

On Tue, Jul 17, 2012 at 5:09 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> Ok, so monitors_max is a guint, the second argument to g_ptr_array_set_size
> is a gint, and the GPtrArray code doesn't seem to handle. A quick reading
> of GPtrArray code didn't make me feel confident that it would do the right
> thing when size > G_INT_MAX / sizeof(void *), which could cause a buffer
> overflow when very huge values are used (I haven't carefully checked that).

G_INT_MAX / sizeof(void *) ? where did you get the / sizeof(void*) from?

Eh ok, so you want to check if max_monitors > G_INT_MAX.. :) a very
unlikely condition that will crash the client anyway. But fine by me.

>> There is nothing wrong in OOM if the server tells us we have 1024
>> maximum monitors on the guest for example,
>
> Is this max value coming from the server, or is it the guest qxl driver
> telling the server about how many monitors it supports?

It's configured at xorg level, with NumHeads option, defaulting to 4.

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 15:41:10 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 17:41:10 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>
References: <20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
	<20120717150959.GU2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>
Message-ID: <20120717154109.GX2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 05:24:02PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 5:09 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > Ok, so monitors_max is a guint, the second argument to g_ptr_array_set_size
> > is a gint, and the GPtrArray code doesn't seem to handle. A quick reading
> > of GPtrArray code didn't make me feel confident that it would do the right
> > thing when size > G_INT_MAX / sizeof(void *), which could cause a buffer
> > overflow when very huge values are used (I haven't carefully checked that).
> 
> G_INT_MAX / sizeof(void *) ? where did you get the / sizeof(void*) from?

You are using a GPtrArray, so my understanding is that the memory allocated
for it is num_elements * sizeof (void*). If num_elements > G_INT_MAX /
sizeof(void*), a quick reading suggests this will cause an overflow in
the amount of memory to allocate causing the allocated array to be much
smaller than what is expected. Maybe that is not an issue, but this is
worrying. My point was that you are saying that allocating too much memory
will only cause swapping and the OOM to trigger, and I wanted to point out
that in this case, the problem could not be easily simplified to "too much
memory allocated, but no buffer overflow".

> 
> Eh ok, so you want to check if max_monitors > G_INT_MAX.. :) a very
> unlikely condition that will crash the client anyway. But fine by me.

Not really, just trying to understand what the extent of the problem is,
and quite worried that you are basically saying this is not an issue at all
and that all is fine.

> 
> >> There is nothing wrong in OOM if the server tells us we have 1024
> >> maximum monitors on the guest for example,
> >
> > Is this max value coming from the server, or is it the guest qxl driver
> > telling the server about how many monitors it supports?
> 
> It's configured at xorg level, with NumHeads option, defaulting to 4.

So what we are trusting here is an arbitrary value provided by the guest
system?

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/ce9379df/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 15:56:46 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 17:56:46 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717154109.GX2162@teriyaki.cdg.redhat.com>
References: <20120717125231.GG2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKqHdi7eXjKUSyNBwCt-H0_Sp74tSF9MEovs88SEWnKAg@mail.gmail.com>
	<20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
	<20120717150959.GU2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>
	<20120717154109.GX2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1C+KGA+yU79NBcuDGhNW9L1WcZGPyB4VtEs8Jf7z_TD04A@mail.gmail.com>

Hi

On Tue, Jul 17, 2012 at 5:41 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
>> G_INT_MAX / sizeof(void *) ? where did you get the / sizeof(void*) from?
>
> You are using a GPtrArray, so my understanding is that the memory allocated
> for it is num_elements * sizeof (void*). If num_elements > G_INT_MAX /
> sizeof(void*), a quick reading suggests this will cause an overflow in
> the amount of memory to allocate causing the allocated array to be much
> smaller than what is expected. Maybe that is not an issue, but this is

It looks like GPtrArry uses guint type, but set_size() is gint.. If
there is an overflow in array length computation, it should be handled
by glib perhaps.

> worrying. My point was that you are saying that allocating too much memory
> will only cause swapping and the OOM to trigger, and I wanted to point out
> that in this case, the problem could not be easily simplified to "too much
> memory allocated, but no buffer overflow".

this is not yet a buffer overflow proof, but you are getting closer perhaps.

>>
>> Eh ok, so you want to check if max_monitors > G_INT_MAX.. :) a very
>> unlikely condition that will crash the client anyway. But fine by me.
>
> Not really, just trying to understand what the extent of the problem is,
> and quite worried that you are basically saying this is not an issue at all
> and that all is fine.

Until we can really point clearly the issue.


>> >> There is nothing wrong in OOM if the server tells us we have 1024
>> >> maximum monitors on the guest for example,
>> >
>> > Is this max value coming from the server, or is it the guest qxl driver
>> > telling the server about how many monitors it supports?
>>
>> It's configured at xorg level, with NumHeads option, defaulting to 4.
>
> So what we are trusting here is an arbitrary value provided by the guest
> system?

Yes, no further checks after that afaict. So a misconfigured guest
could trigger this error perhaps.


-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 16:04:10 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 18:04:10 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1C+KGA+yU79NBcuDGhNW9L1WcZGPyB4VtEs8Jf7z_TD04A@mail.gmail.com>
References: <20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
	<20120717150959.GU2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>
	<20120717154109.GX2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C+KGA+yU79NBcuDGhNW9L1WcZGPyB4VtEs8Jf7z_TD04A@mail.gmail.com>
Message-ID: <20120717160410.GY2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 05:56:46PM +0200, Marc-Andr? Lureau wrote:
> this is not yet a buffer overflow proof, but you are getting closer perhaps.

For this kind of stuff, I prefer to err on the "there's a buffer overflow
unless proven otherwise", it's too dangerous to make the opposite
assumption.

> > So what we are trusting here is an arbitrary value provided by the guest
> > system?
> 
> Yes, no further checks after that afaict. So a misconfigured guest
> could trigger this error perhaps.

I'm more concerned about malicious guests than misconfigured ones.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/28a19839/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 16:06:56 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 18:06:56 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
	Monitors
In-Reply-To: <20120717160410.GY2162@teriyaki.cdg.redhat.com>
References: <20120717135310.GM2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CLYMZUJuHVaH=QkbWL=PuUNA09xVBC+HpsLj-8dkiQR0g@mail.gmail.com>
	<20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
	<20120717150959.GU2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>
	<20120717154109.GX2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C+KGA+yU79NBcuDGhNW9L1WcZGPyB4VtEs8Jf7z_TD04A@mail.gmail.com>
	<20120717160410.GY2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1C++Z=cVJh1vx64+tKtReB2Ww+JroNsxAkU7yMY-0vffoQ@mail.gmail.com>

On Tue, Jul 17, 2012 at 6:04 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> I'm more concerned about malicious guests than misconfigured ones.

Ok, what if we limit this value to 256 and warn if that condition is reached?

Should that be done at spice-gtk level?

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Tue Jul 17 16:19:34 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 17 Jul 2012 18:19:34 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 15/19] Hook up handling of
 Monitors
In-Reply-To: <CAJ+F1C++Z=cVJh1vx64+tKtReB2Ww+JroNsxAkU7yMY-0vffoQ@mail.gmail.com>
References: <20120717140645.GP2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CKwJFtty3Jg6MwULKkeDuwVNYg3dZWHrdv3qPGv3XNo4A@mail.gmail.com>
	<20120717143008.GS2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJUM_uA2wr_JC1gwZ9SZgEa1iKZw-8RJXtXgjLcz=qvow@mail.gmail.com>
	<20120717150959.GU2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJ7hTDa3J4XUjJhQZLJ4QxjoY=DB=b41PO=SY3A4So7hg@mail.gmail.com>
	<20120717154109.GX2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C+KGA+yU79NBcuDGhNW9L1WcZGPyB4VtEs8Jf7z_TD04A@mail.gmail.com>
	<20120717160410.GY2162@teriyaki.cdg.redhat.com>
	<CAJ+F1C++Z=cVJh1vx64+tKtReB2Ww+JroNsxAkU7yMY-0vffoQ@mail.gmail.com>
Message-ID: <20120717161934.GZ2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 06:06:56PM +0200, Marc-Andr? Lureau wrote:
> On Tue, Jul 17, 2012 at 6:04 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> > I'm more concerned about malicious guests than misconfigured ones.
> 
> Ok, what if we limit this value to 256 and warn if that condition is reached?
> 
> Should that be done at spice-gtk level?

That's your call. Doing it in spice-gtk sounds better, but maybe you have
use cases in mind where it does not make sense to do it there, and when
it's better to do it in client applications.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/71b147c2/attachment.sig>

From marcandre.lureau at gmail.com  Tue Jul 17 17:38:06 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 19:38:06 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 16/19] Wait until idle
 handler has run to mark dirty = FALSE
In-Reply-To: <20120717130126.GI2162@teriyaki.cdg.redhat.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-17-git-send-email-marcandre.lureau@gmail.com>
	<20120717130126.GI2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJV1+-f3wREKw=+xp0MNwPjj-KOgiqHFxpgYVY1sXT8+w@mail.gmail.com>

Hi

On Tue, Jul 17, 2012 at 3:01 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> On Mon, Jul 16, 2012 at 06:57:51PM +0200, Marc-Andr? Lureau wrote:
>> This can avoid having multiple idler added

ok that patch description is incorrect, and it probably doesn't work
as intended, however, without it, the resize/constrain behaviour is
buggy with this branch. I need to dig further. I would appreciate help
from Daniel :)

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 18:39:43 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Tue, 17 Jul 2012 20:39:43 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 16/19] Wait until idle
 handler has run to mark dirty = FALSE
In-Reply-To: <CAJ+F1CJV1+-f3wREKw=+xp0MNwPjj-KOgiqHFxpgYVY1sXT8+w@mail.gmail.com>
References: <1342457874-15369-1-git-send-email-marcandre.lureau@gmail.com>
	<1342457874-15369-17-git-send-email-marcandre.lureau@gmail.com>
	<20120717130126.GI2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CJV1+-f3wREKw=+xp0MNwPjj-KOgiqHFxpgYVY1sXT8+w@mail.gmail.com>
Message-ID: <CAJ+F1CJ987ba+U7v56RrUUGpOXhH+KZTQbwzTWegXrkw5o0zKw@mail.gmail.com>

Hi

Ok, forget this one, I can't seem to reproduce the resize problem it
used to work around.

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:43 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:43 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 01/25] build: allow building
	with newer glibc-headers and -O0
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-2-git-send-email-marcandre.lureau@gmail.com>

    Fix copied from libvirt, commit by Eric Blake.

    glibc 2.15 (on Fedora 17) coupled with explicit disabling of
    optimization during development dies a painful death:

    /usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires compiling with optimization (-O) [-Werror=cpp]

    Work around this by only conditionally defining _FORTIFY_SOURCE,
    in the case where glibc can actually use it.  The trick is using
    AH_VERBATIM instead of AC_DEFINE.
---
 acinclude.m4 |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/acinclude.m4 b/acinclude.m4
index 73b5a7b..0c308bc 100644
--- a/acinclude.m4
+++ b/acinclude.m4
@@ -19,7 +19,7 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
 
     warnCFLAGS=
 
-    try_compiler_flags="-Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
+    try_compiler_flags="-fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
 
     case "$enable_compile_warnings" in
     no)
@@ -44,6 +44,13 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
 	;;
     esac
 
+    AH_VERBATIM([FORTIFY_SOURCE],
+                [/* Enable compile-time and run-time bounds-checking, and some warnings. */
+                 #if defined __OPTIMIZE__ && __OPTIMIZE__
+                 # define _FORTIFY_SOURCE 2
+                 #endif
+                ])
+
     compiler_flags=
     for option in $try_compiler_flags; do
 	SAVE_CFLAGS="$CFLAGS"
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:44 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:44 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/25] spice: improve
	fullscreen=auto-conf
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>

Do keep client monitor position, do not align monitors automatically.
The align property is only available since v0.12.101.
---
 configure.ac                    |    2 +-
 src/virt-viewer-session-spice.c |    4 ++++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/configure.ac b/configure.ac
index 494cd13..a2471f4 100644
--- a/configure.ac
+++ b/configure.ac
@@ -19,7 +19,7 @@ GTK2_REQUIRED="2.18.0"
 GTK3_REQUIRED="3.0"
 GTK_VNC1_REQUIRED="0.3.8"
 GTK_VNC2_REQUIRED="0.4.0"
-SPICE_GTK_REQUIRED="0.11"
+SPICE_GTK_REQUIRED="0.12.101"
 SPICE_PROTOCOL_REQUIRED="0.10.1"
 
 AC_MSG_CHECKING([for native Win32])
diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
index 7b47735..6577237 100644
--- a/src/virt-viewer-session-spice.c
+++ b/src/virt-viewer-session-spice.c
@@ -499,6 +499,10 @@ virt_viewer_session_spice_fullscreen_auto_conf(VirtViewerSessionSpice *self)
 
     DEBUG_LOG("Performing full screen auto-conf, %d host monitors",
               gdk_screen_get_n_monitors(screen));
+    g_object_set(G_OBJECT(cmain),
+                 "disable-display-position", FALSE,
+                 "disable-display-align", TRUE,
+                 NULL);
     spice_main_set_display_enabled(cmain, -1, FALSE);
     for (i = 0; i < gdk_screen_get_n_monitors(screen); i++) {
         gdk_screen_get_monitor_geometry(screen, i, &dest);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:47 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:47 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 05/25] Destroy dialog
	immediately after run
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-6-git-send-email-marcandre.lureau@gmail.com>

If the parent is already destroyed, it looks like the dialog is
destroyed too. This avoids a crash when calling app_quit().
---
 src/virt-viewer-app.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 40d1ff7..1447ce7 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -283,6 +283,7 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
                                         GTK_BUTTONS_OK_CANCEL,
                                         _("This is the last visible display. Do you want to quit?"));
             gint result = gtk_dialog_run (GTK_DIALOG (dialog));
+            gtk_widget_destroy(dialog);
             switch (result) {
             case GTK_RESPONSE_OK:
                 virt_viewer_app_quit(self);
@@ -290,7 +291,6 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
             default:
                 break;
             }
-            gtk_widget_destroy(dialog);
             return FALSE;
         } else {
             virt_viewer_app_quit(self);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:48 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:48 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 06/25] Turn display:show-hint
	into flags type
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-7-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c     |    5 +++--
 src/virt-viewer-display.c |   26 ++++++++++++++++----------
 src/virt-viewer-display.h |   11 ++++++-----
 src/virt-viewer-window.c  |    5 ++++-
 4 files changed, 29 insertions(+), 18 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 1447ce7..5588c6a 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -609,7 +609,8 @@ display_show_hint(VirtViewerDisplay *display,
 {
     VirtViewerApp *self;
     VirtViewerNotebook *nb = virt_viewer_window_get_notebook(win);
-    gint nth, hint;
+    gint nth;
+    guint hint;
 
     g_object_get(win,
                  "app", &self,
@@ -619,7 +620,7 @@ display_show_hint(VirtViewerDisplay *display,
                  "show-hint", &hint,
                  NULL);
 
-    if (hint == VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE) {
+    if (hint == 0) {
         if (win != self->priv->main_window &&
             g_getenv("VIRT_VIEWER_HIDE"))
             virt_viewer_window_hide(win);
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 6d6052b..43ec45e 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -41,7 +41,7 @@ struct _VirtViewerDisplayPrivate
     guint zoom_level;
     gboolean zoom;
     gint nth_display;
-    gint show_hint;
+    guint show_hint;
     VirtViewerSession *session;
     gboolean auto_resize;
 };
@@ -151,13 +151,12 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
 
     g_object_class_install_property(object_class,
                                     PROP_SHOW_HINT,
-                                    g_param_spec_int("show-hint",
-                                                     "Show hint",
-                                                     "Show state hint",
-                                                     0,
-                                                     G_MAXINT32,
-                                                     0,
-                                                     G_PARAM_READABLE));
+                                    g_param_spec_flags("show-hint",
+                                                       "Show hint",
+                                                       "Show state hint",
+                                                       VIRT_VIEWER_TYPE_DISPLAY_SHOW_HINT_FLAGS,
+                                                       0,
+                                                       G_PARAM_READABLE));
 
     g_object_class_install_property(object_class,
                                     PROP_SESSION,
@@ -298,7 +297,7 @@ virt_viewer_display_get_property(GObject *object,
         g_value_set_int(value, priv->nth_display);
         break;
     case PROP_SHOW_HINT:
-        g_value_set_int(value, priv->show_hint);
+        g_value_set_flags(value, priv->show_hint);
         break;
     case PROP_SESSION:
         g_value_set_object(value, virt_viewer_display_get_session(display));
@@ -530,7 +529,14 @@ GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display)
     return VIRT_VIEWER_DISPLAY_GET_CLASS(display)->get_pixbuf(display);
 }
 
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, gint hint)
+guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
+{
+    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), 0);
+
+    return self->priv->show_hint;
+}
+
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
 {
     VirtViewerDisplayPrivate *priv;
     g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index ffbaf0e..225080d 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -25,6 +25,7 @@
 #define _VIRT_VIEWER_DISPLAY_H
 
 #include <gtk/gtk.h>
+#include "virt-viewer-enums.h"
 
 G_BEGIN_DECLS
 
@@ -54,10 +55,9 @@ typedef struct _VirtViewerDisplayPrivate VirtViewerDisplayPrivate;
 
 typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
 
-enum {
-    VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE = 0,
-    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY,
-};
+typedef enum {
+    VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
+} VirtViewerDisplayShowHintFlags;
 
 /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
 struct _VirtViewerDisplay {
@@ -108,7 +108,8 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
 void virt_viewer_display_send_keys(VirtViewerDisplay *display,
                                    const guint *keyvals, int nkeyvals);
 GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, gint hint);
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
+guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
 void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
 gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 7b582a0..c873ebc 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1002,9 +1002,12 @@ display_show_hint(VirtViewerDisplay *display,
                   GParamSpec *pspec G_GNUC_UNUSED,
                   VirtViewerWindow *self)
 {
-    gboolean hint;
+    guint hint;
 
     g_object_get(display, "show-hint", &hint, NULL);
+
+    hint = (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
+
     gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-send")), hint);
     gtk_widget_set_sensitive(GTK_WIDGET(gtk_builder_get_object(self->priv->builder, "menu-file-screenshot")), hint);
     gtk_widget_set_sensitive(self->priv->toolbar_send_key, hint);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:49 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:49 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 07/25] Run-time check values
	before doing bad computation
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-8-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-display.c |    6 +++++-
 src/virt-viewer-window.c  |    5 +++++
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 43ec45e..a954671 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -403,6 +403,10 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
     DEBUG_LOG("Allocated %dx%d", allocation->width, allocation->height);
     gtk_widget_set_allocation(widget, allocation);
 
+    if (priv->desktopWidth == 0 ||
+        priv->desktopHeight == 0)
+        goto end;
+
     desktopAspect = (double)priv->desktopWidth / (double)priv->desktopHeight;
 
     if (child && gtk_widget_get_visible(child)) {
@@ -427,7 +431,7 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
         gtk_widget_size_allocate(child, &child_allocation);
     }
 
-
+end:
     /* This unsets the size request, so that the user can
      * manually resize the window smaller again
      */
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index c873ebc..4887e28 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -433,6 +433,11 @@ virt_viewer_window_resize(VirtViewerWindow *self)
                                     (screen, gtk_widget_get_window(priv->window)),
                                     &fullscreen);
 
+    g_return_if_fail(fullscreen.height > 128);
+    g_return_if_fail(fullscreen.width > 128);
+    g_return_if_fail(desktopWidth > 0);
+    g_return_if_fail(desktopHeight > 0);
+
     desktopAspect = (double)desktopWidth / (double)desktopHeight;
     screenAspect = (double)(fullscreen.width - 128) / (double)(fullscreen.height - 128);
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:50 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:50 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 08/25] Mark a menu string as
	translatable
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-9-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 5588c6a..661cdd7 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -1634,7 +1634,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
         gboolean visible;
         gchar *label;
 
-        label = g_strdup_printf("Display %d", *nth);
+        label = g_strdup_printf(_("Display %d"), *nth);
         item = gtk_check_menu_item_new_with_label(label);
         g_free(label);
         visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:51 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:51 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 09/25] Rebuild display menu
	when a window is added or removed
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-10-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 661cdd7..c4f712d 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -520,6 +520,7 @@ virt_viewer_app_remove_nth_window(VirtViewerApp *self, gint nth)
     DEBUG_LOG("Remove window %d %p", nth, win);
     removed = g_hash_table_steal(self->priv->windows, &nth);
     g_warn_if_fail(removed);
+    virt_viewer_app_update_menu_displays(self);
 
     if (removed)
         g_signal_emit(self, signals[SIGNAL_WINDOW_REMOVED], 0, win);
@@ -540,6 +541,7 @@ virt_viewer_app_set_nth_window(VirtViewerApp *self, gint nth, VirtViewerWindow *
     DEBUG_LOG("Insert window %d %p", nth, win);
     g_hash_table_insert(self->priv->windows, key, win);
     virt_viewer_app_set_window_subtitle(self, win, nth);
+    virt_viewer_app_update_menu_displays(self);
 
     g_signal_emit(self, signals[SIGNAL_WINDOW_ADDED], 0, win);
 }
@@ -657,6 +659,7 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
     }
 
     virt_viewer_window_set_display(window, display);
+    virt_viewer_app_update_menu_displays(self);
     g_signal_connect(display, "notify::show-hint",
                      G_CALLBACK(display_show_hint), window);
     g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:52 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:52 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 10/25] Use
	virt_viewer_connect_object() for display
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-11-git-send-email-marcandre.lureau@gmail.com>

When display is released, detach signal automatically.

Fix various crash related to not cleaning up signal handlers properly,
due to no longer 1-1 only relation between display widget and channel.
---
 src/virt-viewer-app.c    |    4 ++--
 src/virt-viewer-window.c |   24 ++++++++++++------------
 2 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index c4f712d..3057116 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -660,8 +660,8 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
 
     virt_viewer_window_set_display(window, display);
     virt_viewer_app_update_menu_displays(self);
-    g_signal_connect(display, "notify::show-hint",
-                     G_CALLBACK(display_show_hint), window);
+    virt_viewer_signal_connect_object(display, "notify::show-hint",
+                                      G_CALLBACK(display_show_hint), window, 0);
     g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */
 }
 
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 4887e28..3f7cbbe 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1042,18 +1042,18 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
         gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
         gtk_widget_show_all(GTK_WIDGET(display));
 
-        g_signal_connect(display, "display-pointer-grab",
-                         G_CALLBACK(virt_viewer_window_pointer_grab), self);
-        g_signal_connect(display, "display-pointer-ungrab",
-                         G_CALLBACK(virt_viewer_window_pointer_ungrab), self);
-        g_signal_connect(display, "display-keyboard-grab",
-                         G_CALLBACK(virt_viewer_window_keyboard_grab), self);
-        g_signal_connect(display, "display-keyboard-ungrab",
-                         G_CALLBACK(virt_viewer_window_keyboard_ungrab), self);
-        g_signal_connect(display, "display-desktop-resize",
-                         G_CALLBACK(virt_viewer_window_desktop_resize), self);
-        g_signal_connect(display, "notify::show-hint",
-                         G_CALLBACK(display_show_hint), self);
+        virt_viewer_signal_connect_object(display, "display-pointer-grab",
+                                          G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-pointer-ungrab",
+                                          G_CALLBACK(virt_viewer_window_pointer_ungrab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-keyboard-grab",
+                                          G_CALLBACK(virt_viewer_window_keyboard_grab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-keyboard-ungrab",
+                                          G_CALLBACK(virt_viewer_window_keyboard_ungrab), self, 0);
+        virt_viewer_signal_connect_object(display, "display-desktop-resize",
+                                          G_CALLBACK(virt_viewer_window_desktop_resize), self, 0);
+        virt_viewer_signal_connect_object(display, "notify::show-hint",
+                                          G_CALLBACK(display_show_hint), self, 0);
     }
 }
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:53 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:53 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 11/25] Add a DISABLED display
	hint
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-12-git-send-email-marcandre.lureau@gmail.com>

This flag will help to track whether the display has been
removed/closed and whether it really has a valid display.

Ready in contrast, is used to "hide" temporarily the display (when
starting or redrawing the display, to avoid artifacts)
---
 src/virt-viewer-app.c           |   12 +++++++-----
 src/virt-viewer-display-spice.c |   20 ++++++++++++++++++++
 src/virt-viewer-display.c       |   15 +++++++++++++++
 src/virt-viewer-display.h       |    2 ++
 src/virt-viewer-window.c        |   12 ++++++++++++
 5 files changed, 56 insertions(+), 5 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 3057116..727eb81 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -622,15 +622,17 @@ display_show_hint(VirtViewerDisplay *display,
                  "show-hint", &hint,
                  NULL);
 
-    if (hint == 0) {
+    if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED) {
+        virt_viewer_window_hide(win);
+    } else if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY) {
+        virt_viewer_notebook_show_display(nb);
+        virt_viewer_window_show(win);
+        gtk_window_present(virt_viewer_window_get_window(win));
+    } else {
         if (win != self->priv->main_window &&
             g_getenv("VIRT_VIEWER_HIDE"))
             virt_viewer_window_hide(win);
         virt_viewer_notebook_show_status(nb, _("Waiting for display %d..."), nth + 1);
-    } else {
-        virt_viewer_notebook_show_display(nb);
-        virt_viewer_window_show(win);
-        gtk_window_present(virt_viewer_window_get_window(win));
     }
 
     g_object_unref(self);
diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 7bf9a9a..101abc5 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -75,9 +75,26 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
 }
 
 static void
+show_hint_changed(VirtViewerDisplay *self)
+{
+    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
+        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self)));
+    guint enabled = TRUE;
+    guint nth;
+
+    g_object_get(self, "nth-display", &nth, NULL);
+    if (virt_viewer_display_get_show_hint(self) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
+        enabled = FALSE;
+
+    spice_main_set_display_enabled(main_channel, nth, enabled);
+}
+
+static void
 virt_viewer_display_spice_init(VirtViewerDisplaySpice *self G_GNUC_UNUSED)
 {
     self->priv = VIRT_VIEWER_DISPLAY_SPICE_GET_PRIVATE(self);
+
+    g_signal_connect(self, "notify::show-hint", G_CALLBACK(show_hint_changed), NULL);
 }
 
 static void
@@ -166,6 +183,9 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
     if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
         return;
 
+    if (virt_viewer_display_get_show_hint(VIRT_VIEWER_DISPLAY(self)) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
+        return;
+
     if (virt_viewer_display_get_zoom(VIRT_VIEWER_DISPLAY(self))) {
         zoom = virt_viewer_display_get_zoom_level(VIRT_VIEWER_DISPLAY(self));
 
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index a954671..18fed19 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -553,6 +553,21 @@ void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
     g_object_notify(G_OBJECT(self), "show-hint");
 }
 
+void virt_viewer_display_set_enabled(VirtViewerDisplay *self, gboolean enabled)
+{
+    guint hint;
+
+    g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
+
+    hint = virt_viewer_display_get_show_hint(self);
+    if (enabled)
+        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
+    else
+        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
+
+    virt_viewer_display_set_show_hint(self, hint);
+}
+
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *self)
 {
     g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), NULL);
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index 225080d..89b4817 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -57,6 +57,7 @@ typedef struct _VirtViewerDisplayChannel VirtViewerDisplayChannel;
 
 typedef enum {
     VIRT_VIEWER_DISPLAY_SHOW_HINT_READY            = 1 << 0,
+    VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED         = 1 << 1,
 } VirtViewerDisplayShowHintFlags;
 
 /* perhaps this become an interface, and be pushed in gtkvnc and spice? */
@@ -116,6 +117,7 @@ gboolean virt_viewer_display_get_auto_resize(VirtViewerDisplay *display);
 void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
 
 void virt_viewer_display_close(VirtViewerDisplay *display);
+void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
 
 G_END_DECLS
 
diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 3f7cbbe..46e44be 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1062,6 +1062,9 @@ virt_viewer_window_show(VirtViewerWindow *self)
 {
     gtk_widget_show(self->priv->window);
 
+    if (self->priv->display)
+        virt_viewer_display_set_enabled(self->priv->display, TRUE);
+
     if (self->priv->desktop_resize_pending) {
         virt_viewer_window_resize(self);
         self->priv->desktop_resize_pending = FALSE;
@@ -1072,6 +1075,15 @@ void
 virt_viewer_window_hide(VirtViewerWindow *self)
 {
     gtk_widget_hide(self->priv->window);
+
+    if (self->priv->display) {
+        VirtViewerDisplay *display = self->priv->display;
+        guint nth;
+
+        g_object_get(display, "nth-display", &nth, NULL);
+        if (nth != 0)
+            virt_viewer_display_set_enabled(display, FALSE);
+    }
 }
 
 void
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:54 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:54 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 12/25] Prevent from adding the
	same display several time in the session
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-13-git-send-email-marcandre.lureau@gmail.com>

In virt_viewer_session_spice_display_monitors(), we (re)add the
display unconditionnaly every time we receive a new MonitorConfig.
---
 src/virt-viewer-session.c |    3 +++
 1 file changed, 3 insertions(+)

diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
index 18b6922..9249a1f 100644
--- a/src/virt-viewer-session.c
+++ b/src/virt-viewer-session.c
@@ -282,6 +282,9 @@ virt_viewer_session_new(void)
 void virt_viewer_session_add_display(VirtViewerSession *session,
                                      VirtViewerDisplay *display)
 {
+    if (g_list_find(session->priv->displays, display))
+        return;
+
     session->priv->displays = g_list_append(session->priv->displays, display);
     g_object_ref(display);
     g_signal_emit_by_name(session, "session-display-added", display);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:55 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:55 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 13/25] Number display starting
	from 1
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-14-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 727eb81..3d7ff68 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -1639,7 +1639,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
         gboolean visible;
         gchar *label;
 
-        label = g_strdup_printf(_("Display %d"), *nth);
+        label = g_strdup_printf(_("Display %d"), *nth + 1);
         item = gtk_check_menu_item_new_with_label(label);
         g_free(label);
         visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:56 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:56 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 14/25] Use SpiceDisplay:ready
	property instead of channel mark
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-15-git-send-email-marcandre.lureau@gmail.com>

The display can now check several conditions before the display can be
shown, use that instead of display mark, which was not high-level
enough.
---
 src/virt-viewer-display-spice.c |   38 +++++++++++++++-----------------------
 1 file changed, 15 insertions(+), 23 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 101abc5..53430dd 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -122,31 +122,25 @@ virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display)
 }
 
 static void
-display_mark(SpiceChannel *channel G_GNUC_UNUSED,
-             gint mark,
-             VirtViewerDisplay *display)
+display_ready(GObject *display,
+              GParamSpec *pspec G_GNUC_UNUSED,
+              VirtViewerDisplay *self)
 {
-    DEBUG_LOG("Toggle monitor visibility %p %d", channel, mark);
+    guint hint;
+    gboolean ready;
 
-    virt_viewer_display_set_show_hint(display, mark);
-}
+    g_object_get(display, "ready", &ready, NULL);
+    DEBUG_LOG("display %p ready:%d", self, ready);
 
-static void
-primary_create(SpiceChannel *channel G_GNUC_UNUSED,
-               gint format G_GNUC_UNUSED,
-               gint width,
-               gint height,
-               gint stride G_GNUC_UNUSED,
-               gint shmid G_GNUC_UNUSED,
-               gpointer imgdata G_GNUC_UNUSED,
-               VirtViewerDisplay *display)
-{
-    DEBUG_LOG("spice desktop resize %dx%d", width, height);
+    hint = virt_viewer_display_get_show_hint(self);
+    if (ready)
+        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
+    else
+        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
 
-    virt_viewer_display_set_desktop_size(display, width, height);
+    virt_viewer_display_set_show_hint(self, hint);
 }
 
-
 static void
 virt_viewer_display_spice_keyboard_grab(SpiceDisplay *display G_GNUC_UNUSED,
                                         int grabbed,
@@ -239,10 +233,8 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
     self->priv->display = spice_display_new(s, channelid);
     g_object_unref(s);
 
-    virt_viewer_signal_connect_object(channel, "display-primary-create",
-                                      G_CALLBACK(primary_create), self, 0);
-    virt_viewer_signal_connect_object(channel, "display-mark",
-                                      G_CALLBACK(display_mark), self, 0);
+    virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
+                                      G_CALLBACK(display_ready), self, 0);
 
     gtk_container_add(GTK_CONTAINER(self), g_object_ref(self->priv->display));
     gtk_widget_show(GTK_WIDGET(self->priv->display));
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:57 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:57 +0200
Subject: [virt-tools-list] =?utf-8?q?=5BPATCH_virt-viewer_15/25=5D_Bump_gl?=
	=?utf-8?q?ib_=3E_2=2E22=2C_add_compat_file?=
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-16-git-send-email-marcandre.lureau@gmail.com>

We use API from 2.22, and some from further version. Add
virt-glib-compat.h fallback file for those.
---
 configure.ac           |    2 +-
 src/Makefile.am        |    1 +
 src/virt-glib-compat.h |   56 ++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 58 insertions(+), 1 deletion(-)
 create mode 100644 src/virt-glib-compat.h

diff --git a/configure.ac b/configure.ac
index 548b945..ca3956e 100644
--- a/configure.ac
+++ b/configure.ac
@@ -12,7 +12,7 @@ AC_CANONICAL_HOST
 m4_ifndef([AM_SILENT_RULES], [m4_define([AM_SILENT_RULES],[])])
 AM_SILENT_RULES([yes])
 
-GLIB2_REQUIRED=2.12.0
+GLIB2_REQUIRED=2.22.0
 LIBXML2_REQUIRED="2.6.0"
 LIBVIRT_REQUIRED="0.9.7"
 GTK2_REQUIRED="2.18.0"
diff --git a/src/Makefile.am b/src/Makefile.am
index 5daa98a..8f9339c 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -32,6 +32,7 @@ $(BUILT_SOURCES): %: %.etemplate $(ENUMS_FILES)
 
 COMMON_SOURCES =					\
 	$(BUILT_SOURCES)				\
+	virt-glib-compat.h				\
 	virt-gtk-compat.h				\
 	virt-viewer-util.h virt-viewer-util.c		\
 	virt-viewer-auth.h virt-viewer-auth.c		\
diff --git a/src/virt-glib-compat.h b/src/virt-glib-compat.h
new file mode 100644
index 0000000..23345a0
--- /dev/null
+++ b/src/virt-glib-compat.h
@@ -0,0 +1,56 @@
+/*
+ * Virt Viewer: A virtual machine console viewer
+ *
+ * Copyright (C) 2007-2009 Red Hat, Inc.
+ * Copyright (C) 2009-2012 Daniel P. Berrange
+ * Copyright (C) 2010 Marc-Andr? Lureau
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Author: Daniel P. Berrange <berrange at redhat.com>
+ */
+
+#include <config.h>
+
+#ifndef _VIRT_GLIB_COMPAT_H
+# define _VIRT_GLIB_COMPAT_H 1
+
+#include <glib-object.h>
+
+G_BEGIN_DECLS
+
+#ifndef g_clear_pointer
+#define g_clear_pointer(pp, destroy) \
+  G_STMT_START {                                                               \
+    G_STATIC_ASSERT (sizeof *(pp) == sizeof (gpointer));                       \
+    /* Only one access, please */                                              \
+    gpointer *_pp = (gpointer *) (pp);                                         \
+    gpointer _p;                                                               \
+    /* This assignment is needed to avoid a gcc warning */                     \
+    GDestroyNotify _destroy = (GDestroyNotify) (destroy);                      \
+                                                                               \
+    (void) (0 ? (gpointer) *(pp) : 0);                                         \
+    do                                                                         \
+      _p = g_atomic_pointer_get (_pp);                                         \
+    while G_UNLIKELY (!g_atomic_pointer_compare_and_exchange (_pp, _p, NULL)); \
+                                                                               \
+    if (_p)                                                                    \
+      _destroy (_p);                                                           \
+  } G_STMT_END
+#endif
+
+G_END_DECLS
+
+#endif // _VIRT_GLIB_COMPAT_H
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:58 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:58 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 16/25] Hook up handling of
	Monitors
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-17-git-send-email-marcandre.lureau@gmail.com>

Rely on spice-gtk display channel monitors property to manage
displays. The same display channel may now provide several monitors,
the SpiceDisplay widget must be told which monitor to display
---
 src/virt-viewer-display-spice.c |   20 +++++----
 src/virt-viewer-display-spice.h |    2 +-
 src/virt-viewer-session-spice.c |   89 ++++++++++++++++++++++++++++++++++-----
 3 files changed, 92 insertions(+), 19 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 53430dd..d06e5cf 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -172,7 +172,7 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
 {
     gdouble dw = allocation->width, dh = allocation->height;
     guint zoom = 100;
-    guint channelid;
+    guint nth;
 
     if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
         return;
@@ -187,13 +187,11 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
         dh /= ((double)zoom / 100.0);
     }
 
-    g_object_get(self->priv->channel, "channel-id", &channelid, NULL);
+    g_object_get(self, "nth-display", &nth, NULL);
 
     SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
         VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
-    spice_main_set_display(main_channel,
-                           channelid,
-                           0, 0, dw, dh);
+    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
 }
 
 static void
@@ -212,7 +210,8 @@ enable_accel_changed(VirtViewerApp *app,
 
 GtkWidget *
 virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
-                              SpiceChannel *channel)
+                              SpiceChannel *channel,
+                              gint monitorid)
 {
     VirtViewerDisplaySpice *self;
     VirtViewerApp *app;
@@ -222,15 +221,20 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
     g_return_val_if_fail(SPICE_IS_DISPLAY_CHANNEL(channel), NULL);
 
     g_object_get(channel, "channel-id", &channelid, NULL);
+    // We don't allow monitorid != 0 && channelid != 0
+    g_return_val_if_fail(channelid == 0 || monitorid == 0, NULL);
 
     self = g_object_new(VIRT_VIEWER_TYPE_DISPLAY_SPICE,
                         "session", session,
-                        "nth-display", channelid,
+                        // either monitorid is always 0 or channelid
+                        // is, we can't have display (0, 2) and (2, 0)
+                        // for example
+                        "nth-display", channelid + monitorid,
                         NULL);
     self->priv->channel = channel;
 
     g_object_get(session, "spice-session", &s, NULL);
-    self->priv->display = spice_display_new(s, channelid);
+    self->priv->display = spice_display_new_with_monitor(s, channelid, monitorid);
     g_object_unref(s);
 
     virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
diff --git a/src/virt-viewer-display-spice.h b/src/virt-viewer-display-spice.h
index 701ed85..c2013ec 100644
--- a/src/virt-viewer-display-spice.h
+++ b/src/virt-viewer-display-spice.h
@@ -66,7 +66,7 @@ struct _VirtViewerDisplaySpiceClass {
 
 GType virt_viewer_display_spice_get_type(void);
 
-GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel);
+GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel, gint monitorid);
 
 G_END_DECLS
 
diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
index 6577237..a728cb1 100644
--- a/src/virt-viewer-session-spice.c
+++ b/src/virt-viewer-session-spice.c
@@ -33,6 +33,7 @@
 #include "virt-viewer-session-spice.h"
 #include "virt-viewer-display-spice.h"
 #include "virt-viewer-auth.h"
+#include "virt-glib-compat.h"
 
 #if !GLIB_CHECK_VERSION(2, 26, 0)
 #include "gbinding.h"
@@ -414,6 +415,78 @@ agent_connected_changed(SpiceChannel *cmain,
 }
 
 static void
+destroy_display(gpointer data)
+{
+    VirtViewerDisplay *display = VIRT_VIEWER_DISPLAY(data);
+    VirtViewerSession *session = virt_viewer_display_get_session(display);
+
+    DEBUG_LOG("Destroying spice display %p", display);
+    virt_viewer_session_remove_display(session, display);
+    g_object_unref(display);
+}
+
+static void
+virt_viewer_session_spice_display_monitors(SpiceChannel *channel,
+                                           GParamSpec *pspec G_GNUC_UNUSED,
+                                           VirtViewerSessionSpice *self)
+{
+    GArray *monitors = NULL;
+    GPtrArray *displays = NULL;
+    GtkWidget *display;
+    guint i, monitors_max;
+
+    g_object_get(channel,
+                 "monitors", &monitors,
+                 "monitors-max", &monitors_max,
+                 NULL);
+    g_return_if_fail(monitors != NULL);
+    g_return_if_fail(monitors->len <= monitors_max);
+
+    displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
+    if (displays == NULL) {
+        displays = g_ptr_array_new();
+        g_ptr_array_set_free_func(displays, destroy_display);
+        g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
+                               displays, (GDestroyNotify)g_ptr_array_unref);
+    }
+
+    g_ptr_array_set_size(displays, monitors_max);
+
+    for (i = 0; i < monitors_max; i++) {
+        display = g_ptr_array_index(displays, i);
+        if (display == NULL) {
+            display = virt_viewer_display_spice_new(self, channel, i);
+            DEBUG_LOG("creating spice display (#:%d)", i);
+            g_ptr_array_index(displays, i) = g_object_ref(display);
+        }
+
+        g_object_freeze_notify(G_OBJECT(display));
+        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), FALSE);
+        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(self),
+                                        VIRT_VIEWER_DISPLAY(display));
+    }
+
+    for (i = 0; i < monitors->len; i++) {
+        SpiceDisplayMonitorConfig *monitor = &g_array_index(monitors, SpiceDisplayMonitorConfig, i);
+        display = g_ptr_array_index(displays, monitor->id);
+        g_return_if_fail(display != NULL);
+
+        if (monitor->width == 0 || monitor->width == 0)
+            continue;
+
+        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), TRUE);
+        virt_viewer_display_set_desktop_size(VIRT_VIEWER_DISPLAY(display),
+                                             monitor->width, monitor->height);
+    }
+
+    for (i = 0; i < monitors_max; i++)
+        g_object_thaw_notify(g_ptr_array_index(displays, i));
+
+    g_clear_pointer(&monitors, g_array_unref);
+
+}
+
+static void
 virt_viewer_session_spice_channel_new(SpiceSession *s,
                                       SpiceChannel *channel,
                                       VirtViewerSession *session)
@@ -441,20 +514,17 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
 
         g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
         agent_connected_changed(channel, NULL, self);
+
+        g_signal_emit_by_name(session, "session-connected");
     }
 
     if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
-        GtkWidget *display;
-
-        g_signal_emit_by_name(session, "session-connected");
+        g_signal_emit_by_name(session, "session-initialized");
 
-        DEBUG_LOG("new display channel (#%d)", id);
-        display = virt_viewer_display_spice_new(self, channel);
-        g_object_set_data(G_OBJECT(channel), "virt-viewer-display", display);
-        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
-                                        VIRT_VIEWER_DISPLAY(display));
+        g_signal_connect(channel, "notify::monitors",
+                         G_CALLBACK(virt_viewer_session_spice_display_monitors), self);
 
-        g_signal_emit_by_name(session, "session-initialized");
+        spice_channel_connect(channel);
     }
 
     if (SPICE_IS_INPUTS_CHANNEL(channel)) {
@@ -538,7 +608,6 @@ virt_viewer_session_spice_channel_destroy(G_GNUC_UNUSED SpiceSession *s,
     if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
         VirtViewerDisplay *display = g_object_get_data(G_OBJECT(channel), "virt-viewer-display");
         DEBUG_LOG("zap display channel (#%d, %p)", id, display);
-        virt_viewer_session_remove_display(session, display);
     }
 
     if (SPICE_IS_PLAYBACK_CHANNEL(channel) && self->priv->audio) {
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:59 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:59 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 17/25] Add
	virt_viewer_window_get_display()
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-18-git-send-email-marcandre.lureau@gmail.com>

Getter used in following changes.
---
 src/virt-viewer-window.c |   13 +++++++++++--
 src/virt-viewer-window.h |    1 +
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 46e44be..4bc427a 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -112,7 +112,8 @@ static void
 virt_viewer_window_get_property (GObject *object, guint property_id,
                                  GValue *value, GParamSpec *pspec)
 {
-    VirtViewerWindowPrivate *priv = VIRT_VIEWER_WINDOW(object)->priv;
+    VirtViewerWindow *self = VIRT_VIEWER_WINDOW(object);
+    VirtViewerWindowPrivate *priv = self->priv;
 
     switch (property_id) {
     case PROP_SUBTITLE:
@@ -124,7 +125,7 @@ virt_viewer_window_get_property (GObject *object, guint property_id,
         break;
 
     case PROP_DISPLAY:
-        g_value_set_object(value, priv->display);
+        g_value_set_object(value, virt_viewer_window_get_display(self));
         break;
 
     case PROP_CONTAINER:
@@ -1117,6 +1118,14 @@ virt_viewer_window_get_builder(VirtViewerWindow *self)
     return self->priv->builder;
 }
 
+VirtViewerDisplay*
+virt_viewer_window_get_display(VirtViewerWindow *self)
+{
+    g_return_val_if_fail(VIRT_VIEWER_WINDOW(self), FALSE);
+
+    return self->priv->display;
+}
+
 /*
  * Local variables:
  *  c-indent-level: 4
diff --git a/src/virt-viewer-window.h b/src/virt-viewer-window.h
index f80c4c7..44db585 100644
--- a/src/virt-viewer-window.h
+++ b/src/virt-viewer-window.h
@@ -63,6 +63,7 @@ GType virt_viewer_window_get_type (void);
 GtkWindow* virt_viewer_window_get_window (VirtViewerWindow* window);
 VirtViewerNotebook* virt_viewer_window_get_notebook (VirtViewerWindow* window);
 void virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *display);
+VirtViewerDisplay* virt_viewer_window_get_display(VirtViewerWindow *self);
 void virt_viewer_window_set_usb_options_sensitive(VirtViewerWindow *self, gboolean sensitive);
 void virt_viewer_window_update_title(VirtViewerWindow *self);
 void virt_viewer_window_show(VirtViewerWindow *self);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:00 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:00 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 18/25] Add
	VirtViewerDisplay::selectable property
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-19-git-send-email-marcandre.lureau@gmail.com>

This property will be set when the display can be selected to be
"enabled" and shown (this can involve creating/connecting an
additional guest monitor, and may need guest agent cooperation for
example).
---
 src/virt-viewer-display-spice.c |   16 ++++++++++++++++
 src/virt-viewer-display.c       |   23 +++++++++++++++++++++++
 src/virt-viewer-display.h       |    2 ++
 3 files changed, 41 insertions(+)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index d06e5cf..81eb16f 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -47,6 +47,7 @@ static void virt_viewer_display_spice_send_keys(VirtViewerDisplay *display,
 static GdkPixbuf *virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display);
 static void virt_viewer_display_spice_release_cursor(VirtViewerDisplay *display);
 static void virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED);
+static gboolean virt_viewer_display_spice_selectable(VirtViewerDisplay *display);
 
 static void
 virt_viewer_display_spice_finalize(GObject *obj)
@@ -70,6 +71,7 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
     dclass->get_pixbuf = virt_viewer_display_spice_get_pixbuf;
     dclass->release_cursor = virt_viewer_display_spice_release_cursor;
     dclass->close = virt_viewer_display_spice_close;
+    dclass->selectable = virt_viewer_display_spice_selectable;
 
     g_type_class_add_private(klass, sizeof(VirtViewerDisplaySpicePrivate));
 }
@@ -279,6 +281,20 @@ virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED)
 {
 }
 
+static gboolean
+virt_viewer_display_spice_selectable(VirtViewerDisplay *self)
+{
+    gboolean agent_connected;
+    SpiceMainChannel *mainc;
+
+    mainc = get_main(self);
+    g_object_get(mainc,
+                 "agent-connected", &agent_connected,
+                 NULL);
+
+    return agent_connected;
+}
+
 
 /*
  * Local variables:
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 18fed19..8d13733 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -80,6 +80,7 @@ enum {
     PROP_ZOOM_LEVEL,
     PROP_SHOW_HINT,
     PROP_SESSION,
+    PROP_SELECTABLE,
 };
 
 static void
@@ -167,6 +168,13 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
                                                         G_PARAM_READWRITE |
                                                         G_PARAM_CONSTRUCT_ONLY));
 
+    g_object_class_install_property(object_class,
+                                    PROP_SELECTABLE,
+                                    g_param_spec_boolean("selectable",
+                                                         "Selectable",
+                                                         "Selectable",
+                                                         FALSE,
+                                                         G_PARAM_READABLE));
 
     g_signal_new("display-pointer-grab",
                  G_OBJECT_CLASS_TYPE(object_class),
@@ -302,6 +310,9 @@ virt_viewer_display_get_property(GObject *object,
     case PROP_SESSION:
         g_value_set_object(value, virt_viewer_display_get_session(display));
         break;
+    case PROP_SELECTABLE:
+        g_value_set_boolean(value, virt_viewer_display_get_selectable(display));
+        break;
 
     default:
         G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
@@ -601,6 +612,18 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *self)
     klass->release_cursor(self);
 }
 
+gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *self)
+{
+    VirtViewerDisplayClass *klass;
+
+    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), FALSE);
+
+    klass = VIRT_VIEWER_DISPLAY_GET_CLASS(self);
+    if (klass->selectable)
+        return klass->selectable(self);
+
+    return TRUE;
+}
 
 void virt_viewer_display_close(VirtViewerDisplay *self)
 {
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index 89b4817..4247570 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -77,6 +77,7 @@ struct _VirtViewerDisplayClass {
     void (*release_cursor)(VirtViewerDisplay *display);
 
     void (*close)(VirtViewerDisplay *display);
+    gboolean (*selectable)(VirtViewerDisplay *display);
 
     /* signals */
     void (*display_pointer_grab)(VirtViewerDisplay *display);
@@ -118,6 +119,7 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
 
 void virt_viewer_display_close(VirtViewerDisplay *display);
 void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
+gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *display);
 
 G_END_DECLS
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:01 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:01 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 19/25] spice: factor out main
	channel lookup code
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-20-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-display-spice.c |   22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 81eb16f..b2f7e6c 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -76,14 +76,27 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
     g_type_class_add_private(klass, sizeof(VirtViewerDisplaySpicePrivate));
 }
 
+static SpiceMainChannel*
+get_main(VirtViewerDisplay *self)
+{
+    VirtViewerSessionSpice *session;
+
+    session = VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self));
+
+    return virt_viewer_session_spice_get_main_channel(session);
+}
+
 static void
 show_hint_changed(VirtViewerDisplay *self)
 {
-    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
-        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(self)));
+    SpiceMainChannel *main_channel = get_main(self);
     guint enabled = TRUE;
     guint nth;
 
+    /* this may happen when finalizing */
+    if (!main_channel)
+        return;
+
     g_object_get(self, "nth-display", &nth, NULL);
     if (virt_viewer_display_get_show_hint(self) & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED)
         enabled = FALSE;
@@ -191,9 +204,8 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
 
     g_object_get(self, "nth-display", &nth, NULL);
 
-    SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
-        VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
-    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
+    spice_main_set_display(get_main(VIRT_VIEWER_DISPLAY(self)),
+                           nth, 0, 0, dw, dh);
 }
 
 static void
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:02 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:02 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 20/25] Make the display
	submenu insensitive if display can't be selected
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-21-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-app.c |   18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index 3d7ff68..a88264b 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -1635,15 +1635,31 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
     while (tmp) {
         int *nth = tmp->data;
         VirtViewerWindow *vwin = VIRT_VIEWER_WINDOW(g_hash_table_lookup(self->priv->windows, nth));
+        VirtViewerDisplay *display = virt_viewer_window_get_display(vwin);
         GtkWidget *item;
-        gboolean visible;
+        gboolean visible, sensitive = FALSE;
         gchar *label;
 
         label = g_strdup_printf(_("Display %d"), *nth + 1);
         item = gtk_check_menu_item_new_with_label(label);
         g_free(label);
+
         visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));
         gtk_check_menu_item_set_active(GTK_CHECK_MENU_ITEM(item), visible);
+
+        if (display) {
+            guint hint = virt_viewer_display_get_show_hint(display);
+
+            if (hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_READY)
+                sensitive = TRUE;
+
+            if ((hint & VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED) &&
+                virt_viewer_display_get_selectable(display))
+                sensitive = TRUE;
+        }
+
+        gtk_widget_set_sensitive(item, sensitive);
+
         g_signal_connect(G_OBJECT(item),
                          "toggled", G_CALLBACK(menu_display_visible_toggled_cb), vwin);
         gtk_menu_shell_append(submenu, item);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:03 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:03 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 21/25] Add
	VirtViewerSession::session-display-updated
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-22-git-send-email-marcandre.lureau@gmail.com>

Rebuild menu when agent is connected. Only when the agent is running
may a display be enabled/disabled.
---
 src/virt-viewer-app.c           |    9 +++++++++
 src/virt-viewer-session-spice.c |   18 ++++++++++++++----
 src/virt-viewer-session.c       |   10 ++++++++++
 src/virt-viewer-session.h       |    1 +
 4 files changed, 34 insertions(+), 4 deletions(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index a88264b..def52c5 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -685,6 +685,13 @@ virt_viewer_app_display_removed(VirtViewerSession *session G_GNUC_UNUSED,
         virt_viewer_app_remove_nth_window(self, nth);
 }
 
+static void
+virt_viewer_app_display_updated(VirtViewerSession *session G_GNUC_UNUSED,
+                                VirtViewerApp *self)
+{
+    virt_viewer_app_update_menu_displays(self);
+}
+
 int
 virt_viewer_app_create_session(VirtViewerApp *self, const gchar *type)
 {
@@ -735,6 +742,8 @@ virt_viewer_app_create_session(VirtViewerApp *self, const gchar *type)
                      G_CALLBACK(virt_viewer_app_display_added), self);
     g_signal_connect(priv->session, "session-display-removed",
                      G_CALLBACK(virt_viewer_app_display_removed), self);
+    g_signal_connect(priv->session, "session-display-updated",
+                     G_CALLBACK(virt_viewer_app_display_updated), self);
 
     g_signal_connect(priv->session, "session-cut-text",
                      G_CALLBACK(virt_viewer_app_server_cut_text), self);
diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
index a728cb1..5fcd7fb 100644
--- a/src/virt-viewer-session-spice.c
+++ b/src/virt-viewer-session-spice.c
@@ -406,12 +406,21 @@ virt_viewer_session_spice_usb_device_selection(VirtViewerSession *session,
 }
 
 static void
-agent_connected_changed(SpiceChannel *cmain,
+agent_connected_changed(SpiceChannel *cmain G_GNUC_UNUSED,
                         GParamSpec *pspec G_GNUC_UNUSED,
                         VirtViewerSessionSpice *self)
 {
+    // this will force refresh of application menu
+    g_signal_emit_by_name(self, "session-display-updated");
+}
+
+static void
+agent_connected_fullscreen_auto_conf(SpiceChannel *cmain,
+                                     GParamSpec *pspec G_GNUC_UNUSED,
+                                     VirtViewerSessionSpice *self)
+{
     if (virt_viewer_session_spice_fullscreen_auto_conf(self))
-        g_signal_handlers_disconnect_by_func(cmain, agent_connected_changed, self);
+        g_signal_handlers_disconnect_by_func(cmain, agent_connected_fullscreen_auto_conf, self);
 }
 
 static void
@@ -512,8 +521,9 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
                          G_CALLBACK(virt_viewer_session_spice_main_channel_event), self);
         self->priv->main_channel = SPICE_MAIN_CHANNEL(channel);
 
-        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
-        agent_connected_changed(channel, NULL, self);
+        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed), self);
+        g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_fullscreen_auto_conf), self);
+        agent_connected_fullscreen_auto_conf(channel, NULL, self);
 
         g_signal_emit_by_name(session, "session-connected");
     }
diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
index 9249a1f..a1d96c2 100644
--- a/src/virt-viewer-session.c
+++ b/src/virt-viewer-session.c
@@ -233,6 +233,16 @@ virt_viewer_session_class_init(VirtViewerSessionClass *class)
                  1,
                  VIRT_VIEWER_TYPE_DISPLAY);
 
+    g_signal_new("session-display-updated",
+                 G_OBJECT_CLASS_TYPE(object_class),
+                 G_SIGNAL_RUN_LAST | G_SIGNAL_NO_HOOKS,
+                 G_STRUCT_OFFSET(VirtViewerSessionClass, session_display_updated),
+                 NULL,
+                 NULL,
+                 g_cclosure_marshal_VOID__VOID,
+                 G_TYPE_NONE,
+                 0);
+
     g_signal_new("session-cut-text",
                  G_OBJECT_CLASS_TYPE(object_class),
                  G_SIGNAL_RUN_LAST | G_SIGNAL_NO_HOOKS,
diff --git a/src/virt-viewer-session.h b/src/virt-viewer-session.h
index 44e4674..38ed988 100644
--- a/src/virt-viewer-session.h
+++ b/src/virt-viewer-session.h
@@ -88,6 +88,7 @@ struct _VirtViewerSessionClass {
                                   VirtViewerDisplay *display);
     void (*session_display_removed)(VirtViewerSession *session,
                                     VirtViewerDisplay *display);
+    void (*session_display_updated)(VirtViewerSession *session);
 
     void (*session_cut_text)(VirtViewerSession *session, const gchar *str);
     void (*session_bell)(VirtViewerSession *session);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:04 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:04 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 22/25] Simplify display flag
	handling
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-23-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-display-spice.c |    9 +--------
 src/virt-viewer-display.c       |   20 ++++++++++----------
 src/virt-viewer-display.h       |    2 +-
 src/virt-viewer-session-vnc.c   |    2 +-
 4 files changed, 13 insertions(+), 20 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index b2f7e6c..8845ee0 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -141,19 +141,12 @@ display_ready(GObject *display,
               GParamSpec *pspec G_GNUC_UNUSED,
               VirtViewerDisplay *self)
 {
-    guint hint;
     gboolean ready;
 
     g_object_get(display, "ready", &ready, NULL);
     DEBUG_LOG("display %p ready:%d", self, ready);
 
-    hint = virt_viewer_display_get_show_hint(self);
-    if (ready)
-        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
-    else
-        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_READY;
-
-    virt_viewer_display_set_show_hint(self, hint);
+    virt_viewer_display_set_show_hint(self, VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, ready);
 }
 
 static void
diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
index 8d13733..fe7ce4f 100644
--- a/src/virt-viewer-display.c
+++ b/src/virt-viewer-display.c
@@ -551,12 +551,20 @@ guint virt_viewer_display_get_show_hint(VirtViewerDisplay *self)
     return self->priv->show_hint;
 }
 
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint mask, gboolean enable)
 {
     VirtViewerDisplayPrivate *priv;
+    guint hint;
     g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
 
     priv = self->priv;
+    hint = priv->show_hint;
+
+    if (enable)
+        hint |= mask;
+    else
+        hint &= ~mask;
+
     if (priv->show_hint == hint)
         return;
 
@@ -566,17 +574,9 @@ void virt_viewer_display_set_show_hint(VirtViewerDisplay *self, guint hint)
 
 void virt_viewer_display_set_enabled(VirtViewerDisplay *self, gboolean enabled)
 {
-    guint hint;
-
     g_return_if_fail(VIRT_VIEWER_IS_DISPLAY(self));
 
-    hint = virt_viewer_display_get_show_hint(self);
-    if (enabled)
-        hint &= ~VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
-    else
-        hint |= VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED;
-
-    virt_viewer_display_set_show_hint(self, hint);
+    virt_viewer_display_set_show_hint(self, VIRT_VIEWER_DISPLAY_SHOW_HINT_DISABLED, !enabled);
 }
 
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *self)
diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
index 4247570..af78709 100644
--- a/src/virt-viewer-display.h
+++ b/src/virt-viewer-display.h
@@ -110,7 +110,7 @@ gboolean virt_viewer_display_get_zoom(VirtViewerDisplay *display);
 void virt_viewer_display_send_keys(VirtViewerDisplay *display,
                                    const guint *keyvals, int nkeyvals);
 GdkPixbuf* virt_viewer_display_get_pixbuf(VirtViewerDisplay *display);
-void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint hint);
+void virt_viewer_display_set_show_hint(VirtViewerDisplay *display, guint mask, gboolean enable);
 guint virt_viewer_display_get_show_hint(VirtViewerDisplay *display);
 VirtViewerSession* virt_viewer_display_get_session(VirtViewerDisplay *display);
 void virt_viewer_display_set_auto_resize(VirtViewerDisplay *display, gboolean auto_resize);
diff --git a/src/virt-viewer-session-vnc.c b/src/virt-viewer-session-vnc.c
index 238c865..a24d65d 100644
--- a/src/virt-viewer-session-vnc.c
+++ b/src/virt-viewer-session-vnc.c
@@ -95,7 +95,7 @@ virt_viewer_session_vnc_connected(VncDisplay *vnc G_GNUC_UNUSED,
     GtkWidget *display = virt_viewer_display_vnc_new(session->priv->vnc);
     g_signal_emit_by_name(session, "session-connected");
     virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(display),
-                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY);
+                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, TRUE);
     virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
                                     VIRT_VIEWER_DISPLAY(display));
 }
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:05 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:05 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 23/25] vnc: when session is
	disconnected, make the display as non-ready
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-24-git-send-email-marcandre.lureau@gmail.com>

---
 src/virt-viewer-session-vnc.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/src/virt-viewer-session-vnc.c b/src/virt-viewer-session-vnc.c
index a24d65d..52baf5c 100644
--- a/src/virt-viewer-session-vnc.c
+++ b/src/virt-viewer-session-vnc.c
@@ -104,11 +104,11 @@ static void
 virt_viewer_session_vnc_disconnected(VncDisplay *vnc G_GNUC_UNUSED,
                                      VirtViewerSessionVnc *session)
 {
+    GtkWidget *display = virt_viewer_display_vnc_new(session->priv->vnc);
     DEBUG_LOG("Disconnected");
     g_signal_emit_by_name(session, "session-disconnected");
-    /* TODO perhaps? */
-    /* virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(session->priv->vnc), */
-    /*                                   VIRT_VIEWER_DISPLAY_SHOW_HINT_HIDE); */
+    virt_viewer_display_set_show_hint(VIRT_VIEWER_DISPLAY(display),
+                                      VIRT_VIEWER_DISPLAY_SHOW_HINT_READY, FALSE);
 }
 
 static void
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:06 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:06 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 24/25] Make status widget
	visible immediately
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-25-git-send-email-marcandre.lureau@gmail.com>

GtkNotebook will use the currently visible widget as default page.
If we don't show status widget before we append the display, the
current page will be on display. Quoting Gtk+ documentation:

"Note that due to historical reasons, GtkNotebook refuses to switch to
a page unless the child widget is visible. Therefore, it is
recommended to show child widgets before adding them to a notebook."
---
 src/virt-viewer-notebook.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/src/virt-viewer-notebook.c b/src/virt-viewer-notebook.c
index 09e8ec4..cc7136b 100644
--- a/src/virt-viewer-notebook.c
+++ b/src/virt-viewer-notebook.c
@@ -86,6 +86,7 @@ virt_viewer_notebook_init (VirtViewerNotebook *self)
     priv->status = gtk_label_new("");
     gtk_notebook_set_show_tabs(GTK_NOTEBOOK(self), FALSE);
     gtk_notebook_set_show_border(GTK_NOTEBOOK(self), FALSE);
+    gtk_widget_show_all(priv->status);
     gtk_notebook_append_page(GTK_NOTEBOOK(self), priv->status, NULL);
     gdk_color_parse("white", &color);
     gtk_widget_modify_fg(priv->status, GTK_STATE_NORMAL, &color);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:25:07 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:25:07 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 25/25] Show status page by
	default if display not ready
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-26-git-send-email-marcandre.lureau@gmail.com>

Switch back to status page if display is not ready.
---
 src/virt-viewer-window.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
index 4bc427a..7c6e41f 100644
--- a/src/virt-viewer-window.c
+++ b/src/virt-viewer-window.c
@@ -1040,8 +1040,12 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
         virt_viewer_display_set_zoom_level(VIRT_VIEWER_DISPLAY(priv->display), priv->zoomlevel);
         virt_viewer_display_set_auto_resize(VIRT_VIEWER_DISPLAY(priv->display), priv->auto_resize);
 
-        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
         gtk_widget_show_all(GTK_WIDGET(display));
+        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
+        /* switch back to non-display if not ready */
+        if (!(virt_viewer_display_get_show_hint(display) &
+              VIRT_VIEWER_DISPLAY_SHOW_HINT_READY))
+            gtk_notebook_set_current_page(GTK_NOTEBOOK(priv->notebook), 0);
 
         virt_viewer_signal_connect_object(display, "display-pointer-grab",
                                           G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:42 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:42 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 00/25] Add new Spice monitors
	handling support v2
Message-ID: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>

Hi,

Spice now supports multiple arbitrary-sized monitors per display
channel. The client has to learn how to handle the new events in order
to create, to show, and to manage the displays appropriately.

The feature is currently being merged in the various projects, to test
it you'll need the following projects update with monitor config
support: spice-server, qemu, spice-gtk, qxl driver and vdagent.

Changes since v1, after Christophe review:
 - splitted into more smaller patches
 - bumped to glib 2.22 and removed fallback code
 - added g_clear_pointer() virt-glib-compat.h fallback
 - renamed s/cmain/main_channel, s/c/monitor, s/enabled/enable
 - removed the display->dirty idle flag change, it was borked
 - more comments and spelling fixes
 - monitor count/max range is now checked by spice-gtk

cheers


Marc-Andr? Lureau (25):
  build: allow building with newer glibc-headers and -O0
  spice: improve fullscreen=auto-conf
  spice: disconnect signal handlers when either object is destroyed
  build: generate enums type boilerplate
  Destroy dialog immediately after run
  Turn display:show-hint into flags type
  Run-time check values before doing bad computation
  Mark a menu string as translatable
  Rebuild display menu when a window is added or removed
  Use virt_viewer_connect_object() for display
  Add a DISABLED display hint
  Prevent from adding the same display several time in the session
  Number display starting from 1
  Use SpiceDisplay:ready property instead of channel mark
  Bump glib > 2.22, add compat file
  Hook up handling of Monitors
  Add virt_viewer_window_get_display()
  Add VirtViewerDisplay::selectable property
  spice: factor out main channel lookup code
  Make the display submenu insensitive if display can't be selected
  Add VirtViewerSession::session-display-updated
  Simplify display flag handling
  vnc: when session is disconnected, make the display as non-ready
  Make status widget visible immediately
  Show status page by default if display not ready

 acinclude.m4                      |    9 ++-
 configure.ac                      |    7 ++-
 src/Makefile.am                   |   24 +++++++-
 src/virt-glib-compat.h            |   56 +++++++++++++++++
 src/virt-viewer-app.c             |   53 ++++++++++++----
 src/virt-viewer-display-spice.c   |  123 ++++++++++++++++++++++++-------------
 src/virt-viewer-display-spice.h   |    2 +-
 src/virt-viewer-display.c         |   70 +++++++++++++++++----
 src/virt-viewer-display.h         |   15 +++--
 src/virt-viewer-enums.c.etemplate |   60 ++++++++++++++++++
 src/virt-viewer-enums.h.etemplate |   41 +++++++++++++
 src/virt-viewer-notebook.c        |    1 +
 src/virt-viewer-session-spice.c   |  111 ++++++++++++++++++++++++++++-----
 src/virt-viewer-session-vnc.c     |    8 +--
 src/virt-viewer-session.c         |   13 ++++
 src/virt-viewer-session.h         |    1 +
 src/virt-viewer-util.c            |  117 +++++++++++++++++++++++++++++++++++
 src/virt-viewer-util.h            |    6 ++
 src/virt-viewer-window.c          |   67 +++++++++++++++-----
 src/virt-viewer-window.h          |    1 +
 20 files changed, 672 insertions(+), 113 deletions(-)
 create mode 100644 src/virt-glib-compat.h
 create mode 100644 src/virt-viewer-enums.c.etemplate
 create mode 100644 src/virt-viewer-enums.h.etemplate

-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:45 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:45 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 03/25] spice: disconnect
	signal handlers when either object is destroyed
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-4-git-send-email-marcandre.lureau@gmail.com>

Use virt_viewer_signal_connect_object(), a copy of telepathy
utility function tp_g_signal_connect_object(). This function
will take care of removing signal handler if any of emitter or
attached object are destroyed.

The following patches will have this condition met, since there is no
longer 1-1 relation between channel and display. The channels can
continue to be around when some of the display are removed.
---
 src/virt-viewer-display-spice.c |   26 ++++-----
 src/virt-viewer-util.c          |  117 +++++++++++++++++++++++++++++++++++++++
 src/virt-viewer-util.h          |    6 ++
 3 files changed, 135 insertions(+), 14 deletions(-)

diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
index 0b6949b..7bf9a9a 100644
--- a/src/virt-viewer-display-spice.c
+++ b/src/virt-viewer-display-spice.c
@@ -219,10 +219,10 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
     self->priv->display = spice_display_new(s, channelid);
     g_object_unref(s);
 
-    g_signal_connect(channel, "display-primary-create",
-                     G_CALLBACK(primary_create), self);
-    g_signal_connect(channel, "display-mark",
-                     G_CALLBACK(display_mark), self);
+    virt_viewer_signal_connect_object(channel, "display-primary-create",
+                                      G_CALLBACK(primary_create), self, 0);
+    virt_viewer_signal_connect_object(channel, "display-mark",
+                                      G_CALLBACK(display_mark), self, 0);
 
     gtk_container_add(GTK_CONTAINER(self), g_object_ref(self->priv->display));
     gtk_widget_show(GTK_WIDGET(self->priv->display));
@@ -233,19 +233,17 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
                  "scaling", TRUE,
                  NULL);
 
-    g_signal_connect(self->priv->display,
-                     "keyboard-grab",
-                     G_CALLBACK(virt_viewer_display_spice_keyboard_grab), self);
-    g_signal_connect(self->priv->display,
-                     "mouse-grab",
-                     G_CALLBACK(virt_viewer_display_spice_mouse_grab), self);
-    g_signal_connect(self,
-                     "size-allocate",
-                     G_CALLBACK(virt_viewer_display_spice_size_allocate), self);
+    virt_viewer_signal_connect_object(self->priv->display, "keyboard-grab",
+                                      G_CALLBACK(virt_viewer_display_spice_keyboard_grab), self, 0);
+    virt_viewer_signal_connect_object(self->priv->display, "mouse-grab",
+                                      G_CALLBACK(virt_viewer_display_spice_mouse_grab), self, 0);
+    virt_viewer_signal_connect_object(self, "size-allocate",
+                                      G_CALLBACK(virt_viewer_display_spice_size_allocate), self, 0);
 
 
     app = virt_viewer_session_get_app(VIRT_VIEWER_SESSION(session));
-    g_signal_connect(app, "notify::enable-accel", G_CALLBACK(enable_accel_changed), self);
+    virt_viewer_signal_connect_object(app, "notify::enable-accel",
+                                      G_CALLBACK(enable_accel_changed), self, 0);
     enable_accel_changed(app, NULL, self);
 
     return GTK_WIDGET(self);
diff --git a/src/virt-viewer-util.c b/src/virt-viewer-util.c
index c1182af..9a163fa 100644
--- a/src/virt-viewer-util.c
+++ b/src/virt-viewer-util.c
@@ -136,6 +136,123 @@ virt_viewer_util_extract_host(const char *uristr,
     return 0;
 }
 
+typedef struct {
+    GObject *instance;
+    GObject *observer;
+    GClosure *closure;
+    gulong handler_id;
+} WeakHandlerCtx;
+
+static WeakHandlerCtx *
+whc_new(GObject *instance,
+        GObject *observer)
+{
+    WeakHandlerCtx *ctx = g_slice_new0(WeakHandlerCtx);
+
+    ctx->instance = instance;
+    ctx->observer = observer;
+
+    return ctx;
+}
+
+static void
+whc_free(WeakHandlerCtx *ctx)
+{
+    g_slice_free(WeakHandlerCtx, ctx);
+}
+
+static void observer_destroyed_cb(gpointer, GObject *);
+static void closure_invalidated_cb(gpointer, GClosure *);
+
+/*
+ * If signal handlers are removed before the object is destroyed, this
+ * callback will never get triggered.
+ */
+static void
+instance_destroyed_cb(gpointer ctx_,
+                      GObject *where_the_instance_was G_GNUC_UNUSED)
+{
+    WeakHandlerCtx *ctx = ctx_;
+
+    /* No need to disconnect the signal here, the instance has gone away. */
+    g_object_weak_unref(ctx->observer, observer_destroyed_cb, ctx);
+    g_closure_remove_invalidate_notifier(ctx->closure, ctx,
+                                         closure_invalidated_cb);
+    whc_free(ctx);
+}
+
+/* Triggered when the observer is destroyed. */
+static void
+observer_destroyed_cb(gpointer ctx_,
+                      GObject *where_the_observer_was G_GNUC_UNUSED)
+{
+    WeakHandlerCtx *ctx = ctx_;
+
+    g_closure_remove_invalidate_notifier(ctx->closure, ctx,
+                                         closure_invalidated_cb);
+    g_signal_handler_disconnect(ctx->instance, ctx->handler_id);
+    g_object_weak_unref(ctx->instance, instance_destroyed_cb, ctx);
+    whc_free(ctx);
+}
+
+/* Triggered when either object is destroyed or the handler is disconnected. */
+static void
+closure_invalidated_cb(gpointer ctx_,
+                       GClosure *where_the_closure_was G_GNUC_UNUSED)
+{
+    WeakHandlerCtx *ctx = ctx_;
+
+    g_object_weak_unref(ctx->instance, instance_destroyed_cb, ctx);
+    g_object_weak_unref(ctx->observer, observer_destroyed_cb, ctx);
+    whc_free(ctx);
+}
+
+/* Copied from tp_g_signal_connect_object. */
+/**
+  * virt_viewer_signal_connect_object: (skip)
+  * @instance: the instance to connect to.
+  * @detailed_signal: a string of the form "signal-name::detail".
+  * @c_handler: the #GCallback to connect.
+  * @gobject: the object to pass as data to @c_handler.
+  * @connect_flags: a combination of #GConnectFlags.
+  *
+  * Similar to g_signal_connect_object() but will delete connection
+  * when any of the objects is destroyed.
+  *
+  * Returns: the handler id.
+  */
+gulong virt_viewer_signal_connect_object(gpointer instance,
+                                         const gchar *detailed_signal,
+                                         GCallback c_handler,
+                                         gpointer gobject,
+                                         GConnectFlags connect_flags)
+{
+    GObject *instance_obj = G_OBJECT(instance);
+    WeakHandlerCtx *ctx = whc_new(instance_obj, gobject);
+
+    g_return_val_if_fail(G_TYPE_CHECK_INSTANCE (instance), 0);
+    g_return_val_if_fail(detailed_signal != NULL, 0);
+    g_return_val_if_fail(c_handler != NULL, 0);
+    g_return_val_if_fail(G_IS_OBJECT (gobject), 0);
+    g_return_val_if_fail((connect_flags & ~(G_CONNECT_AFTER|G_CONNECT_SWAPPED)) == 0, 0);
+
+    if (connect_flags & G_CONNECT_SWAPPED)
+        ctx->closure = g_cclosure_new_object_swap(c_handler, gobject);
+    else
+        ctx->closure = g_cclosure_new_object(c_handler, gobject);
+
+    ctx->handler_id = g_signal_connect_closure(instance, detailed_signal,
+                                               ctx->closure, (connect_flags & G_CONNECT_AFTER) ? TRUE : FALSE);
+
+    g_object_weak_ref(instance_obj, instance_destroyed_cb, ctx);
+    g_object_weak_ref(gobject, observer_destroyed_cb, ctx);
+    g_closure_add_invalidate_notifier(ctx->closure, ctx,
+                                      closure_invalidated_cb);
+
+    return ctx->handler_id;
+}
+
+
 /*
  * Local variables:
  *  c-indent-level: 4
diff --git a/src/virt-viewer-util.h b/src/virt-viewer-util.h
index 3a40651..38c8078 100644
--- a/src/virt-viewer-util.h
+++ b/src/virt-viewer-util.h
@@ -40,6 +40,12 @@ int virt_viewer_util_extract_host(const char *uristr,
                                   char **user,
                                   int *port);
 
+gulong virt_viewer_signal_connect_object(gpointer instance,
+                                         const gchar *detailed_signal,
+                                         GCallback c_handler,
+                                         gpointer gobject,
+                                         GConnectFlags connect_flags);
+
 #endif
 
 /*
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 17 19:24:46 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 17 Jul 2012 21:24:46 +0200
Subject: [virt-tools-list] =?utf-8?q?=5BPATCH_virt-viewer_04/25=5D_build?=
	=?utf-8?q?=3A_generate_enums_type_boilerplate?=
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1342553107-27947-5-git-send-email-marcandre.lureau@gmail.com>

---
 configure.ac                      |    3 ++
 src/Makefile.am                   |   23 ++++++++++++--
 src/virt-viewer-enums.c.etemplate |   60 +++++++++++++++++++++++++++++++++++++
 src/virt-viewer-enums.h.etemplate |   41 +++++++++++++++++++++++++
 4 files changed, 124 insertions(+), 3 deletions(-)
 create mode 100644 src/virt-viewer-enums.c.etemplate
 create mode 100644 src/virt-viewer-enums.h.etemplate

diff --git a/configure.ac b/configure.ac
index a2471f4..548b945 100644
--- a/configure.ac
+++ b/configure.ac
@@ -72,6 +72,9 @@ AC_DEFINE_UNQUOTED([GETTEXT_PACKAGE],"$GETTEXT_PACKAGE", [GETTEXT package name])
 AM_GLIB_GNU_GETTEXT
 IT_PROG_INTLTOOL([0.35.0])
 
+PKG_PROG_PKG_CONFIG
+GLIB_MKENUMS=`$PKG_CONFIG --variable=glib_mkenums glib-2.0`
+AC_SUBST(GLIB_MKENUMS)
 
 PKG_CHECK_MODULES(GLIB2, glib-2.0 >= $GLIB2_REQUIRED gthread-2.0 gmodule-export-2.0)
 PKG_CHECK_MODULES(LIBXML2, libxml-2.0 >= $LIBXML2_REQUIRED)
diff --git a/src/Makefile.am b/src/Makefile.am
index d99b043..5daa98a 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -10,11 +10,28 @@ builderxml_DATA =				\
 	virt-viewer-auth.xml			\
 	$(NULL)
 
-EXTRA_DIST = $(builderxml_DATA) \
-	gbinding.c \
-	gbinding.h
+EXTRA_DIST =					\
+	$(builderxml_DATA)			\
+	virt-viewer-enums.c.etemplate		\
+	virt-viewer-enums.h.etemplate		\
+	gbinding.c				\
+	gbinding.h				\
+	$(NULL)
+
+ENUMS_FILES =					\
+	virt-viewer-display.h			\
+	$(NULL)
+
+BUILT_SOURCES =					\
+	virt-viewer-enums.h			\
+	virt-viewer-enums.c			\
+	$(NULL)
+
+$(BUILT_SOURCES): %: %.etemplate $(ENUMS_FILES)
+	$(AM_V_GEN)$(GLIB_MKENUMS) --template $^ | sed -e 's/VIRT_TYPE_VIEWER/VIRT_VIEWER_TYPE/' > $@
 
 COMMON_SOURCES =					\
+	$(BUILT_SOURCES)				\
 	virt-gtk-compat.h				\
 	virt-viewer-util.h virt-viewer-util.c		\
 	virt-viewer-auth.h virt-viewer-auth.c		\
diff --git a/src/virt-viewer-enums.c.etemplate b/src/virt-viewer-enums.c.etemplate
new file mode 100644
index 0000000..6f9925f
--- /dev/null
+++ b/src/virt-viewer-enums.c.etemplate
@@ -0,0 +1,60 @@
+/*** BEGIN file-header ***/
+/*
+ * Virt Viewer: A virtual machine console viewer
+ *
+ * Copyright (C) 2007-2012 Red Hat, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Author: Marc-Andr? Lureau <marcandre.lureau at redhat.com>
+ */
+
+#include "virt-viewer-enums.h"
+
+/*** END file-header ***/
+
+/*** BEGIN file-production ***/
+#include "@filename@"
+/*** END file-production ***/
+
+
+/*** BEGIN value-header ***/
+
+GType
+ at enum_name@_get_type (void)
+{
+  static volatile gsize g_define_type_id__volatile = 0;
+
+  if (g_once_init_enter (&g_define_type_id__volatile))
+    {
+      static const G at Type@Value values[] = {
+/*** END value-header ***/
+
+/*** BEGIN value-production ***/
+        { @VALUENAME@, "@VALUENAME@", "@valuenick@" },
+/*** END value-production ***/
+
+/*** BEGIN value-tail ***/
+        { 0, NULL, NULL }
+      };
+      GType g_define_type_id =
+        g_ at type@_register_static (g_intern_static_string ("@EnumName@"), values);
+      g_once_init_leave (&g_define_type_id__volatile, g_define_type_id);
+    }
+
+  return g_define_type_id__volatile;
+}
+
+/*** END value-tail ***/
diff --git a/src/virt-viewer-enums.h.etemplate b/src/virt-viewer-enums.h.etemplate
new file mode 100644
index 0000000..f33158d
--- /dev/null
+++ b/src/virt-viewer-enums.h.etemplate
@@ -0,0 +1,41 @@
+/*** BEGIN file-header ***/
+/*
+ * Virt Viewer: A virtual machine console viewer
+ *
+ * Copyright (C) 2007-2012 Red Hat, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Author: Marc-Andr? Lureau <marcandre.lureau at redhat.com>
+ */
+
+#ifndef VIRT_VIEWER_ENUMS_H
+#define VIRT_VIEWER_ENUMS_H
+
+#include <glib-object.h>
+
+G_BEGIN_DECLS
+/*** END file-header ***/
+
+/*** BEGIN value-header ***/
+GType @enum_name at _get_type (void) G_GNUC_CONST;
+#define @ENUMPREFIX at _TYPE_@ENUMSHORT@ (@enum_name at _get_type ())
+/*** END value-header ***/
+
+/*** BEGIN file-tail ***/
+G_END_DECLS
+
+#endif
+/*** END file-tail ***/
-- 
1.7.10.4



From felicitus at felicitus.org  Tue Jul 17 23:10:02 2012
From: felicitus at felicitus.org (Felicitus)
Date: Wed, 18 Jul 2012 01:10:02 +0200
Subject: [virt-tools-list] virt-manager slow via SSH - but NOT the console
Message-ID: <5005F0CA.4050301@felicitus.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I'm using virt-manager via SSH. The docs say that the _initial_
connect could be slow, which is correct in my case.

However, anything I do within virt-manager is also slow, for example,
adding new hardware takes almost a minute to display the extra window,
and navigating to the "network device" section takes 5 minutes.

The odd thing is that the VNC console is actually very fast.

This seems to be very odd, and I really don't know where to start
debugging. Any hints how to get more debug output from virt-manager?

Thanks,
Felicitus
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJQBfDKAAoJEJYpjrA42x19MY4H/3HDbgcz7vG0sGgxWLqG11pk
yDEIzbLZliaxiHlGgWBoEUtD/o8rlUlEhXLFCMhqhgn2BQ341xXvuTUMuqKAvlmm
6UPKHtJIZALLMTAqsoarPH16QgGo+vgKg5RLHSGUvz6HHfnPSrGnBN3tQfRUC+pN
BzZsw6iWo4FuxKNT7eA6QD3elX4dLfixq/b+Uiwka4ICySQAbojnZPgvmGY48QaJ
bl8Zkd/r69FquwSmSsOwyr95xpu3hdPIsXfgsPeZjAEpAbVVTQVjST8zpoz85W1N
GXeBjd/v571QW7n1wy4f+OAqHRyK+DBdceWq9i4jekFIXlUKXTSseIoDbWXxWOo=
=G22s
-----END PGP SIGNATURE-----



From crobinso at redhat.com  Tue Jul 17 23:16:58 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Tue, 17 Jul 2012 19:16:58 -0400
Subject: [virt-tools-list] virt-manager slow via SSH - but NOT the
	console
In-Reply-To: <5005F0CA.4050301@felicitus.org>
References: <5005F0CA.4050301@felicitus.org>
Message-ID: <5005F26A.3070602@redhat.com>

On 07/17/2012 07:10 PM, Felicitus wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi,
> 
> I'm using virt-manager via SSH. The docs say that the _initial_
> connect could be slow, which is correct in my case.
> 
> However, anything I do within virt-manager is also slow, for example,
> adding new hardware takes almost a minute to display the extra window,
> and navigating to the "network device" section takes 5 minutes.
> 
> The odd thing is that the VNC console is actually very fast.
> 
> This seems to be very odd, and I really don't know where to start
> debugging. Any hints how to get more debug output from virt-manager?
> 
> Thanks,
> Felicitus

I assume the latency to the remote system is pretty high, ping will confirm.
If that's the case, virt-manager is known to be difficult to use, since all
the libvirt API calls we do behind the scenes need to do round trips which
really slows things down. Unfortunately there really isn't a good solution
that I know of.

- Cole




From eblake at redhat.com  Tue Jul 17 23:25:11 2012
From: eblake at redhat.com (Eric Blake)
Date: Tue, 17 Jul 2012 17:25:11 -0600
Subject: [virt-tools-list] virt-manager slow via SSH - but NOT the
	console
In-Reply-To: <5005F26A.3070602@redhat.com>
References: <5005F0CA.4050301@felicitus.org> <5005F26A.3070602@redhat.com>
Message-ID: <5005F457.9060902@redhat.com>

On 07/17/2012 05:16 PM, Cole Robinson wrote:
> On 07/17/2012 07:10 PM, Felicitus wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Hi,
>>
>> I'm using virt-manager via SSH. The docs say that the _initial_
>> connect could be slow, which is correct in my case.
>>
>> However, anything I do within virt-manager is also slow, for example,
>> adding new hardware takes almost a minute to display the extra window,
>> and navigating to the "network device" section takes 5 minutes.
>>
>> The odd thing is that the VNC console is actually very fast.
>>
>> This seems to be very odd, and I really don't know where to start
>> debugging. Any hints how to get more debug output from virt-manager?
>>
>> Thanks,
>> Felicitus
> 
> I assume the latency to the remote system is pretty high, ping will confirm.
> If that's the case, virt-manager is known to be difficult to use, since all
> the libvirt API calls we do behind the scenes need to do round trips which
> really slows things down. Unfortunately there really isn't a good solution
> that I know of.

There are, however, some improvements being made to reduce some of this
pain.  For example, 0.9.13 added virConnectListAllDomains, which greatly
reduces the number of round trips needed for listing the names of all
domains.  It may be also worth designing an API for batching together
other API into a single RPC call, for example making it possible to
iterate over an array of virDomainPtr in one API call instead of having
to make one RPC call per domain pointer to do the same action on each
domain in the array (such as the action of querying status).

-- 
Eric Blake   eblake at redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 620 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120717/702f1730/attachment.sig>

From felicitus at felicitus.org  Tue Jul 17 23:35:56 2012
From: felicitus at felicitus.org (Felicitus)
Date: Wed, 18 Jul 2012 01:35:56 +0200
Subject: [virt-tools-list] virt-manager slow via SSH - but NOT the
	console
In-Reply-To: <5005F457.9060902@redhat.com>
References: <5005F0CA.4050301@felicitus.org> <5005F26A.3070602@redhat.com>
	<5005F457.9060902@redhat.com>
Message-ID: <5005F6DC.2090703@felicitus.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>> I assume the latency to the remote system is pretty high, ping
>> will confirm. If that's the case, virt-manager is known to be
>> difficult to use, since all the libvirt API calls we do behind
>> the scenes need to do round trips which really slows things down.
>> Unfortunately there really isn't a good solution that I know of.
> 
> There are, however, some improvements being made to reduce some of
> this pain.  For example, 0.9.13 added virConnectListAllDomains,
> which greatly reduces the number of round trips needed for listing
> the names of all domains.  It may be also worth designing an API
> for batching together other API into a single RPC call, for example
> making it possible to iterate over an array of virDomainPtr in one
> API call instead of having to make one RPC call per domain pointer
> to do the same action on each domain in the array (such as the
> action of querying status).
> 

The latency is 80-100ms for an ICMP ping. I have no idea if that's a
low or high value.

I played a bit around and noticed that when I increase the Update
interval from 60 seconds to 30 minutes, the lags are heavily reduced.
Probably getting the stats from the machines took somewhere between 30
seconds and one minute, causing the connection to be full with status
updates.

I also assume that remote calls are somewhat blocking, because when I
attempt to move around the window, it doesn't get re-drawn. Probably
different threads for stats update and management functions could help.

cheers,
Felicitus
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJQBfbcAAoJEJYpjrA42x19vZAH/Rbh1pTjA5lFlmUb75FIBZqv
zMIt3cHMMObhG3lECbljdagtDnSfyQlCPx87pT6yDR/vqmS1Fa2JgVUH4IMk2Rsp
5L2+2TmSSBFP7e3TmQrjw5ZKGI9Xv+p1xf9JFds+9XB1toEoX1ta9Z9QodHZEn0r
bl3YVxkfr2ccRXB6Trbe4w0UF+iy3OWXTaGfq2mTrQuE68bVoEay2tpqyaavRU5A
u9rRJpQv/lMQ+jiWK+kZCxcY9MJE9st+KMBjsoSyrecqSym1xMv2ogoB14LkXKdb
lZtdGSOapTQCuBs0aodLHFM1w4+7nMOjYAY4K6fzGIFB/NOfOxcTSCr3h/XME8w=
=mrPN
-----END PGP SIGNATURE-----



From jhelfman at e-e.com  Tue Jul 17 23:39:21 2012
From: jhelfman at e-e.com (Jason Helfman)
Date: Tue, 17 Jul 2012 16:39:21 -0700
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <20120710214028.GO20543@dormouse.experts-exchange.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
Message-ID: <20120717233921.GA65524@dormouse.experts-exchange.com>

Hi,

I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
this is /usr/local/bin/python. Is there a way I can change this either
through a configure flag at build time, or patch a Makefile?

One more bit. NLS on FreeBSD is installed to /usr/local/share/locale, and
noticed in the po/Makefile:

libdir = ${exec_prefix}/lib
DATADIRNAME = lib
itlocaledir = $(prefix)/$(DATADIRNAME)/locale

I was wondering what flag I could use to get NLS to go to
/usr/local/share/locale and not /usr/local/lib/locale

   --bindir=DIR            user executables [EPREFIX/bin]
   --sbindir=DIR           system admin executables [EPREFIX/sbin]
   --libexecdir=DIR        program executables [EPREFIX/libexec]
   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
   --sharedstatedir=DIR    modifiable architecture-independent data [PREFIX/com]
   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
   --libdir=DIR            object code libraries [EPREFIX/lib]
   --includedir=DIR        C header files [PREFIX/include]
   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
   --datarootdir=DIR       read-only arch.-independent data root [PREFIX/share]
   --datadir=DIR           read-only architecture-independent data [DATAROOTDIR]
   --infodir=DIR           info documentation [DATAROOTDIR/info]
   --localedir=DIR         locale-dependent data [DATAROOTDIR/locale]
   --mandir=DIR            man documentation [DATAROOTDIR/man]
   --docdir=DIR            documentation root [DATAROOTDIR/doc/virt-manager]
   --htmldir=DIR           html documentation [DOCDIR]
   --dvidir=DIR            dvi documentation [DOCDIR]
   --pdfdir=DIR            pdf documentation [DOCDIR]
   --psdir=DIR             ps documentation [DOCDIR]

Thanks,
Jason

-- 
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5



From cfergeau at redhat.com  Wed Jul 18 08:38:46 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 18 Jul 2012 10:38:46 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/25] spice: improve
 fullscreen=auto-conf
In-Reply-To: <1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120718083845.GC2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 09:24:44PM +0200, Marc-Andr? Lureau wrote:
> Do keep client monitor position, do not align monitors automatically.
> The align property is only available since v0.12.101.

I'm still curious about the plans to get this in a released tarball. Any
idea about an ETA?

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/9e6a82e8/attachment.sig>

From cfergeau at redhat.com  Wed Jul 18 08:54:54 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 18 Jul 2012 10:54:54 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 18/25] Add
 VirtViewerDisplay::selectable property
In-Reply-To: <1342553107-27947-19-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-19-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120718085453.GD2162@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 09:25:00PM +0200, Marc-Andr? Lureau wrote:
> This property will be set when the display can be selected to be
> "enabled" and shown (this can involve creating/connecting an
> additional guest monitor, and may need guest agent cooperation for
> example).
> ---
>  src/virt-viewer-display-spice.c |   16 ++++++++++++++++
>  src/virt-viewer-display.c       |   23 +++++++++++++++++++++++
>  src/virt-viewer-display.h       |    2 ++
>  3 files changed, 41 insertions(+)
> 
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index d06e5cf..81eb16f 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -47,6 +47,7 @@ static void virt_viewer_display_spice_send_keys(VirtViewerDisplay *display,
>  static GdkPixbuf *virt_viewer_display_spice_get_pixbuf(VirtViewerDisplay *display);
>  static void virt_viewer_display_spice_release_cursor(VirtViewerDisplay *display);
>  static void virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED);
> +static gboolean virt_viewer_display_spice_selectable(VirtViewerDisplay *display);
>  
>  static void
>  virt_viewer_display_spice_finalize(GObject *obj)
> @@ -70,6 +71,7 @@ virt_viewer_display_spice_class_init(VirtViewerDisplaySpiceClass *klass)
>      dclass->get_pixbuf = virt_viewer_display_spice_get_pixbuf;
>      dclass->release_cursor = virt_viewer_display_spice_release_cursor;
>      dclass->close = virt_viewer_display_spice_close;
> +    dclass->selectable = virt_viewer_display_spice_selectable;
>  
>      g_type_class_add_private(klass, sizeof(VirtViewerDisplaySpicePrivate));
>  }
> @@ -279,6 +281,20 @@ virt_viewer_display_spice_close(VirtViewerDisplay *display G_GNUC_UNUSED)
>  {
>  }
>  
> +static gboolean
> +virt_viewer_display_spice_selectable(VirtViewerDisplay *self)
> +{
> +    gboolean agent_connected;
> +    SpiceMainChannel *mainc;
> +
> +    mainc = get_main(self);
> +    g_object_get(mainc,
> +                 "agent-connected", &agent_connected,
> +                 NULL);
> +
> +    return agent_connected;

Can all agents (older agents, windows agent, ...) handle multiple displays
or is a more specific check needed?


> +}
> +
>  
>  /*
>   * Local variables:
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 18fed19..8d13733 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -80,6 +80,7 @@ enum {
>      PROP_ZOOM_LEVEL,
>      PROP_SHOW_HINT,
>      PROP_SESSION,
> +    PROP_SELECTABLE,
>  };
>  
>  static void
> @@ -167,6 +168,13 @@ virt_viewer_display_class_init(VirtViewerDisplayClass *class)
>                                                          G_PARAM_READWRITE |
>                                                          G_PARAM_CONSTRUCT_ONLY));
>  
> +    g_object_class_install_property(object_class,
> +                                    PROP_SELECTABLE,
> +                                    g_param_spec_boolean("selectable",
> +                                                         "Selectable",
> +                                                         "Selectable",
> +                                                         FALSE,
> +                                                         G_PARAM_READABLE));
>  
>      g_signal_new("display-pointer-grab",
>                   G_OBJECT_CLASS_TYPE(object_class),
> @@ -302,6 +310,9 @@ virt_viewer_display_get_property(GObject *object,
>      case PROP_SESSION:
>          g_value_set_object(value, virt_viewer_display_get_session(display));
>          break;
> +    case PROP_SELECTABLE:
> +        g_value_set_boolean(value, virt_viewer_display_get_selectable(display));
> +        break;
>  
>      default:
>          G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
> @@ -601,6 +612,18 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *self)
>      klass->release_cursor(self);
>  }
>  
> +gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *self)
> +{
> +    VirtViewerDisplayClass *klass;
> +
> +    g_return_val_if_fail(VIRT_VIEWER_IS_DISPLAY(self), FALSE);
> +
> +    klass = VIRT_VIEWER_DISPLAY_GET_CLASS(self);
> +    if (klass->selectable)
> +        return klass->selectable(self);
> +
> +    return TRUE;

Isn't it better to default to FALSE when the method is not overridden?


Christophe

> +}
>  
>  void virt_viewer_display_close(VirtViewerDisplay *self)
>  {
> diff --git a/src/virt-viewer-display.h b/src/virt-viewer-display.h
> index 89b4817..4247570 100644
> --- a/src/virt-viewer-display.h
> +++ b/src/virt-viewer-display.h
> @@ -77,6 +77,7 @@ struct _VirtViewerDisplayClass {
>      void (*release_cursor)(VirtViewerDisplay *display);
>  
>      void (*close)(VirtViewerDisplay *display);
> +    gboolean (*selectable)(VirtViewerDisplay *display);
>  
>      /* signals */
>      void (*display_pointer_grab)(VirtViewerDisplay *display);
> @@ -118,6 +119,7 @@ void virt_viewer_display_release_cursor(VirtViewerDisplay *display);
>  
>  void virt_viewer_display_close(VirtViewerDisplay *display);
>  void virt_viewer_display_set_enabled(VirtViewerDisplay *display, gboolean enabled);
> +gboolean virt_viewer_display_get_selectable(VirtViewerDisplay *display);
>  
>  G_END_DECLS
>  
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/e9066a21/attachment.sig>

From cfergeau at redhat.com  Wed Jul 18 09:38:27 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 18 Jul 2012 11:38:27 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 16/25] Hook up handling of
 Monitors
In-Reply-To: <1342553107-27947-17-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-17-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120718093827.GA14996@teriyaki.cdg.redhat.com>

On Tue, Jul 17, 2012 at 09:24:58PM +0200, Marc-Andr? Lureau wrote:
> Rely on spice-gtk display channel monitors property to manage
> displays. The same display channel may now provide several monitors,
> the SpiceDisplay widget must be told which monitor to display
> ---
>  src/virt-viewer-display-spice.c |   20 +++++----
>  src/virt-viewer-display-spice.h |    2 +-
>  src/virt-viewer-session-spice.c |   89 ++++++++++++++++++++++++++++++++++-----
>  3 files changed, 92 insertions(+), 19 deletions(-)
> 
> diff --git a/src/virt-viewer-display-spice.c b/src/virt-viewer-display-spice.c
> index 53430dd..d06e5cf 100644
> --- a/src/virt-viewer-display-spice.c
> +++ b/src/virt-viewer-display-spice.c
> @@ -172,7 +172,7 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>  {
>      gdouble dw = allocation->width, dh = allocation->height;
>      guint zoom = 100;
> -    guint channelid;
> +    guint nth;
>  
>      if (virt_viewer_display_get_auto_resize(VIRT_VIEWER_DISPLAY(self)) == FALSE)
>          return;
> @@ -187,13 +187,11 @@ virt_viewer_display_spice_size_allocate(VirtViewerDisplaySpice *self,
>          dh /= ((double)zoom / 100.0);
>      }
>  
> -    g_object_get(self->priv->channel, "channel-id", &channelid, NULL);
> +    g_object_get(self, "nth-display", &nth, NULL);
>  
>      SpiceMainChannel *main_channel = virt_viewer_session_spice_get_main_channel(
>          VIRT_VIEWER_SESSION_SPICE(virt_viewer_display_get_session(VIRT_VIEWER_DISPLAY(self))));
> -    spice_main_set_display(main_channel,
> -                           channelid,
> -                           0, 0, dw, dh);
> +    spice_main_set_display(main_channel, nth, 0, 0, dw, dh);
>  }
>  
>  static void
> @@ -212,7 +210,8 @@ enable_accel_changed(VirtViewerApp *app,
>  
>  GtkWidget *
>  virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
> -                              SpiceChannel *channel)
> +                              SpiceChannel *channel,
> +                              gint monitorid)
>  {
>      VirtViewerDisplaySpice *self;
>      VirtViewerApp *app;
> @@ -222,15 +221,20 @@ virt_viewer_display_spice_new(VirtViewerSessionSpice *session,
>      g_return_val_if_fail(SPICE_IS_DISPLAY_CHANNEL(channel), NULL);
>  
>      g_object_get(channel, "channel-id", &channelid, NULL);
> +    // We don't allow monitorid != 0 && channelid != 0
> +    g_return_val_if_fail(channelid == 0 || monitorid == 0, NULL);
>  
>      self = g_object_new(VIRT_VIEWER_TYPE_DISPLAY_SPICE,
>                          "session", session,
> -                        "nth-display", channelid,
> +                        // either monitorid is always 0 or channelid
> +                        // is, we can't have display (0, 2) and (2, 0)
> +                        // for example

I would have put the comment together with the comment above the
g_return_val_if_fail, but that's not important ;)

Christophe

> +                        "nth-display", channelid + monitorid,
>                          NULL);
>      self->priv->channel = channel;
>  
>      g_object_get(session, "spice-session", &s, NULL);
> -    self->priv->display = spice_display_new(s, channelid);
> +    self->priv->display = spice_display_new_with_monitor(s, channelid, monitorid);
>      g_object_unref(s);
>  
>      virt_viewer_signal_connect_object(self->priv->display, "notify::ready",
> diff --git a/src/virt-viewer-display-spice.h b/src/virt-viewer-display-spice.h
> index 701ed85..c2013ec 100644
> --- a/src/virt-viewer-display-spice.h
> +++ b/src/virt-viewer-display-spice.h
> @@ -66,7 +66,7 @@ struct _VirtViewerDisplaySpiceClass {
>  
>  GType virt_viewer_display_spice_get_type(void);
>  
> -GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel);
> +GtkWidget* virt_viewer_display_spice_new(VirtViewerSessionSpice *session, SpiceChannel *channel, gint monitorid);
>  
>  G_END_DECLS
>  
> diff --git a/src/virt-viewer-session-spice.c b/src/virt-viewer-session-spice.c
> index 6577237..a728cb1 100644
> --- a/src/virt-viewer-session-spice.c
> +++ b/src/virt-viewer-session-spice.c
> @@ -33,6 +33,7 @@
>  #include "virt-viewer-session-spice.h"
>  #include "virt-viewer-display-spice.h"
>  #include "virt-viewer-auth.h"
> +#include "virt-glib-compat.h"
>  
>  #if !GLIB_CHECK_VERSION(2, 26, 0)
>  #include "gbinding.h"
> @@ -414,6 +415,78 @@ agent_connected_changed(SpiceChannel *cmain,
>  }
>  
>  static void
> +destroy_display(gpointer data)
> +{
> +    VirtViewerDisplay *display = VIRT_VIEWER_DISPLAY(data);
> +    VirtViewerSession *session = virt_viewer_display_get_session(display);
> +
> +    DEBUG_LOG("Destroying spice display %p", display);
> +    virt_viewer_session_remove_display(session, display);
> +    g_object_unref(display);
> +}
> +
> +static void
> +virt_viewer_session_spice_display_monitors(SpiceChannel *channel,
> +                                           GParamSpec *pspec G_GNUC_UNUSED,
> +                                           VirtViewerSessionSpice *self)
> +{
> +    GArray *monitors = NULL;
> +    GPtrArray *displays = NULL;
> +    GtkWidget *display;
> +    guint i, monitors_max;
> +
> +    g_object_get(channel,
> +                 "monitors", &monitors,
> +                 "monitors-max", &monitors_max,
> +                 NULL);
> +    g_return_if_fail(monitors != NULL);
> +    g_return_if_fail(monitors->len <= monitors_max);
> +
> +    displays = g_object_get_data(G_OBJECT(channel), "virt-viewer-displays");
> +    if (displays == NULL) {
> +        displays = g_ptr_array_new();
> +        g_ptr_array_set_free_func(displays, destroy_display);
> +        g_object_set_data_full(G_OBJECT(channel), "virt-viewer-displays",
> +                               displays, (GDestroyNotify)g_ptr_array_unref);
> +    }
> +
> +    g_ptr_array_set_size(displays, monitors_max);
> +
> +    for (i = 0; i < monitors_max; i++) {
> +        display = g_ptr_array_index(displays, i);
> +        if (display == NULL) {
> +            display = virt_viewer_display_spice_new(self, channel, i);
> +            DEBUG_LOG("creating spice display (#:%d)", i);
> +            g_ptr_array_index(displays, i) = g_object_ref(display);
> +        }
> +
> +        g_object_freeze_notify(G_OBJECT(display));
> +        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), FALSE);
> +        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(self),
> +                                        VIRT_VIEWER_DISPLAY(display));
> +    }
> +
> +    for (i = 0; i < monitors->len; i++) {
> +        SpiceDisplayMonitorConfig *monitor = &g_array_index(monitors, SpiceDisplayMonitorConfig, i);
> +        display = g_ptr_array_index(displays, monitor->id);
> +        g_return_if_fail(display != NULL);
> +
> +        if (monitor->width == 0 || monitor->width == 0)
> +            continue;
> +
> +        virt_viewer_display_set_enabled(VIRT_VIEWER_DISPLAY(display), TRUE);
> +        virt_viewer_display_set_desktop_size(VIRT_VIEWER_DISPLAY(display),
> +                                             monitor->width, monitor->height);
> +    }
> +
> +    for (i = 0; i < monitors_max; i++)
> +        g_object_thaw_notify(g_ptr_array_index(displays, i));
> +
> +    g_clear_pointer(&monitors, g_array_unref);
> +
> +}
> +
> +static void
>  virt_viewer_session_spice_channel_new(SpiceSession *s,
>                                        SpiceChannel *channel,
>                                        VirtViewerSession *session)
> @@ -441,20 +514,17 @@ virt_viewer_session_spice_channel_new(SpiceSession *s,
>  
>          g_signal_connect(channel, "notify::agent-connected", G_CALLBACK(agent_connected_changed),  self);
>          agent_connected_changed(channel, NULL, self);
> +
> +        g_signal_emit_by_name(session, "session-connected");
>      }
>  
>      if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
> -        GtkWidget *display;
> -
> -        g_signal_emit_by_name(session, "session-connected");
> +        g_signal_emit_by_name(session, "session-initialized");
>  
> -        DEBUG_LOG("new display channel (#%d)", id);
> -        display = virt_viewer_display_spice_new(self, channel);
> -        g_object_set_data(G_OBJECT(channel), "virt-viewer-display", display);
> -        virt_viewer_session_add_display(VIRT_VIEWER_SESSION(session),
> -                                        VIRT_VIEWER_DISPLAY(display));
> +        g_signal_connect(channel, "notify::monitors",
> +                         G_CALLBACK(virt_viewer_session_spice_display_monitors), self);
>  
> -        g_signal_emit_by_name(session, "session-initialized");
> +        spice_channel_connect(channel);
>      }
>  
>      if (SPICE_IS_INPUTS_CHANNEL(channel)) {
> @@ -538,7 +608,6 @@ virt_viewer_session_spice_channel_destroy(G_GNUC_UNUSED SpiceSession *s,
>      if (SPICE_IS_DISPLAY_CHANNEL(channel)) {
>          VirtViewerDisplay *display = g_object_get_data(G_OBJECT(channel), "virt-viewer-display");
>          DEBUG_LOG("zap display channel (#%d, %p)", id, display);
> -        virt_viewer_session_remove_display(session, display);
>      }
>  
>      if (SPICE_IS_PLAYBACK_CHANNEL(channel) && self->priv->audio) {
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/91c65fb7/attachment.sig>

From cfergeau at redhat.com  Wed Jul 18 09:40:27 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 18 Jul 2012 11:40:27 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 00/25] Add new Spice
 monitors handling support v2
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120718094027.GB14996@teriyaki.cdg.redhat.com>

Series look good to me apart from the few patches on which I've commented.
For what it's worth, the READY/DISCONNECTED changes feel a bit awkward, but
I don't have better suggestions so they should be good enough ;)
Similarly, I'm not a big fan of "Prevent from adding the same display
several time in the session"

ACK series with the few comments I made addressed.

Christophe

On Tue, Jul 17, 2012 at 09:24:42PM +0200, Marc-Andr? Lureau wrote:
> Hi,
> 
> Spice now supports multiple arbitrary-sized monitors per display
> channel. The client has to learn how to handle the new events in order
> to create, to show, and to manage the displays appropriately.
> 
> The feature is currently being merged in the various projects, to test
> it you'll need the following projects update with monitor config
> support: spice-server, qemu, spice-gtk, qxl driver and vdagent.
> 
> Changes since v1, after Christophe review:
>  - splitted into more smaller patches
>  - bumped to glib 2.22 and removed fallback code
>  - added g_clear_pointer() virt-glib-compat.h fallback
>  - renamed s/cmain/main_channel, s/c/monitor, s/enabled/enable
>  - removed the display->dirty idle flag change, it was borked
>  - more comments and spelling fixes
>  - monitor count/max range is now checked by spice-gtk
> 
> cheers
> 
> 
> Marc-Andr? Lureau (25):
>   build: allow building with newer glibc-headers and -O0
>   spice: improve fullscreen=auto-conf
>   spice: disconnect signal handlers when either object is destroyed
>   build: generate enums type boilerplate
>   Destroy dialog immediately after run
>   Turn display:show-hint into flags type
>   Run-time check values before doing bad computation
>   Mark a menu string as translatable
>   Rebuild display menu when a window is added or removed
>   Use virt_viewer_connect_object() for display
>   Add a DISABLED display hint
>   Prevent from adding the same display several time in the session
>   Number display starting from 1
>   Use SpiceDisplay:ready property instead of channel mark
>   Bump glib > 2.22, add compat file
>   Hook up handling of Monitors
>   Add virt_viewer_window_get_display()
>   Add VirtViewerDisplay::selectable property
>   spice: factor out main channel lookup code
>   Make the display submenu insensitive if display can't be selected
>   Add VirtViewerSession::session-display-updated
>   Simplify display flag handling
>   vnc: when session is disconnected, make the display as non-ready
>   Make status widget visible immediately
>   Show status page by default if display not ready
> 
>  acinclude.m4                      |    9 ++-
>  configure.ac                      |    7 ++-
>  src/Makefile.am                   |   24 +++++++-
>  src/virt-glib-compat.h            |   56 +++++++++++++++++
>  src/virt-viewer-app.c             |   53 ++++++++++++----
>  src/virt-viewer-display-spice.c   |  123 ++++++++++++++++++++++++-------------
>  src/virt-viewer-display-spice.h   |    2 +-
>  src/virt-viewer-display.c         |   70 +++++++++++++++++----
>  src/virt-viewer-display.h         |   15 +++--
>  src/virt-viewer-enums.c.etemplate |   60 ++++++++++++++++++
>  src/virt-viewer-enums.h.etemplate |   41 +++++++++++++
>  src/virt-viewer-notebook.c        |    1 +
>  src/virt-viewer-session-spice.c   |  111 ++++++++++++++++++++++++++++-----
>  src/virt-viewer-session-vnc.c     |    8 +--
>  src/virt-viewer-session.c         |   13 ++++
>  src/virt-viewer-session.h         |    1 +
>  src/virt-viewer-util.c            |  117 +++++++++++++++++++++++++++++++++++
>  src/virt-viewer-util.h            |    6 ++
>  src/virt-viewer-window.c          |   67 +++++++++++++++-----
>  src/virt-viewer-window.h          |    1 +
>  20 files changed, 672 insertions(+), 113 deletions(-)
>  create mode 100644 src/virt-glib-compat.h
>  create mode 100644 src/virt-viewer-enums.c.etemplate
>  create mode 100644 src/virt-viewer-enums.h.etemplate
> 
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/8880b6ce/attachment.sig>

From marcandre.lureau at gmail.com  Wed Jul 18 09:48:53 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Wed, 18 Jul 2012 11:48:53 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/25] spice: improve
	fullscreen=auto-conf
In-Reply-To: <20120718083845.GC2162@teriyaki.cdg.redhat.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>
	<20120718083845.GC2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CK7mGf8p85cVA6QDw_s8X96U75mCU_X4jqPUQoJO1f9sQ@mail.gmail.com>

Hi

On Wed, Jul 18, 2012 at 10:38 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> On Tue, Jul 17, 2012 at 09:24:44PM +0200, Marc-Andr? Lureau wrote:
>> Do keep client monitor position, do not align monitors automatically.
>> The align property is only available since v0.12.101.
>
> I'm still curious about the plans to get this in a released tarball. Any
> idea about an ETA?

http://spice-space.org/download/gtk/spice-gtk-0.12.101.tar.bz2


-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Wed Jul 18 09:54:07 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Wed, 18 Jul 2012 11:54:07 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 18/25] Add
 VirtViewerDisplay::selectable property
In-Reply-To: <20120718085453.GD2162@teriyaki.cdg.redhat.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-19-git-send-email-marcandre.lureau@gmail.com>
	<20120718085453.GD2162@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CKno_qw93Ef_owjGuOcjSS-ptC1qAiaUQJuMQYhN-nieQ@mail.gmail.com>

Hi

On Wed, Jul 18, 2012 at 10:54 AM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> Can all agents (older agents, windows agent, ...) handle multiple displays
> or is a more specific check needed?

Good point, and indeed not. We will want to check the agent
capability. I'll see if I can update the patch.

> Isn't it better to default to FALSE when the method is not overridden?

I would prefer keeping the current behaviour in this case, so yes
default to TRUE until something else is needed.

-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Wed Jul 18 10:18:40 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 18 Jul 2012 12:18:40 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/25] spice: improve
 fullscreen=auto-conf
In-Reply-To: <CAJ+F1CK7mGf8p85cVA6QDw_s8X96U75mCU_X4jqPUQoJO1f9sQ@mail.gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>
	<20120718083845.GC2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CK7mGf8p85cVA6QDw_s8X96U75mCU_X4jqPUQoJO1f9sQ@mail.gmail.com>
Message-ID: <20120718101839.GC14996@teriyaki.cdg.redhat.com>

On Wed, Jul 18, 2012 at 11:48:53AM +0200, Marc-Andr? Lureau wrote:
> On Wed, Jul 18, 2012 at 10:38 AM, Christophe Fergeau
> <cfergeau at redhat.com> wrote:
> > On Tue, Jul 17, 2012 at 09:24:44PM +0200, Marc-Andr? Lureau wrote:
> >> Do keep client monitor position, do not align monitors automatically.
> >> The align property is only available since v0.12.101.
> >
> > I'm still curious about the plans to get this in a released tarball. Any
> > idea about an ETA?
> 
> http://spice-space.org/download/gtk/spice-gtk-0.12.101.tar.bz2

Can we please stop having this discussion every time? If it's not tagged in
git, it's just a random git snapshot, not a proper release. I don't see
a tag in http://cgit.freedesktop.org/spice/spice-gtk/ , so I'm assuming
this is just a snapshot with no API/ABI guarantees for the newly added
stuff.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/662540a1/attachment.sig>

From morgancoxuk at gmail.com  Wed Jul 18 10:22:00 2012
From: morgancoxuk at gmail.com (Morgan Cox)
Date: Wed, 18 Jul 2012 11:22:00 +0100
Subject: [virt-tools-list] [libvirt-users] I can't virt-clone into an
 existing LVM now (ERROR: clone onto existing storage volume is not
 supported:) - i could in previous lbvirt versions? (deployment scripts no
 longer work...)
In-Reply-To: <20120712220215.GA11400@amd.home.annexia.org>
References: <CAGfHYAS43gT4ZsrOrC5O5VSvXaNJrExriwOzynPZeXiLJbFgDg@mail.gmail.com>
	<4FFD8FE7.8000104@redhat.com>
	<20120712220215.GA11400@amd.home.annexia.org>
Message-ID: <CAGfHYAQDynnwJTH5j0iBhz3UPXUdhkijGquOo6FpKpF1Beea3A@mail.gmail.com>

Hi

I was under the impression that virt-clone would create new UUId,s etc
where as virt-resize doesn't ?

All I am trying to do is use a VM (in LVM) as a template to use to clone .

The templates are only about 3GB - so to deploy the template I need to
resize them.

This is why i did this

TEMPLATE -> [virt-clone] -> new_vm.tmp  (to create new UUIDs , etc)

new_vm.tmp -> [virt-clone] -> new_vm (a clone of the vm with new uuids
- this creates .xml for qemu, etc, correct name . etc)

new_vm.tmp -> [virt-resize] -> new_vm  - new vm has correct partition size.

If there is a more sensible way of doing this I would like to know !

Anyway is there a way of forcing being able to copy to an existing LVM
partition ??  As mentioned I could do it fine in Ubuntu 10.04 not
12.04 - i.e do I need to recompile libvirt, etc ?

Any help will be welcomed



From rjones at redhat.com  Wed Jul 18 10:35:06 2012
From: rjones at redhat.com (Richard W.M. Jones)
Date: Wed, 18 Jul 2012 11:35:06 +0100
Subject: [virt-tools-list] [libvirt-users] I can't virt-clone into an
 existing LVM now (ERROR: clone onto existing storage volume is not
 supported:) - i could in previous lbvirt versions? (deployment scripts no
 longer work...)
In-Reply-To: <CAGfHYAQDynnwJTH5j0iBhz3UPXUdhkijGquOo6FpKpF1Beea3A@mail.gmail.com>
References: <CAGfHYAS43gT4ZsrOrC5O5VSvXaNJrExriwOzynPZeXiLJbFgDg@mail.gmail.com>
	<4FFD8FE7.8000104@redhat.com>
	<20120712220215.GA11400@amd.home.annexia.org>
	<CAGfHYAQDynnwJTH5j0iBhz3UPXUdhkijGquOo6FpKpF1Beea3A@mail.gmail.com>
Message-ID: <20120718103506.GP2701@amd.home.annexia.org>

On Wed, Jul 18, 2012 at 11:22:00AM +0100, Morgan Cox wrote:
> Hi
> 
> I was under the impression that virt-clone would create new UUId,s etc
> where as virt-resize doesn't ?

In my previous message I said:

  "[...] plus it [virt-clone] makes a handful of changes to the
  libvirt config.  You can do most of this by alternate means."

> All I am trying to do is use a VM (in LVM) as a template to use to clone .
> 
> The templates are only about 3GB - so to deploy the template I need to
> resize them.
> 
> This is why i did this
> 
> TEMPLATE -> [virt-clone] -> new_vm.tmp  (to create new UUIDs , etc)

If virt-clone doesn't work then you should just copy and resize the
disk image using one step of virt-resize, create a new libvirt XML
configuration, and finally 'virsh define' the new guest.  This would
involve just a single copy and is about the most efficient way of
doing it (even using qcow2 or LUN clones / FlexClone is unlikely to be
better).

> If there is a more sensible way of doing this I would like to know !

See above.

> Anyway is there a way of forcing being able to copy to an existing LVM
> partition ??  As mentioned I could do it fine in Ubuntu 10.04 not
> 12.04 - i.e do I need to recompile libvirt, etc ?

$ virt-resize /dev/vg/lv_template /dev/vg/lv_new

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
virt-p2v converts physical machines to virtual machines.  Boot with a
live CD or over the network (PXE) and turn machines into Xen guests.
http://et.redhat.com/~rjones/virt-p2v



From rjones at redhat.com  Wed Jul 18 10:40:43 2012
From: rjones at redhat.com (Richard W.M. Jones)
Date: Wed, 18 Jul 2012 11:40:43 +0100
Subject: [virt-tools-list] virt-manager slow via SSH - but NOT the
	console
In-Reply-To: <5005F6DC.2090703@felicitus.org>
References: <5005F0CA.4050301@felicitus.org> <5005F26A.3070602@redhat.com>
	<5005F457.9060902@redhat.com> <5005F6DC.2090703@felicitus.org>
Message-ID: <20120718104043.GA29650@amd.home.annexia.org>

On Wed, Jul 18, 2012 at 01:35:56AM +0200, Felicitus wrote:
> The latency is 80-100ms for an ICMP ping. I have no idea if that's a
> low or high value.

Very high.

For LANs you should see < 1ms.  For example, I get ~ 0.2 ms between
two machines connected directly through an ethernet switch.

Even for two machines located on opposite sides of London I'm getting
ping times under 30ms.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine.  Supports Linux and Windows.
http://et.redhat.com/~rjones/virt-df/



From marcandre.lureau at gmail.com  Wed Jul 18 10:47:00 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Wed, 18 Jul 2012 12:47:00 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/25] spice: improve
	fullscreen=auto-conf
In-Reply-To: <20120718101839.GC14996@teriyaki.cdg.redhat.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>
	<20120718083845.GC2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CK7mGf8p85cVA6QDw_s8X96U75mCU_X4jqPUQoJO1f9sQ@mail.gmail.com>
	<20120718101839.GC14996@teriyaki.cdg.redhat.com>
Message-ID: <CAJ+F1CJn4Kxdp_HepfTr2kbwU9i8c+ZRZ_m_sX7fXeHHN77-8Q@mail.gmail.com>

Hi

On Wed, Jul 18, 2012 at 12:18 PM, Christophe Fergeau
<cfergeau at redhat.com> wrote:
> On Wed, Jul 18, 2012 at 11:48:53AM +0200, Marc-Andr? Lureau wrote:
>> On Wed, Jul 18, 2012 at 10:38 AM, Christophe Fergeau
>> <cfergeau at redhat.com> wrote:
>> > On Tue, Jul 17, 2012 at 09:24:44PM +0200, Marc-Andr? Lureau wrote:
>> >> Do keep client monitor position, do not align monitors automatically.
>> >> The align property is only available since v0.12.101.
>> >
>> > I'm still curious about the plans to get this in a released tarball. Any
>> > idea about an ETA?
>>
>> http://spice-space.org/download/gtk/spice-gtk-0.12.101.tar.bz2
>
> Can we please stop having this discussion every time? If it's not tagged in
> git, it's just a random git snapshot, not a proper release. I don't see
> a tag in http://cgit.freedesktop.org/spice/spice-gtk/ , so I'm assuming
> this is just a snapshot with no API/ABI guarantees for the newly added
> stuff.


We are soon to release 0.13, we just need a bit more testing.

In the very unlikely event of having API issues discovered, we may
break. In this sense, it just work like any glib/gtk/or gnome library
release during it's unstable period.


-- 
Marc-Andr? Lureau



From cfergeau at redhat.com  Wed Jul 18 12:09:18 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 18 Jul 2012 14:09:18 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 02/25] spice: improve
 fullscreen=auto-conf
In-Reply-To: <CAJ+F1CJn4Kxdp_HepfTr2kbwU9i8c+ZRZ_m_sX7fXeHHN77-8Q@mail.gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-3-git-send-email-marcandre.lureau@gmail.com>
	<20120718083845.GC2162@teriyaki.cdg.redhat.com>
	<CAJ+F1CK7mGf8p85cVA6QDw_s8X96U75mCU_X4jqPUQoJO1f9sQ@mail.gmail.com>
	<20120718101839.GC14996@teriyaki.cdg.redhat.com>
	<CAJ+F1CJn4Kxdp_HepfTr2kbwU9i8c+ZRZ_m_sX7fXeHHN77-8Q@mail.gmail.com>
Message-ID: <20120718120918.GD14996@teriyaki.cdg.redhat.com>

On Wed, Jul 18, 2012 at 12:47:00PM +0200, Marc-Andr? Lureau wrote:
> 
> We are soon to release 0.13, we just need a bit more testing.

Ok, thanks, this is all that I was asking, I wasn't necessarily requesting
a tarball...

> In the very unlikely event of having API issues discovered, we may
> break. In this sense, it just work like any glib/gtk/or gnome library
> release during it's unstable period.

Yes, except that they are actual releases, they are announced as unstable
releases, and the rules are clear, they are not git snapshots. I don't have
anything against such a rule, I have something against pretending these git
snapshots are 'releases'.

Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/398ff505/attachment.sig>

From pheriko.support at gmail.com  Wed Jul 18 14:33:15 2012
From: pheriko.support at gmail.com (Periko Support)
Date: Wed, 18 Jul 2012 07:33:15 -0700
Subject: [virt-tools-list] virt-install options, are important?
Message-ID: <CAK2yrTYuj+wcb_9x7YWRB7SAWg36cvgne9b7ybyxg1-CPjUV+w@mail.gmail.com>

Hi.

 This is related to hvm.

 I have read virt-install man page, I had read that we have a lot of
flags, we can be very specific on how we build a hvm for win 2k3, 3k3,
etc.

 Now, is important to be very specific on what we want to build or
virt-install is smart enough to detect our vm OS we are building?

 Running this:

virt-install -n win8-1 -r 2048 -f /opt/win8/win8-disk.img -b eth0
--vnc --vnclisten=192.168.50.2 -v -c /opt/win8/win8.iso

or

virt-install -n win8-2 -r 2048 --hvm --vcpus=2 --cpuset=2-3 --disk
path=/opt/win8-2/win8-disk.img --os-type=windows --os-variant=win2k8
-b bridge0 --noautoconsole --vnc --vnclisten=192.168.50.107 -v --cdrom
/dev/dvdrw

is important?

Exist a way to know how we build our vm? or is not important?

Does the options we give to virt-install affect our vm daily
operation(Performance)?

Centos 6.

Thanks!!!



From crobinso at redhat.com  Wed Jul 18 16:56:10 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Wed, 18 Jul 2012 12:56:10 -0400
Subject: [virt-tools-list] virt-install options, are important?
In-Reply-To: <CAK2yrTYuj+wcb_9x7YWRB7SAWg36cvgne9b7ybyxg1-CPjUV+w@mail.gmail.com>
References: <CAK2yrTYuj+wcb_9x7YWRB7SAWg36cvgne9b7ybyxg1-CPjUV+w@mail.gmail.com>
Message-ID: <5006EAAA.4070000@redhat.com>

On 07/18/2012 10:33 AM, Periko Support wrote:
> Hi.
> 
>  This is related to hvm.
> 
>  I have read virt-install man page, I had read that we have a lot of
> flags, we can be very specific on how we build a hvm for win 2k3, 3k3,
> etc.
> 
>  Now, is important to be very specific on what we want to build or
> virt-install is smart enough to detect our vm OS we are building?
> 
>  Running this:
> 
> virt-install -n win8-1 -r 2048 -f /opt/win8/win8-disk.img -b eth0
> --vnc --vnclisten=192.168.50.2 -v -c /opt/win8/win8.iso
> 
> or
> 
> virt-install -n win8-2 -r 2048 --hvm --vcpus=2 --cpuset=2-3 --disk
> path=/opt/win8-2/win8-disk.img --os-type=windows --os-variant=win2k8
> -b bridge0 --noautoconsole --vnc --vnclisten=192.168.50.107 -v --cdrom
> /dev/dvdrw
> 
> is important?
> 
> Exist a way to know how we build our vm? or is not important?
> 
> Does the options we give to virt-install affect our vm daily
> operation(Performance)?
> 

Yes, you always want to specify --os-variant if there is an appropriate value
(--os-type isn't really required). It is used for setting up the most
performant defaults. We do some detection if you are installing with
--location, but not --cdrom. libosinfo provides that and we should hopefully
be switching over soon.

- Cole



From crobinso at redhat.com  Wed Jul 18 17:17:29 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Wed, 18 Jul 2012 13:17:29 -0400
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <20120717233921.GA65524@dormouse.experts-exchange.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
Message-ID: <5006EFA9.7020009@redhat.com>

On 07/17/2012 07:39 PM, Jason Helfman wrote:
> Hi,
> 
> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
> this is /usr/local/bin/python. Is there a way I can change this either
> through a configure flag at build time, or patch a Makefile?
> 

Not at the moment.

Hmm, the system python binary is under /usr/local? That doesn't seem right,
/usr/local is supposed to be where things go when a user installs things by
hand. For example none of the software the Fedora distributes put anything in
/usr/local, but by default ./configure && make install for most software roots
things under /usr/local.

If freebsd is really doing that, it would probably require some makefile and
configure tweaking to work on both platforms. Are you just trying to get
virt-manager working personally, or packaging it (whatever that entails) to be
distributed with freebsd?

> One more bit. NLS on FreeBSD is installed to /usr/local/share/locale, and
> noticed in the po/Makefile:
> 
> libdir = ${exec_prefix}/lib
> DATADIRNAME = lib
> itlocaledir = $(prefix)/$(DATADIRNAME)/locale
> 
> I was wondering what flag I could use to get NLS to go to
> /usr/local/share/locale and not /usr/local/lib/locale
> 
>   --bindir=DIR            user executables [EPREFIX/bin]
>   --sbindir=DIR           system admin executables [EPREFIX/sbin]
>   --libexecdir=DIR        program executables [EPREFIX/libexec]
>   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
>   --sharedstatedir=DIR    modifiable architecture-independent data [PREFIX/com]
>   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
>   --libdir=DIR            object code libraries [EPREFIX/lib]
>   --includedir=DIR        C header files [PREFIX/include]
>   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
>   --datarootdir=DIR       read-only arch.-independent data root [PREFIX/share]
>   --datadir=DIR           read-only architecture-independent data [DATAROOTDIR]
>   --infodir=DIR           info documentation [DATAROOTDIR/info]
>   --localedir=DIR         locale-dependent data [DATAROOTDIR/locale]
>   --mandir=DIR            man documentation [DATAROOTDIR/man]
>   --docdir=DIR            documentation root [DATAROOTDIR/doc/virt-manager]
>   --htmldir=DIR           html documentation [DOCDIR]
>   --dvidir=DIR            dvi documentation [DOCDIR]
>   --pdfdir=DIR            pdf documentation [DOCDIR]
>   --psdir=DIR             ps documentation [DOCDIR]
>

I'm not entirely sure I follow, but I'd try playing around with --libdir or
--datadir.

- Cole

> Thanks,
> Jason
> 




From eblake at redhat.com  Wed Jul 18 17:26:13 2012
From: eblake at redhat.com (Eric Blake)
Date: Wed, 18 Jul 2012 11:26:13 -0600
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <5006EFA9.7020009@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com>
Message-ID: <5006F1B5.5050601@redhat.com>

On 07/18/2012 11:17 AM, Cole Robinson wrote:
> On 07/17/2012 07:39 PM, Jason Helfman wrote:
>> Hi,
>>
>> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
>> this is /usr/local/bin/python. Is there a way I can change this either
>> through a configure flag at build time, or patch a Makefile?
>>
> 
> Not at the moment.

I think you're asking for something like what libvirt did in commit
e4384459c, by using:

#!/usr/bin/env python

instead of

#!/usr/bin/python

> Hmm, the system python binary is under /usr/local? That doesn't seem right,
> /usr/local is supposed to be where things go when a user installs things by
> hand. For example none of the software the Fedora distributes put anything in
> /usr/local, but by default ./configure && make install for most software roots
> things under /usr/local.

Yeah, that sounds like it is not FreeBSD's default python installation
location, but your hand-installed version, or else a bug in how FreeBSD
packages things.  At any rate, using /usr/bin/env instead of hard-coded
paths would work around that just as easily as it will for python being
installed anywhere else other than /usr/bin/.

-- 
Eric Blake   eblake at redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 620 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120718/8f557ba7/attachment.sig>

From jhelfman at e-e.com  Wed Jul 18 17:37:02 2012
From: jhelfman at e-e.com (Jason Helfman)
Date: Wed, 18 Jul 2012 10:37:02 -0700
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <5006F1B5.5050601@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
Message-ID: <20120718173702.GC76254@dormouse.experts-exchange.com>

On Wed, Jul 18, 2012 at 11:26:13AM -0600, Eric Blake thus spake:
>On 07/18/2012 11:17 AM, Cole Robinson wrote:
>> On 07/17/2012 07:39 PM, Jason Helfman wrote:
>>> Hi,
>>>
>>> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
>>> this is /usr/local/bin/python. Is there a way I can change this either
>>> through a configure flag at build time, or patch a Makefile?
>>>
>>
>> Not at the moment.

Oy. Can this be configurable in the next release?
>
>I think you're asking for something like what libvirt did in commit
>e4384459c, by using:
>
>#!/usr/bin/env python
>
>instead of
>
>#!/usr/bin/python
>
>> Hmm, the system python binary is under /usr/local? That doesn't seem right,
>> /usr/local is supposed to be where things go when a user installs things by
>> hand. For example none of the software the Fedora distributes put anything in
>> /usr/local, but by default ./configure && make install for most software roots
>> things under /usr/local.
>
>Yeah, that sounds like it is not FreeBSD's default python installation
>location, but your hand-installed version, or else a bug in how FreeBSD
>packages things.  At any rate, using /usr/bin/env instead of hard-coded
>paths would work around that just as easily as it will for python being
>installed anywhere else other than /usr/bin/.
>

Default for FreeBSD is /usr/local for all portstree software, because this
software isn't part of the base operating system, so it can't go in the
base distribution.

Also, do you have any ideas for the NLS location?

Thanks!
-jgh

>-- 
>Eric Blake   eblake at redhat.com    +1-919-301-3266
>Libvirt virtualization library http://libvirt.org
>

-- 
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5



From jhelfman at e-e.com  Wed Jul 18 17:50:20 2012
From: jhelfman at e-e.com (Jason Helfman)
Date: Wed, 18 Jul 2012 10:50:20 -0700
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <5006EFA9.7020009@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com>
Message-ID: <20120718175020.GE76254@dormouse.experts-exchange.com>

On Wed, Jul 18, 2012 at 01:17:29PM -0400, Cole Robinson thus spake:
>On 07/17/2012 07:39 PM, Jason Helfman wrote:
>> Hi,
>>
>> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
>> this is /usr/local/bin/python. Is there a way I can change this either
>> through a configure flag at build time, or patch a Makefile?
>>
>
>Not at the moment.

Doh!

>
>Hmm, the system python binary is under /usr/local? That doesn't seem right,
>/usr/local is supposed to be where things go when a user installs things by
>hand. For example none of the software the Fedora distributes put anything in
>/usr/local, but by default ./configure && make install for most software roots
>things under /usr/local.
>
>If freebsd is really doing that, it would probably require some makefile and
>configure tweaking to work on both platforms. Are you just trying to get
>virt-manager working personally, or packaging it (whatever that entails) to be
>distributed with freebsd?

Both. The default for FreeBSD is to install to /usr/local, as no software in
our ports collection is part of our base operating system. Having
configurability would be optimal.
>
>> One more bit. NLS on FreeBSD is installed to /usr/local/share/locale, and
>> noticed in the po/Makefile:
>>
>> libdir = ${exec_prefix}/lib
>> DATADIRNAME = lib
>> itlocaledir = $(prefix)/$(DATADIRNAME)/locale
>>
>> I was wondering what flag I could use to get NLS to go to
>> /usr/local/share/locale and not /usr/local/lib/locale
>>
>>   --bindir=DIR            user executables [EPREFIX/bin]
>>   --sbindir=DIR           system admin executables [EPREFIX/sbin]
>>   --libexecdir=DIR        program executables [EPREFIX/libexec]
>>   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
>>   --sharedstatedir=DIR    modifiable architecture-independent data [PREFIX/com]
>>   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
>>   --libdir=DIR            object code libraries [EPREFIX/lib]
>>   --includedir=DIR        C header files [PREFIX/include]
>>   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
>>   --datarootdir=DIR       read-only arch.-independent data root [PREFIX/share]
>>   --datadir=DIR           read-only architecture-independent data [DATAROOTDIR]
>>   --infodir=DIR           info documentation [DATAROOTDIR/info]
>>   --localedir=DIR         locale-dependent data [DATAROOTDIR/locale]
>>   --mandir=DIR            man documentation [DATAROOTDIR/man]
>>   --docdir=DIR            documentation root [DATAROOTDIR/doc/virt-manager]
>>   --htmldir=DIR           html documentation [DOCDIR]
>>   --dvidir=DIR            dvi documentation [DOCDIR]
>>   --pdfdir=DIR            pdf documentation [DOCDIR]
>>   --psdir=DIR             ps documentation [DOCDIR]
>>
>
>I'm not entirely sure I follow, but I'd try playing around with --libdir or
>--datadir.

I figured I would do that, but would this change the installation path of
anything else?

Thanks!

-- 
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5



From crobinso at redhat.com  Wed Jul 18 18:03:00 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Wed, 18 Jul 2012 14:03:00 -0400
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <20120718173702.GC76254@dormouse.experts-exchange.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
Message-ID: <5006FA54.5060306@redhat.com>

On 07/18/2012 01:37 PM, Jason Helfman wrote:
> On Wed, Jul 18, 2012 at 11:26:13AM -0600, Eric Blake thus spake:
>> On 07/18/2012 11:17 AM, Cole Robinson wrote:
>>> On 07/17/2012 07:39 PM, Jason Helfman wrote:
>>>> Hi,
>>>>
>>>> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
>>>> this is /usr/local/bin/python. Is there a way I can change this either
>>>> through a configure flag at build time, or patch a Makefile?
>>>>
>>>
>>> Not at the moment.
> 
> Oy. Can this be configurable in the next release?

I'll take a patch for the /usr/bin/env change like Eric suggested.

- Cole




From crobinso at redhat.com  Wed Jul 18 18:04:55 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Wed, 18 Jul 2012 14:04:55 -0400
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <20120718175020.GE76254@dormouse.experts-exchange.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com>
	<20120718175020.GE76254@dormouse.experts-exchange.com>
Message-ID: <5006FAC7.8000707@redhat.com>

On 07/18/2012 01:50 PM, Jason Helfman wrote:
> On Wed, Jul 18, 2012 at 01:17:29PM -0400, Cole Robinson thus spake:
>> On 07/17/2012 07:39 PM, Jason Helfman wrote:
>>> Hi,
>>>
>>> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
>>> this is /usr/local/bin/python. Is there a way I can change this either
>>> through a configure flag at build time, or patch a Makefile?
>>>
>>
>> Not at the moment.
> 
> Doh!
> 
>>
>> Hmm, the system python binary is under /usr/local? That doesn't seem right,
>> /usr/local is supposed to be where things go when a user installs things by
>> hand. For example none of the software the Fedora distributes put anything in
>> /usr/local, but by default ./configure && make install for most software roots
>> things under /usr/local.
>>
>> If freebsd is really doing that, it would probably require some makefile and
>> configure tweaking to work on both platforms. Are you just trying to get
>> virt-manager working personally, or packaging it (whatever that entails) to be
>> distributed with freebsd?
> 
> Both. The default for FreeBSD is to install to /usr/local, as no software in
> our ports collection is part of our base operating system. Having
> configurability would be optimal.
>>
>>> One more bit. NLS on FreeBSD is installed to /usr/local/share/locale, and
>>> noticed in the po/Makefile:
>>>
>>> libdir = ${exec_prefix}/lib
>>> DATADIRNAME = lib
>>> itlocaledir = $(prefix)/$(DATADIRNAME)/locale
>>>
>>> I was wondering what flag I could use to get NLS to go to
>>> /usr/local/share/locale and not /usr/local/lib/locale
>>>
>>>   --bindir=DIR            user executables [EPREFIX/bin]
>>>   --sbindir=DIR           system admin executables [EPREFIX/sbin]
>>>   --libexecdir=DIR        program executables [EPREFIX/libexec]
>>>   --sysconfdir=DIR        read-only single-machine data [PREFIX/etc]
>>>   --sharedstatedir=DIR    modifiable architecture-independent data
>>> [PREFIX/com]
>>>   --localstatedir=DIR     modifiable single-machine data [PREFIX/var]
>>>   --libdir=DIR            object code libraries [EPREFIX/lib]
>>>   --includedir=DIR        C header files [PREFIX/include]
>>>   --oldincludedir=DIR     C header files for non-gcc [/usr/include]
>>>   --datarootdir=DIR       read-only arch.-independent data root [PREFIX/share]
>>>   --datadir=DIR           read-only architecture-independent data
>>> [DATAROOTDIR]
>>>   --infodir=DIR           info documentation [DATAROOTDIR/info]
>>>   --localedir=DIR         locale-dependent data [DATAROOTDIR/locale]
>>>   --mandir=DIR            man documentation [DATAROOTDIR/man]
>>>   --docdir=DIR            documentation root [DATAROOTDIR/doc/virt-manager]
>>>   --htmldir=DIR           html documentation [DOCDIR]
>>>   --dvidir=DIR            dvi documentation [DOCDIR]
>>>   --pdfdir=DIR            pdf documentation [DOCDIR]
>>>   --psdir=DIR             ps documentation [DOCDIR]
>>>
>>
>> I'm not entirely sure I follow, but I'd try playing around with --libdir or
>> --datadir.
> 
> I figured I would do that, but would this change the installation path of
> anything else?
> 

Possibly, you'll just have to give it a go and see. Another option would be to
find another app in bsd ports that installs gettext files, and see what they
are doing to install things to the correct location.

- Cole



From jhelfman at e-e.com  Wed Jul 18 18:39:09 2012
From: jhelfman at e-e.com (Jason Helfman)
Date: Wed, 18 Jul 2012 11:39:09 -0700
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <5006FA54.5060306@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
	<5006FA54.5060306@redhat.com>
Message-ID: <20120718183909.GG76254@dormouse.experts-exchange.com>

On Wed, Jul 18, 2012 at 02:03:00PM -0400, Cole Robinson thus spake:
>On 07/18/2012 01:37 PM, Jason Helfman wrote:
>> On Wed, Jul 18, 2012 at 11:26:13AM -0600, Eric Blake thus spake:
>>> On 07/18/2012 11:17 AM, Cole Robinson wrote:
>>>> On 07/17/2012 07:39 PM, Jason Helfman wrote:
>>>>> Hi,
>>>>>
>>>>> I noticed in virt-manager, it has an exec on /usr/bin/python. In FreeBSD,
>>>>> this is /usr/local/bin/python. Is there a way I can change this either
>>>>> through a configure flag at build time, or patch a Makefile?
>>>>>
>>>>
>>>> Not at the moment.
>>
>> Oy. Can this be configurable in the next release?
>
>I'll take a patch for the /usr/bin/env change like Eric suggested.
>
>- Cole

Attached is patch for compatibility.

Thanks!
-jgh

-- 
Jason Helfman
System Administrator
experts-exchange.com
http://www.experts-exchange.com/M_4830110.html
E4AD 7CF1 1396 27F6 79DD  4342 5E92 AD66 8C8C FBA5
-------------- next part --------------
commit bab825e555383f150f5967b22707dc9ea4123cf8
Author: Jason Helfman <jgh at FreeBSD.org>
Date:   Wed Jul 18 11:36:36 2012 -0700

    Revert "Specify /usr/bin/python explicitly"
    
    This reverts commit 5104669795788a1685006761b437e9b296840f89.

diff --git a/src/virt-manager-tui.in b/src/virt-manager-tui.in
index fb3caeb..192cc22 100644
--- a/src/virt-manager-tui.in
+++ b/src/virt-manager-tui.in
@@ -1,3 +1,3 @@
 #!/bin/sh
 
-exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
+exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"
diff --git a/src/virt-manager.in b/src/virt-manager.in
index fb3caeb..192cc22 100755
--- a/src/virt-manager.in
+++ b/src/virt-manager.in
@@ -1,3 +1,3 @@
 #!/bin/sh
 
-exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
+exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"

From crobinso at redhat.com  Thu Jul 19 19:26:19 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Thu, 19 Jul 2012 15:26:19 -0400
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <20120718183909.GG76254@dormouse.experts-exchange.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
	<5006FA54.5060306@redhat.com>
	<20120718183909.GG76254@dormouse.experts-exchange.com>
Message-ID: <50085F5B.3060508@redhat.com>

On 07/18/2012 02:39 PM, Jason Helfman wrote:
> commit bab825e555383f150f5967b22707dc9ea4123cf8
> Author: Jason Helfman <jgh at FreeBSD.org>
> Date:   Wed Jul 18 11:36:36 2012 -0700
> 
>     Revert "Specify /usr/bin/python explicitly"
>     
>     This reverts commit 5104669795788a1685006761b437e9b296840f89.
> 
> diff --git a/src/virt-manager-tui.in b/src/virt-manager-tui.in
> index fb3caeb..192cc22 100644
> --- a/src/virt-manager-tui.in
> +++ b/src/virt-manager-tui.in
> @@ -1,3 +1,3 @@
>  #!/bin/sh
>  
> -exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
> +exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"
> diff --git a/src/virt-manager.in b/src/virt-manager.in
> index fb3caeb..192cc22 100755
> --- a/src/virt-manager.in
> +++ b/src/virt-manager.in
> @@ -1,3 +1,3 @@
>  #!/bin/sh
>  
> -exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
> +exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"

One instance was missing, so I fixed that, added some more details, and pushed:

http://git.fedorahosted.org/git?p=virt-manager.git;a=commit;h=30282a17091014ba5a96d20f4260df55779933c4

Thanks,
Cole



From andersonaa at gmail.com  Thu Jul 19 19:37:16 2012
From: andersonaa at gmail.com (Anderson Alves de Albuquerque)
Date: Thu, 19 Jul 2012 16:37:16 -0300
Subject: [virt-tools-list] Problem with shutdown
Message-ID: <CACBBpwTrd6+s18m+UxBq26qC2uNvyhnWGK_UoXb-WLtEbtx3bw@mail.gmail.com>

 I am using centos 6.3 (Host machine) and virt-manger and guest machines
with openbsd 5.1, centos (Zimbra) and another with Openbsd 5.1.

 I am having problem with shutdown in Host machine. Host machine turn off
after 15 minutes or more to  shutdown. the host machine has a lot of time
to shutdowsn. This is a problem because I want that my nobreak APC shutdown
Host machine 5 minutes before battery turn off.

  What Do i have do of the shutdown Be more fast?




-- 
[], Anderson Alves de Albuquerque.
---------------------------------------------------------------
E-mails: andersonalvesdealbuquerque#hotmail.com (replace # by @)
andersonaa#gmail.com (replace # by @)
Skype: andersonalvesdealbuquerque
ICQ: 73222660
---------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120719/67eabe83/attachment.htm>

From eblake at redhat.com  Thu Jul 19 19:54:25 2012
From: eblake at redhat.com (Eric Blake)
Date: Thu, 19 Jul 2012 13:54:25 -0600
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <50085F5B.3060508@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
	<5006FA54.5060306@redhat.com>
	<20120718183909.GG76254@dormouse.experts-exchange.com>
	<50085F5B.3060508@redhat.com>
Message-ID: <500865F1.4090902@redhat.com>

On 07/19/2012 01:26 PM, Cole Robinson wrote:
>>  #!/bin/sh
>>  
>> -exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
>> +exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"
> 
> One instance was missing, so I fixed that, added some more details, and pushed:
> 
> http://git.fedorahosted.org/git?p=virt-manager.git;a=commit;h=30282a17091014ba5a96d20f4260df55779933c4

> 
> The change was originally made for unknown reasons: it was requested
> through a RHEL bug that didn't have any justification and I just changed
> it in ignorance:
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=708181
> 
>> How reproducible:
> Always, if /usr/bin is removed off $PATH or if a different/custom Python release is placed on another directory and this directory preceeds /usr/bin in $PATH.

Actually, calling the first 'python' on PATH _is_ potentially
problematic, if you mix more than one python stack on your system.  It
may mean that we need a ./configure --with-python=/path/to/python, then
install the tools with a hard-coded substitution to the absolute path
passed at configure time instead of the use of env to run the first
python on PATH.  But let's not worry about that unless someone actually
complains that it broke their setup.

-- 
Eric Blake   eblake at redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 620 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120719/19dda03d/attachment.sig>

From crobinso at redhat.com  Thu Jul 19 19:56:33 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Thu, 19 Jul 2012 15:56:33 -0400
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <500865F1.4090902@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
	<5006FA54.5060306@redhat.com>
	<20120718183909.GG76254@dormouse.experts-exchange.com>
	<50085F5B.3060508@redhat.com> <500865F1.4090902@redhat.com>
Message-ID: <50086671.7020300@redhat.com>

On 07/19/2012 03:54 PM, Eric Blake wrote:
> On 07/19/2012 01:26 PM, Cole Robinson wrote:
>>>  #!/bin/sh
>>>  
>>> -exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
>>> +exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"
>>
>> One instance was missing, so I fixed that, added some more details, and pushed:
>>
>> http://git.fedorahosted.org/git?p=virt-manager.git;a=commit;h=30282a17091014ba5a96d20f4260df55779933c4
> 
>>
>> The change was originally made for unknown reasons: it was requested
>> through a RHEL bug that didn't have any justification and I just changed
>> it in ignorance:
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=708181
>>
>>> How reproducible:
>> Always, if /usr/bin is removed off $PATH or if a different/custom Python release is placed on another directory and this directory preceeds /usr/bin in $PATH.
> 
> Actually, calling the first 'python' on PATH _is_ potentially
> problematic, if you mix more than one python stack on your system.  It
> may mean that we need a ./configure --with-python=/path/to/python, then
> install the tools with a hard-coded substitution to the absolute path
> passed at configure time instead of the use of env to run the first
> python on PATH.  But let's not worry about that unless someone actually
> complains that it broke their setup.
> 

Is that any different from /usr/bin/env python though, the change libvirt made?

- Cole



From eblake at redhat.com  Thu Jul 19 20:01:15 2012
From: eblake at redhat.com (Eric Blake)
Date: Thu, 19 Jul 2012 14:01:15 -0600
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <50086671.7020300@redhat.com>
References: <4FFAF3DC.3000301@redhat.com> <4FFB4D0E.7050409@redhat.com>
	<20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
	<5006FA54.5060306@redhat.com>
	<20120718183909.GG76254@dormouse.experts-exchange.com>
	<50085F5B.3060508@redhat.com> <500865F1.4090902@redhat.com>
	<50086671.7020300@redhat.com>
Message-ID: <5008678B.5030909@redhat.com>

On 07/19/2012 01:56 PM, Cole Robinson wrote:
>>>> -exec /usr/bin/python "::PYTHONDIR::/::PACKAGE::.py" "$@"
>>>> +exec python "::PYTHONDIR::/::PACKAGE::.py" "$@"

> 
> Is that any different from /usr/bin/env python though, the change libvirt made?

#! lines have to use an absolute path, but use of /usr/bin/env is a
great way to allow a PATH lookup.  Your code is not in a #! line.
Either way, libvirt's '#!/usr/bin/env python' and your 'exec python' are
both doing the same thing - a PATH lookup for the first available
python.  Like I said, it might not always do the right thing, but let's
wait until someone actually files a bug report about it, because it is
at least better than forcing a hard-coded call to /usr/bin/python.

-- 
Eric Blake   eblake at redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 620 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120719/043f45ab/attachment.sig>

From crobinso at redhat.com  Thu Jul 19 20:25:47 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Thu, 19 Jul 2012 16:25:47 -0400
Subject: [virt-tools-list] Problem with shutdown
In-Reply-To: <CACBBpwTrd6+s18m+UxBq26qC2uNvyhnWGK_UoXb-WLtEbtx3bw@mail.gmail.com>
References: <CACBBpwTrd6+s18m+UxBq26qC2uNvyhnWGK_UoXb-WLtEbtx3bw@mail.gmail.com>
Message-ID: <50086D4B.2080700@redhat.com>

On 07/19/2012 03:37 PM, Anderson Alves de Albuquerque wrote:
> 
> ? I am using centos 6.3 (Host machine) and virt-manger and guest machines with
> openbsd 5.1, centos (Zimbra) and another with Openbsd 5.1.
> 
> ? I am having problem with shutdown in Host machine. Host machine turn off
> after 15 minutes or more to?  shutdown. the host machine has a lot of time to
> shutdowsn. This is a problem because I want that my nobreak APC shutdown Host
> machine 5 minutes before battery turn off.

I'm confused... you are talking about the host shutdown time? Is there a
reason you think virt is involved?

Maybe check if the libvirt-guests service is running, and stop it. Otherwise
host shutdown has to wait for your VMs to migrate to disk before the host will
shutdown. If you need that functionality, maybe you can move the
libvirt-guests service to run before the UPS notification in the shutdown
process (but maybe that's not possible, I'm just guessing).

- Cole

> 
> 
> 
> 
> -- 
> [], Anderson Alves de Albuquerque.
> ---------------------------------------------------------------
> E-mails: andersonalvesdealbuquerque#hotmail.com <http://hotmail.com> (replace
> # by @)
> andersonaa#gmail.com <http://gmail.com> (replace # by @)? 
> Skype: andersonalvesdealbuquerque
> ICQ: 73222660
> ---------------------------------------------------------------
> 
> 
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
> 




From raahuja at au1.ibm.com  Fri Jul 20 05:02:48 2012
From: raahuja at au1.ibm.com (Rahul Ahuja)
Date: Fri, 20 Jul 2012 15:02:48 +1000
Subject: [virt-tools-list] Question regarding the virtual client.
Message-ID: <OF8025A291.E1A12A78-ONCA257A41.001B899A-CA257A41.001B4EB3@au1.ibm.com>


Hello Guys,

I have a linux redhat base machine running the virtual client.
I wanted to know the way to copy files from my linux base machine and guest
virtual windows XP machine using the virtual client.


Regards,

Rahu




From list-virt at legroom.net  Sat Jul 21 08:40:27 2012
From: list-virt at legroom.net (Jared)
Date: Sat, 21 Jul 2012 03:40:27 -0500
Subject: [virt-tools-list] virt-install cpuset=auto error
Message-ID: <500A6AFB.8030500@legroom.net>

Hi.

I'm trying to create new KVM images using the --cpuset=auto option "to
automatically determine the optimal cpu pinning using numa data" (from the
man page).  Problem is, I can't get this to work.  The following always
appears in my virt-install.log file when creating a new image:

DEBUG (cli:833) Not setting cpuset: unsupported operand type(s) for *:
'NoneType' and 'int'

I'm pretty sure I do have NUMA support properly configured and enabled and
all of that, but unfortunately I don't have a great deal of experience with
this stuff yet, so I can't guarantee that this isn't a user problem.

I've primarily been following the advice in this guide:

http://berrange.com/posts/2010/02/12/controlling-guest-cpu-numa-affinity-in-libvirt-with-qemu-kvm-xen/

When I run the nodeinfo and capabilities commands, I do get the appropriate
NUM cell(s) and topology output as shown in his examples.  Similarly for
freecell, memtune, numatune, and all the other related commands I could find
discussed on the internet.  However, no matter what I try, I can't get
virt-install to do the auto pinning for me.

Is there anything else I could be missing?  Some library or process
dependency outside of libvirt that I'm missing?  or could this be just a bug?

Would appreciate any guidance in troubleshooting and identifying the
problem.  I've been working on this off-and-on for a couple nights now and
have had no luck.

Thanks.

-- 
Jared



From rjones at redhat.com  Sun Jul 22 19:48:33 2012
From: rjones at redhat.com (Richard W.M. Jones)
Date: Sun, 22 Jul 2012 20:48:33 +0100
Subject: [virt-tools-list] [libvirt] ANNOUNCE: virt-manager 0.9.3
In-Reply-To: <500865F1.4090902@redhat.com>
References: <20120710213101.GN20543@dormouse.experts-exchange.com>
	<20120710214028.GO20543@dormouse.experts-exchange.com>
	<20120717233921.GA65524@dormouse.experts-exchange.com>
	<5006EFA9.7020009@redhat.com> <5006F1B5.5050601@redhat.com>
	<20120718173702.GC76254@dormouse.experts-exchange.com>
	<5006FA54.5060306@redhat.com>
	<20120718183909.GG76254@dormouse.experts-exchange.com>
	<50085F5B.3060508@redhat.com> <500865F1.4090902@redhat.com>
Message-ID: <20120722194833.GA779@amd.home.annexia.org>

On Thu, Jul 19, 2012 at 01:54:25PM -0600, Eric Blake wrote:
> Actually, calling the first 'python' on PATH _is_ potentially
> problematic, if you mix more than one python stack on your system.  It
> may mean that we need a ./configure --with-python=/path/to/python, then
> install the tools with a hard-coded substitution to the absolute path
> passed at configure time instead of the use of env to run the first
> python on PATH.

This is essentially what we do in libguestfs for python and ruby
binaries.  The reason was because of Debian where parallel installs of
different {Python,Ruby} stacks are possible and perhaps even routine.

When you do this, you also need to key several other things off the
binary, eg. paths to libraries.  See this in configure.ac:

https://github.com/libguestfs/libguestfs/blob/master/configure.ac#L896
https://github.com/libguestfs/libguestfs/blob/master/configure.ac#L965

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming blog: http://rwmj.wordpress.com
Fedora now supports 80 OCaml packages (the OPEN alternative to F#)
http://cocan.org/getting_started_with_ocaml_on_red_hat_and_fedora



From yupzhang at redhat.com  Mon Jul 23 07:47:12 2012
From: yupzhang at redhat.com (yupzhang)
Date: Mon, 23 Jul 2012 15:47:12 +0800
Subject: [virt-tools-list] virt-install cpuset=auto error
In-Reply-To: <500A6AFB.8030500@legroom.net>
References: <500A6AFB.8030500@legroom.net>
Message-ID: <500D0180.4040508@redhat.com>

There is a bug about cpuset=auto *Bug 834495* 
<https://bugzilla.redhat.com/show_bug.cgi?id=834495> -virt-install 
--cpuset=auto breaks [ASSIGNED]

Thanks
yuping
On 07/21/2012 04:40 PM, Jared wrote:
> Hi.
>
> I'm trying to create new KVM images using the --cpuset=auto option "to
> automatically determine the optimal cpu pinning using numa data" (from the
> man page).  Problem is, I can't get this to work.  The following always
> appears in my virt-install.log file when creating a new image:
>
> DEBUG (cli:833) Not setting cpuset: unsupported operand type(s) for *:
> 'NoneType' and 'int'
>
> I'm pretty sure I do have NUMA support properly configured and enabled and
> all of that, but unfortunately I don't have a great deal of experience with
> this stuff yet, so I can't guarantee that this isn't a user problem.
>
> I've primarily been following the advice in this guide:
>
> http://berrange.com/posts/2010/02/12/controlling-guest-cpu-numa-affinity-in-libvirt-with-qemu-kvm-xen/
>
> When I run the nodeinfo and capabilities commands, I do get the appropriate
> NUM cell(s) and topology output as shown in his examples.  Similarly for
> freecell, memtune, numatune, and all the other related commands I could find
> discussed on the internet.  However, no matter what I try, I can't get
> virt-install to do the auto pinning for me.
>
> Is there anything else I could be missing?  Some library or process
> dependency outside of libvirt that I'm missing?  or could this be just a bug?
>
> Would appreciate any guidance in troubleshooting and identifying the
> problem.  I've been working on this off-and-on for a couple nights now and
> have had no luck.
>
> Thanks.
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120723/d3f5e635/attachment.htm>

From list-virt at legroom.net  Mon Jul 23 07:58:30 2012
From: list-virt at legroom.net (Jared)
Date: Mon, 23 Jul 2012 02:58:30 -0500
Subject: [virt-tools-list] virt-install cpuset=auto error
In-Reply-To: <500D0180.4040508@redhat.com>
References: <500A6AFB.8030500@legroom.net> <500D0180.4040508@redhat.com>
Message-ID: <500D0426.7090509@legroom.net>

On 07/23/2012 02:47 AM, yupzhang wrote:
> There is a bug about cpuset=auto *Bug 834495*
> <https://bugzilla.redhat.com/show_bug.cgi?id=834495> -virt-install
> --cpuset=auto breaks [ASSIGNED]
> 
> Thanks
> yuping


Thanks, yuping.  That certainly sounds related to my problem, but I'm unable
to view the bug you linked to.  Even after logging in to bugzilla, I just
get an unauthorized error.

Do you know how I can view the details of this bug?  or, failing that, could
you please let me know if there's any discussion of a fix/workaround?

Thanks!

-- 
Jared



From yupzhang at redhat.com  Mon Jul 23 08:36:07 2012
From: yupzhang at redhat.com (yupzhang)
Date: Mon, 23 Jul 2012 16:36:07 +0800
Subject: [virt-tools-list] virt-install cpuset=auto error
In-Reply-To: <500D0426.7090509@legroom.net>
References: <500A6AFB.8030500@legroom.net> <500D0180.4040508@redhat.com>
	<500D0426.7090509@legroom.net>
Message-ID: <500D0CF7.7050706@redhat.com>

On 07/23/2012 03:58 PM, Jared wrote:
> On 07/23/2012 02:47 AM, yupzhang wrote:
>> There is a bug about cpuset=auto *Bug 834495*
>> <https://bugzilla.redhat.com/show_bug.cgi?id=834495>  -virt-install
>> --cpuset=auto breaks [ASSIGNED]
>>
>> Thanks
>> yuping
>
> Thanks, yuping.  That certainly sounds related to my problem, but I'm unable
> to view the bug you linked to.  Even after logging in to bugzilla, I just
> get an unauthorized error.
>
> Do you know how I can view the details of this bug?  or, failing that, could
> you please let me know if there's any discussion of a fix/workaround?
>
> Thanks!
>
Hi Jared,

Sorry,I don't know the permission about bugzilla account.
People are focusing this issue.I think it will be fixed soon in upstream.

Thanks.
Yuping



From crobinso at redhat.com  Mon Jul 23 13:16:44 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Mon, 23 Jul 2012 09:16:44 -0400
Subject: [virt-tools-list] virt-install cpuset=auto error
In-Reply-To: <500D0426.7090509@legroom.net>
References: <500A6AFB.8030500@legroom.net> <500D0180.4040508@redhat.com>
	<500D0426.7090509@legroom.net>
Message-ID: <500D4EBC.8000902@redhat.com>

On 07/23/2012 03:58 AM, Jared wrote:
> On 07/23/2012 02:47 AM, yupzhang wrote:
>> There is a bug about cpuset=auto *Bug 834495*
>> <https://bugzilla.redhat.com/show_bug.cgi?id=834495> -virt-install
>> --cpuset=auto breaks [ASSIGNED]
>>
>> Thanks
>> yuping
> 
> 
> Thanks, yuping.  That certainly sounds related to my problem, but I'm unable
> to view the bug you linked to.  Even after logging in to bugzilla, I just
> get an unauthorized error.
> 
> Do you know how I can view the details of this bug?  or, failing that, could
> you please let me know if there's any discussion of a fix/workaround?
> 
> Thanks!
> 

Unfortunately that bug has some customer related stuff attached but it can't
be made private. That said, the latest release of virtinst, 0.600.2, should
fix this issue. So I'd recommend giving that a spin.

- Cole



From berrange at redhat.com  Mon Jul 23 13:54:20 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 14:54:20 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 01/25] build: allow
 building with newer glibc-headers and -O0
In-Reply-To: <1342553107-27947-2-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-2-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723135420.GU7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:43PM +0200, Marc-Andr? Lureau wrote:
>     Fix copied from libvirt, commit by Eric Blake.
> 
>     glibc 2.15 (on Fedora 17) coupled with explicit disabling of
>     optimization during development dies a painful death:
> 
>     /usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires compiling with optimization (-O) [-Werror=cpp]
> 
>     Work around this by only conditionally defining _FORTIFY_SOURCE,
>     in the case where glibc can actually use it.  The trick is using
>     AH_VERBATIM instead of AC_DEFINE.
> ---
>  acinclude.m4 |    9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/acinclude.m4 b/acinclude.m4
> index 73b5a7b..0c308bc 100644
> --- a/acinclude.m4
> +++ b/acinclude.m4
> @@ -19,7 +19,7 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
>  
>      warnCFLAGS=
>  
> -    try_compiler_flags="-Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
> +    try_compiler_flags="-fexceptions -fstack-protector --param=ssp-buffer-size=4 -fasynchronous-unwind-tables"
>  
>      case "$enable_compile_warnings" in
>      no)
> @@ -44,6 +44,13 @@ AC_DEFUN([VIRT_VIEWER_COMPILE_WARNINGS],[
>  	;;
>      esac
>  
> +    AH_VERBATIM([FORTIFY_SOURCE],
> +                [/* Enable compile-time and run-time bounds-checking, and some warnings. */
> +                 #if defined __OPTIMIZE__ && __OPTIMIZE__
> +                 # define _FORTIFY_SOURCE 2
> +                 #endif
> +                ])
> +
>      compiler_flags=
>      for option in $try_compiler_flags; do
>  	SAVE_CFLAGS="$CFLAGS"

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 13:58:11 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 14:58:11 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 03/25] spice: disconnect
 signal handlers when either object is destroyed
In-Reply-To: <1342553107-27947-4-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-4-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723135811.GV7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:45PM +0200, Marc-Andr? Lureau wrote:
> Use virt_viewer_signal_connect_object(), a copy of telepathy
> utility function tp_g_signal_connect_object(). This function
> will take care of removing signal handler if any of emitter or
> attached object are destroyed.
> 
> The following patches will have this condition met, since there is no
> longer 1-1 relation between channel and display. The channels can
> continue to be around when some of the display are removed.
> ---
>  src/virt-viewer-display-spice.c |   26 ++++-----
>  src/virt-viewer-util.c          |  117 +++++++++++++++++++++++++++++++++++++++
>  src/virt-viewer-util.h          |    6 ++
>  3 files changed, 135 insertions(+), 14 deletions(-)
> diff --git a/src/virt-viewer-util.h b/src/virt-viewer-util.h
> index 3a40651..38c8078 100644
> --- a/src/virt-viewer-util.h
> +++ b/src/virt-viewer-util.h
> @@ -40,6 +40,12 @@ int virt_viewer_util_extract_host(const char *uristr,
>                                    char **user,
>                                    int *port);
>  
> +gulong virt_viewer_signal_connect_object(gpointer instance,
> +                                         const gchar *detailed_signal,
> +                                         GCallback c_handler,
> +                                         gpointer gobject,
> +                                         GConnectFlags connect_flags);
> +
>  #endif

Superficially this looks like it tries todo the same as
g_signal_connect_object ?

http://developer.gnome.org/gobject/stable/gobject-Signals.html#g-signal-connect-object

Does it avoids the flaws mentioned about that API ?


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 13:59:20 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 14:59:20 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 04/25] build: generate
 enums type boilerplate
In-Reply-To: <1342553107-27947-5-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-5-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723135920.GW7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:46PM +0200, Marc-Andr? Lureau wrote:
> ---
>  configure.ac                      |    3 ++
>  src/Makefile.am                   |   23 ++++++++++++--
>  src/virt-viewer-enums.c.etemplate |   60 +++++++++++++++++++++++++++++++++++++
>  src/virt-viewer-enums.h.etemplate |   41 +++++++++++++++++++++++++
>  4 files changed, 124 insertions(+), 3 deletions(-)
>  create mode 100644 src/virt-viewer-enums.c.etemplate
>  create mode 100644 src/virt-viewer-enums.h.etemplate

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 13:59:57 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 14:59:57 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 05/25] Destroy dialog
 immediately after run
In-Reply-To: <1342553107-27947-6-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-6-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723135957.GX7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:47PM +0200, Marc-Andr? Lureau wrote:
> If the parent is already destroyed, it looks like the dialog is
> destroyed too. This avoids a crash when calling app_quit().
> ---
>  src/virt-viewer-app.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 40d1ff7..1447ce7 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -283,6 +283,7 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
>                                          GTK_BUTTONS_OK_CANCEL,
>                                          _("This is the last visible display. Do you want to quit?"));
>              gint result = gtk_dialog_run (GTK_DIALOG (dialog));
> +            gtk_widget_destroy(dialog);
>              switch (result) {
>              case GTK_RESPONSE_OK:
>                  virt_viewer_app_quit(self);
> @@ -290,7 +291,6 @@ virt_viewer_app_window_set_visible(VirtViewerApp *self,
>              default:
>                  break;
>              }
> -            gtk_widget_destroy(dialog);
>              return FALSE;
>          } else {
>              virt_viewer_app_quit(self);

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:00:33 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:00:33 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 06/25] Turn
 display:show-hint into flags type
In-Reply-To: <1342553107-27947-7-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-7-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140033.GY7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:48PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c     |    5 +++--
>  src/virt-viewer-display.c |   26 ++++++++++++++++----------
>  src/virt-viewer-display.h |   11 ++++++-----
>  src/virt-viewer-window.c  |    5 ++++-
>  4 files changed, 29 insertions(+), 18 deletions(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:00:59 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:00:59 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 07/25] Run-time check
 values before doing bad computation
In-Reply-To: <1342553107-27947-8-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-8-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140059.GZ7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:49PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-display.c |    6 +++++-
>  src/virt-viewer-window.c  |    5 +++++
>  2 files changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-display.c b/src/virt-viewer-display.c
> index 43ec45e..a954671 100644
> --- a/src/virt-viewer-display.c
> +++ b/src/virt-viewer-display.c
> @@ -403,6 +403,10 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
>      DEBUG_LOG("Allocated %dx%d", allocation->width, allocation->height);
>      gtk_widget_set_allocation(widget, allocation);
>  
> +    if (priv->desktopWidth == 0 ||
> +        priv->desktopHeight == 0)
> +        goto end;
> +
>      desktopAspect = (double)priv->desktopWidth / (double)priv->desktopHeight;
>  
>      if (child && gtk_widget_get_visible(child)) {
> @@ -427,7 +431,7 @@ virt_viewer_display_size_allocate(GtkWidget *widget,
>          gtk_widget_size_allocate(child, &child_allocation);
>      }
>  
> -
> +end:
>      /* This unsets the size request, so that the user can
>       * manually resize the window smaller again
>       */
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index c873ebc..4887e28 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -433,6 +433,11 @@ virt_viewer_window_resize(VirtViewerWindow *self)
>                                      (screen, gtk_widget_get_window(priv->window)),
>                                      &fullscreen);
>  
> +    g_return_if_fail(fullscreen.height > 128);
> +    g_return_if_fail(fullscreen.width > 128);
> +    g_return_if_fail(desktopWidth > 0);
> +    g_return_if_fail(desktopHeight > 0);
> +
>      desktopAspect = (double)desktopWidth / (double)desktopHeight;
>      screenAspect = (double)(fullscreen.width - 128) / (double)(fullscreen.height - 128);


ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:01:16 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:01:16 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 08/25] Mark a menu string
 as translatable
In-Reply-To: <1342553107-27947-9-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-9-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140116.GA7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:50PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 5588c6a..661cdd7 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -1634,7 +1634,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
>          gboolean visible;
>          gchar *label;
>  
> -        label = g_strdup_printf("Display %d", *nth);
> +        label = g_strdup_printf(_("Display %d"), *nth);
>          item = gtk_check_menu_item_new_with_label(label);
>          g_free(label);
>          visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:01:44 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:01:44 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 09/25] Rebuild display
 menu when a window is added or removed
In-Reply-To: <1342553107-27947-10-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-10-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140143.GB7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:51PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |    3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 661cdd7..c4f712d 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -520,6 +520,7 @@ virt_viewer_app_remove_nth_window(VirtViewerApp *self, gint nth)
>      DEBUG_LOG("Remove window %d %p", nth, win);
>      removed = g_hash_table_steal(self->priv->windows, &nth);
>      g_warn_if_fail(removed);
> +    virt_viewer_app_update_menu_displays(self);
>  
>      if (removed)
>          g_signal_emit(self, signals[SIGNAL_WINDOW_REMOVED], 0, win);
> @@ -540,6 +541,7 @@ virt_viewer_app_set_nth_window(VirtViewerApp *self, gint nth, VirtViewerWindow *
>      DEBUG_LOG("Insert window %d %p", nth, win);
>      g_hash_table_insert(self->priv->windows, key, win);
>      virt_viewer_app_set_window_subtitle(self, win, nth);
> +    virt_viewer_app_update_menu_displays(self);
>  
>      g_signal_emit(self, signals[SIGNAL_WINDOW_ADDED], 0, win);
>  }
> @@ -657,6 +659,7 @@ virt_viewer_app_display_added(VirtViewerSession *session G_GNUC_UNUSED,
>      }
>  
>      virt_viewer_window_set_display(window, display);
> +    virt_viewer_app_update_menu_displays(self);
>      g_signal_connect(display, "notify::show-hint",
>                       G_CALLBACK(display_show_hint), window);
>      g_object_notify(G_OBJECT(display), "show-hint"); /* call display_show_hint */

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:02:40 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:02:40 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 11/25] Add a DISABLED
 display hint
In-Reply-To: <1342553107-27947-12-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-12-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140240.GC7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:53PM +0200, Marc-Andr? Lureau wrote:
> This flag will help to track whether the display has been
> removed/closed and whether it really has a valid display.
> 
> Ready in contrast, is used to "hide" temporarily the display (when
> starting or redrawing the display, to avoid artifacts)
> ---
>  src/virt-viewer-app.c           |   12 +++++++-----
>  src/virt-viewer-display-spice.c |   20 ++++++++++++++++++++
>  src/virt-viewer-display.c       |   15 +++++++++++++++
>  src/virt-viewer-display.h       |    2 ++
>  src/virt-viewer-window.c        |   12 ++++++++++++
>  5 files changed, 56 insertions(+), 5 deletions(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:03:02 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:03:02 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 12/25] Prevent from adding
 the same display several time in the session
In-Reply-To: <1342553107-27947-13-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-13-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140302.GD7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:54PM +0200, Marc-Andr? Lureau wrote:
> In virt_viewer_session_spice_display_monitors(), we (re)add the
> display unconditionnaly every time we receive a new MonitorConfig.
> ---
>  src/virt-viewer-session.c |    3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/src/virt-viewer-session.c b/src/virt-viewer-session.c
> index 18b6922..9249a1f 100644
> --- a/src/virt-viewer-session.c
> +++ b/src/virt-viewer-session.c
> @@ -282,6 +282,9 @@ virt_viewer_session_new(void)
>  void virt_viewer_session_add_display(VirtViewerSession *session,
>                                       VirtViewerDisplay *display)
>  {
> +    if (g_list_find(session->priv->displays, display))
> +        return;
> +
>      session->priv->displays = g_list_append(session->priv->displays, display);
>      g_object_ref(display);
>      g_signal_emit_by_name(session, "session-display-added", display);

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:03:19 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:03:19 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 13/25] Number display
 starting from 1
In-Reply-To: <1342553107-27947-14-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-14-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140319.GE7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:55PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index 727eb81..3d7ff68 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -1639,7 +1639,7 @@ window_update_menu_displays_cb(gpointer key G_GNUC_UNUSED,
>          gboolean visible;
>          gchar *label;
>  
> -        label = g_strdup_printf(_("Display %d"), *nth);
> +        label = g_strdup_printf(_("Display %d"), *nth + 1);
>          item = gtk_check_menu_item_new_with_label(label);
>          g_free(label);
>          visible = gtk_widget_get_visible(GTK_WIDGET(virt_viewer_window_get_window(vwin)));

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:03:47 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:03:47 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 14/25] Use
 SpiceDisplay:ready property instead of channel mark
In-Reply-To: <1342553107-27947-15-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-15-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140347.GF7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:56PM +0200, Marc-Andr? Lureau wrote:
> The display can now check several conditions before the display can be
> shown, use that instead of display mark, which was not high-level
> enough.
> ---
>  src/virt-viewer-display-spice.c |   38 +++++++++++++++-----------------------
>  1 file changed, 15 insertions(+), 23 deletions(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:04:38 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:04:38 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 15/25] Bump glib > 2.22,
 add compat file
In-Reply-To: <1342553107-27947-16-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-16-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140438.GG7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:57PM +0200, Marc-Andr? Lureau wrote:
> We use API from 2.22, and some from further version. Add
> virt-glib-compat.h fallback file for those.
> ---
>  configure.ac           |    2 +-
>  src/Makefile.am        |    1 +
>  src/virt-glib-compat.h |   56 ++++++++++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 58 insertions(+), 1 deletion(-)
>  create mode 100644 src/virt-glib-compat.h

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:08:02 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:08:02 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 15/25] Bump glib > 2.22,
 add compat file
In-Reply-To: <20120723140438.GG7339@redhat.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-16-git-send-email-marcandre.lureau@gmail.com>
	<20120723140438.GG7339@redhat.com>
Message-ID: <20120723140802.GH7339@redhat.com>

On Mon, Jul 23, 2012 at 03:04:38PM +0100, Daniel P. Berrange wrote:
> On Tue, Jul 17, 2012 at 09:24:57PM +0200, Marc-Andr? Lureau wrote:
> > We use API from 2.22, and some from further version. Add
> > virt-glib-compat.h fallback file for those.
> > ---
> >  configure.ac           |    2 +-
> >  src/Makefile.am        |    1 +
> >  src/virt-glib-compat.h |   56 ++++++++++++++++++++++++++++++++++++++++++++++++
> >  3 files changed, 58 insertions(+), 1 deletion(-)
> >  create mode 100644 src/virt-glib-compat.h
> 
> ACK

Actually you need to update  virt-viewer.spec.in and mingw-virt-viewer.spec.in
with the new version number too.  The same applies to the patch changing the
spice-gtk min version number.

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:08:59 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:08:59 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 17/25] Add
 virt_viewer_window_get_display()
In-Reply-To: <1342553107-27947-18-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-18-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140859.GI7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:59PM +0200, Marc-Andr? Lureau wrote:
> Getter used in following changes.
> ---
>  src/virt-viewer-window.c |   13 +++++++++++--
>  src/virt-viewer-window.h |    1 +
>  2 files changed, 12 insertions(+), 2 deletions(-)

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:09:43 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:09:43 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 19/25] spice: factor out
 main channel lookup code
In-Reply-To: <1342553107-27947-20-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-20-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723140943.GJ7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:01PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-display-spice.c |   22 +++++++++++++++++-----
>  1 file changed, 17 insertions(+), 5 deletions(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:10:13 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:10:13 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 20/25] Make the display
 submenu insensitive if display can't be selected
In-Reply-To: <1342553107-27947-21-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-21-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141013.GK7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:02PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-app.c |   18 +++++++++++++++++-
>  1 file changed, 17 insertions(+), 1 deletion(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:10:45 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:10:45 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 21/25] Add
 VirtViewerSession::session-display-updated
In-Reply-To: <1342553107-27947-22-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-22-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141045.GL7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:03PM +0200, Marc-Andr? Lureau wrote:
> Rebuild menu when agent is connected. Only when the agent is running
> may a display be enabled/disabled.
> ---
>  src/virt-viewer-app.c           |    9 +++++++++
>  src/virt-viewer-session-spice.c |   18 ++++++++++++++----
>  src/virt-viewer-session.c       |   10 ++++++++++
>  src/virt-viewer-session.h       |    1 +
>  4 files changed, 34 insertions(+), 4 deletions(-)

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:11:22 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:11:22 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 22/25] Simplify display
 flag handling
In-Reply-To: <1342553107-27947-23-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-23-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141122.GM7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:04PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-display-spice.c |    9 +--------
>  src/virt-viewer-display.c       |   20 ++++++++++----------
>  src/virt-viewer-display.h       |    2 +-
>  src/virt-viewer-session-vnc.c   |    2 +-
>  4 files changed, 13 insertions(+), 20 deletions(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:11:40 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:11:40 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 23/25] vnc: when session
 is disconnected, make the display as non-ready
In-Reply-To: <1342553107-27947-24-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-24-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141140.GN7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:05PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/virt-viewer-session-vnc.c |    6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:12:10 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:12:10 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 24/25] Make status widget
 visible immediately
In-Reply-To: <1342553107-27947-25-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-25-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141210.GO7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:06PM +0200, Marc-Andr? Lureau wrote:
> GtkNotebook will use the currently visible widget as default page.
> If we don't show status widget before we append the display, the
> current page will be on display. Quoting Gtk+ documentation:
> 
> "Note that due to historical reasons, GtkNotebook refuses to switch to
> a page unless the child widget is visible. Therefore, it is
> recommended to show child widgets before adding them to a notebook."
> ---
>  src/virt-viewer-notebook.c |    1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/src/virt-viewer-notebook.c b/src/virt-viewer-notebook.c
> index 09e8ec4..cc7136b 100644
> --- a/src/virt-viewer-notebook.c
> +++ b/src/virt-viewer-notebook.c
> @@ -86,6 +86,7 @@ virt_viewer_notebook_init (VirtViewerNotebook *self)
>      priv->status = gtk_label_new("");
>      gtk_notebook_set_show_tabs(GTK_NOTEBOOK(self), FALSE);
>      gtk_notebook_set_show_border(GTK_NOTEBOOK(self), FALSE);
> +    gtk_widget_show_all(priv->status);
>      gtk_notebook_append_page(GTK_NOTEBOOK(self), priv->status, NULL);
>      gdk_color_parse("white", &color);
>      gtk_widget_modify_fg(priv->status, GTK_STATE_NORMAL, &color);

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:12:26 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:12:26 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 25/25] Show status page by
 default if display not ready
In-Reply-To: <1342553107-27947-26-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<1342553107-27947-26-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141226.GP7339@redhat.com>

On Tue, Jul 17, 2012 at 09:25:07PM +0200, Marc-Andr? Lureau wrote:
> Switch back to status page if display is not ready.
> ---
>  src/virt-viewer-window.c |    6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-window.c b/src/virt-viewer-window.c
> index 4bc427a..7c6e41f 100644
> --- a/src/virt-viewer-window.c
> +++ b/src/virt-viewer-window.c
> @@ -1040,8 +1040,12 @@ virt_viewer_window_set_display(VirtViewerWindow *self, VirtViewerDisplay *displa
>          virt_viewer_display_set_zoom_level(VIRT_VIEWER_DISPLAY(priv->display), priv->zoomlevel);
>          virt_viewer_display_set_auto_resize(VIRT_VIEWER_DISPLAY(priv->display), priv->auto_resize);
>  
> -        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
>          gtk_widget_show_all(GTK_WIDGET(display));
> +        gtk_notebook_append_page(GTK_NOTEBOOK(priv->notebook), GTK_WIDGET(display), NULL);
> +        /* switch back to non-display if not ready */
> +        if (!(virt_viewer_display_get_show_hint(display) &
> +              VIRT_VIEWER_DISPLAY_SHOW_HINT_READY))
> +            gtk_notebook_set_current_page(GTK_NOTEBOOK(priv->notebook), 0);
>  
>          virt_viewer_signal_connect_object(display, "display-pointer-grab",
>                                            G_CALLBACK(virt_viewer_window_pointer_grab), self, 0);

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Mon Jul 23 14:14:30 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 23 Jul 2012 15:14:30 +0100
Subject: [virt-tools-list] [PATCH virt-viewer 00/25] Add new Spice
 monitors handling support v2
In-Reply-To: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120723141430.GQ7339@redhat.com>

On Tue, Jul 17, 2012 at 09:24:42PM +0200, Marc-Andr? Lureau wrote:
> Hi,
> 
> Spice now supports multiple arbitrary-sized monitors per display
> channel. The client has to learn how to handle the new events in order
> to create, to show, and to manage the displays appropriately.
> 
> The feature is currently being merged in the various projects, to test
> it you'll need the following projects update with monitor config
> support: spice-server, qemu, spice-gtk, qxl driver and vdagent.

My main concern is probably about what level of testing this has
had. In particular can you verify that the '--wait' and '--reconnect'
flags still work, with both VNC and SPICE

ACK assuming those are working.

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From mlureau at redhat.com  Mon Jul 23 14:22:00 2012
From: mlureau at redhat.com (=?utf-8?Q?Marc-Andr=C3=A9?= Lureau)
Date: Mon, 23 Jul 2012 10:22:00 -0400 (EDT)
Subject: [virt-tools-list] [PATCH virt-viewer 03/25] spice: disconnect
 signal handlers when either object is destroyed
In-Reply-To: <20120723135811.GV7339@redhat.com>
Message-ID: <392710529.1811795.1343053320002.JavaMail.root@redhat.com>



----- Mensaje original -----
> On Tue, Jul 17, 2012 at 09:24:45PM +0200, Marc-Andr? Lureau wrote:
> > Use virt_viewer_signal_connect_object(), a copy of telepathy
> > utility function tp_g_signal_connect_object(). This function
> > will take care of removing signal handler if any of emitter or
> > attached object are destroyed.
> > 
> > The following patches will have this condition met, since there is
> > no
> > longer 1-1 relation between channel and display. The channels can
> > continue to be around when some of the display are removed.
> > ---
> >  src/virt-viewer-display-spice.c |   26 ++++-----
> >  src/virt-viewer-util.c          |  117
> >  +++++++++++++++++++++++++++++++++++++++
> >  src/virt-viewer-util.h          |    6 ++
> >  3 files changed, 135 insertions(+), 14 deletions(-)
> > diff --git a/src/virt-viewer-util.h b/src/virt-viewer-util.h
> > index 3a40651..38c8078 100644
> > --- a/src/virt-viewer-util.h
> > +++ b/src/virt-viewer-util.h
> > @@ -40,6 +40,12 @@ int virt_viewer_util_extract_host(const char
> > *uristr,
> >                                    char **user,
> >                                    int *port);
> >  
> > +gulong virt_viewer_signal_connect_object(gpointer instance,
> > +                                         const gchar
> > *detailed_signal,
> > +                                         GCallback c_handler,
> > +                                         gpointer gobject,
> > +                                         GConnectFlags
> > connect_flags);
> > +
> >  #endif
> 
> Superficially this looks like it tries todo the same as
> g_signal_connect_object ?
> 
> http://developer.gnome.org/gobject/stable/gobject-Signals.html#g-signal-connect-object
> 
> Does it avoids the flaws mentioned about that API ?

Yes, see also: http://cgit.freedesktop.org/telepathy/telepathy-glib/tree/telepathy-glib/util.c#n984

And bug https://bugzilla.gnome.org/show_bug.cgi?id=118536



From mlureau at redhat.com  Mon Jul 23 14:24:50 2012
From: mlureau at redhat.com (=?utf-8?Q?Marc-Andr=C3=A9?= Lureau)
Date: Mon, 23 Jul 2012 10:24:50 -0400 (EDT)
Subject: [virt-tools-list] [PATCH virt-viewer 15/25] Bump glib > 2.22,
 add compat file
In-Reply-To: <20120723140802.GH7339@redhat.com>
Message-ID: <1721473024.1822735.1343053490265.JavaMail.root@redhat.com>



----- Mensaje original -----
> On Mon, Jul 23, 2012 at 03:04:38PM +0100, Daniel P. Berrange wrote:
> > On Tue, Jul 17, 2012 at 09:24:57PM +0200, Marc-Andr? Lureau wrote:
> > > We use API from 2.22, and some from further version. Add
> > > virt-glib-compat.h fallback file for those.
> > > ---
> > >  configure.ac           |    2 +-
> > >  src/Makefile.am        |    1 +
> > >  src/virt-glib-compat.h |   56
> > >  ++++++++++++++++++++++++++++++++++++++++++++++++
> > >  3 files changed, 58 insertions(+), 1 deletion(-)
> > >  create mode 100644 src/virt-glib-compat.h
> > 
> > ACK
> 
> Actually you need to update  virt-viewer.spec.in and
> mingw-virt-viewer.spec.in
> with the new version number too.  The same applies to the patch
> changing the
> spice-gtk min version number.

done, thanks



From marcandre.lureau at gmail.com  Mon Jul 23 14:47:04 2012
From: marcandre.lureau at gmail.com (=?ISO-8859-1?Q?Marc=2DAndr=E9_Lureau?=)
Date: Mon, 23 Jul 2012 16:47:04 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 00/25] Add new Spice
 monitors handling support v2
In-Reply-To: <20120723141430.GQ7339@redhat.com>
References: <1342553107-27947-1-git-send-email-marcandre.lureau@gmail.com>
	<20120723141430.GQ7339@redhat.com>
Message-ID: <CAJ+F1CKKb6SBmH57MkdiQkwR0ejOm=0qU=T0LM_2neRWHAq4aQ@mail.gmail.com>

On Mon, Jul 23, 2012 at 4:14 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> ACK assuming those are working.

I just tested successfully with both VNC & Spice, now pushing the series.

thanks

-- 
Marc-Andr? Lureau



From marcandre.lureau at gmail.com  Tue Jul 24 12:23:08 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 24 Jul 2012 14:23:08 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 1/4] win32: there is no
	CONERR$, only CONOUT$
Message-ID: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>

---
 src/remote-viewer-main.c      |    2 +-
 src/windows-cmdline-wrapper.c |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/remote-viewer-main.c b/src/remote-viewer-main.c
index 156bd0b..aee90ea 100644
--- a/src/remote-viewer-main.c
+++ b/src/remote-viewer-main.c
@@ -231,7 +231,7 @@ main(int argc, char **argv)
     if (AttachConsole(ATTACH_PARENT_PROCESS) != 0) {
         freopen("CONIN$", "r", stdin);
         freopen("CONOUT$", "w", stdout);
-        freopen("CONERR$", "w", stderr);
+        freopen("CONOUT$", "w", stderr);
         dup2(fileno(stdin), STDIN_FILENO);
         dup2(fileno(stdout), STDOUT_FILENO);
         dup2(fileno(stderr), STDERR_FILENO);
diff --git a/src/windows-cmdline-wrapper.c b/src/windows-cmdline-wrapper.c
index 510763b..704e2d7 100644
--- a/src/windows-cmdline-wrapper.c
+++ b/src/windows-cmdline-wrapper.c
@@ -30,7 +30,7 @@
  *   if (AttachConsole(ATTACH_PARENT_PROCESS) != 0) {
  *       freopen("CONIN$", "r", stdin);
  *       freopen("CONOUT$", "w", stdout);
- *       freopen("CONERR$", "w", stderr);
+ *       freopen("CONOUT$", "w", stderr);
  *       dup2(fileno(stdin), STDIN_FILENO);
  *       dup2(fileno(stdout), STDOUT_FILENO);
  *       dup2(fileno(stderr), STDERR_FILENO);
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 24 12:23:09 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 24 Jul 2012 14:23:09 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 2/4] nsis: update required
	files and libraries
In-Reply-To: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1343132591-1039-2-git-send-email-marcandre.lureau@gmail.com>

---
 data/virt-viewer.nsis.in |   46 +++++++++++++++++++++++++++++-----------------
 1 file changed, 29 insertions(+), 17 deletions(-)

diff --git a/data/virt-viewer.nsis.in b/data/virt-viewer.nsis.in
index e438159..f221a20 100644
--- a/data/virt-viewer.nsis.in
+++ b/data/virt-viewer.nsis.in
@@ -66,6 +66,7 @@ Section "VirtViewer"
   File "@prefix@/bin/libcairo-2.dll"
   File "@prefix@/bin/libcelt051-0.dll"
   File "@prefix@/bin/libcrypto-10.dll"
+  File "@prefix@/bin/libexpat-1.dll"
   File "@prefix@/bin/libffi-6.dll"
   File "@prefix@/bin/libfontconfig-1.dll"
   File "@prefix@/bin/libfreetype-6.dll"
@@ -99,8 +100,8 @@ Section "VirtViewer"
   File "@prefix@/bin/libpangowin32-1.0-0.dll"
   File "@prefix@/bin/libpixman-1-0.dll"
   File "@prefix@/bin/libpng15-15.dll"
-  File "@prefix@/bin/libspice-client-glib-2.0-1.dll"
-  File "@prefix@/bin/libspice-client-gtk-2.0-1.dll"
+  File "@prefix@/bin/libspice-client-glib-2.0-8.dll"
+  File "@prefix@/bin/libspice-client-gtk-2.0-4.dll"
   File "@prefix@/bin/libspice-controller-0.dll"
   File "@prefix@/bin/libssl-10.dll"
   File "@prefix@/bin/libtasn1-3.dll"
@@ -128,6 +129,7 @@ Section "VirtViewer"
   File /oname=remote-viewer.com "${DESTDIR}@prefix@/bin/windows-cmdline-wrapper.exe"
   File "${DESTDIR}@prefix@/bin/remote-viewer.exe"
   File "${DESTDIR}@prefix@/bin/virt-viewer.exe"
+  File "@prefix@/bin/readline.dll"
   File "@prefix@/bin/zlib1.dll"
 
   SetOutPath "$INSTDIR\etc"
@@ -212,15 +214,11 @@ Section "Uninstall"
   DeleteRegKey HKCU "${REG_UNINSTALL}"
   DeleteRegKey HKCU "${REG_IEPOLICY}"
 
-  RMDir /r "$INSTDIR\lib\pango"
-
-  RMDir /r "$INSTDIR\lib\gtk-2.0"
-
-  RMDir /r "$INSTDIR\share\icons\hicolor"
 
-  Delete /rebootok "$INSTDIR\share\virt-viewer\ui\*.xml"
-  Delete /rebootok "$INSTDIR\share\themes\MS-Windows\gtk-2.0\gtkrc"
+  RMDir /r "$INSTDIR\share"
 
+  RMDir /r "$INSTDIR\lib\pango"
+  RMDir /r "$INSTDIR\lib\gtk-2.0"
   Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstdirectsoundsrc.dll"
   Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstdirectsoundsink.dll"
   Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstcoreelements.dll"
@@ -231,18 +229,19 @@ Section "Uninstall"
   Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstaudioconvert.dll"
   Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstapp.dll"
   RMDir "$INSTDIR\lib\gstreamer-0.10"
-
   RMDir /r "$INSTDIR\lib\gdk-pixbuf-2.0"
 
   RMDir /r "$INSTDIR\etc\pango"
-
   RMDir /r "$INSTDIR\etc\gtk-2.0"
-
   RMDir /r "$INSTDIR\etc\fonts"
 
   Delete /rebootok "$INSTDIR\bin\remote-viewer.com"
   Delete /rebootok "$INSTDIR\bin\remote-viewer.exe"
+  Delete /rebootok "$INSTDIR\bin\virt-viewer.exe"
+  Delete /rebootok "$INSTDIR\bin\virsh.exe"
   Delete /rebootok "$INSTDIR\bin\pango-querymodules.exe"
+  Delete /rebootok "$INSTDIR\bin\virt-pki-validate"
+  Delete /rebootok "$INSTDIR\bin\virt-xml-validate"
   Delete /rebootok "$INSTDIR\bin\libpangoft2-1.0-0.dll"
   Delete /rebootok "$INSTDIR\bin\libgdk_pixbuf-2.0-0.dll"
   Delete /rebootok "$INSTDIR\bin\libgdk-win32-2.0-0.dll"
@@ -258,16 +257,15 @@ Section "Uninstall"
   Delete /rebootok "$INSTDIR\bin\gst-inspect-0.10.exe"
   Delete /rebootok "$INSTDIR\bin\gst-feedback.exe"
   Delete /rebootok "$INSTDIR\bin\gst-discoverer-0.10.exe"
-  RMDir "$INSTDIR\bin"
-
-  Delete /rebootok "$INSTDIR\bin\libffi-6.dll"
   Delete /rebootok "$INSTDIR\bin\zlib1.dll"
+  Delete /rebootok "$INSTDIR\bin\readline.dll"
+  Delete /rebootok "$INSTDIR\bin\libffi-6.dll"
   Delete /rebootok "$INSTDIR\bin\libp11-kit-0.dll"
   Delete /rebootok "$INSTDIR\bin\libxml2-2.dll"
   Delete /rebootok "$INSTDIR\bin\libtasn1-3.dll"
   Delete /rebootok "$INSTDIR\bin\libssl-10.dll"
-  Delete /rebootok "$INSTDIR\bin\libspice-client-gtk-2.0-1.dll"
-  Delete /rebootok "$INSTDIR\bin\libspice-client-glib-2.0-1.dll"
+  Delete /rebootok "$INSTDIR\bin\libspice-client-gtk-2.0-4.dll"
+  Delete /rebootok "$INSTDIR\bin\libspice-client-glib-2.0-8.dll"
   Delete /rebootok "$INSTDIR\bin\libspice-controller-0.dll"
   Delete /rebootok "$INSTDIR\bin\libpng15-15.dll"
   Delete /rebootok "$INSTDIR\bin\libpixman-1-0.dll"
@@ -278,6 +276,16 @@ Section "Uninstall"
   Delete /rebootok "$INSTDIR\bin\liborc-0.4-0.dll"
   Delete /rebootok "$INSTDIR\bin\libjpeg-62.dll"
   Delete /rebootok "$INSTDIR\bin\libintl-8.dll"
+  Delete /rebootok "$INSTDIR\bin\libvirt-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libcurl-4.dll"
+  Delete /rebootok "$INSTDIR\bin\libidn-11.dll"
+  Delete /rebootok "$INSTDIR\bin\libssh2-1.dll"
+  Delete /rebootok "$INSTDIR\bin\libgvnc-1.0-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libgtk-vnc-1.0-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libgpg-error-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libportablexdr-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libvirt-qemu-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libssp-0.dll"
   Delete /rebootok "$INSTDIR\bin\iconv.dll"
   Delete /rebootok "$INSTDIR\bin\libgtk-win32-2.0-0.dll"
   Delete /rebootok "$INSTDIR\bin\libgthread-2.0-0.dll"
@@ -291,6 +299,8 @@ Section "Uninstall"
   Delete /rebootok "$INSTDIR\bin\libgstaudio-0.10-0.dll"
   Delete /rebootok "$INSTDIR\bin\libgstapp-0.10-0.dll"
   Delete /rebootok "$INSTDIR\bin\libgobject-2.0-0.dll"
+  Delete /rebootok "$INSTDIR\bin\libgnutls-26.dll"
+  Delete /rebootok "$INSTDIR\bin\libgcrypt-11.dll"
   Delete /rebootok "$INSTDIR\bin\libgmodule-2.0-0.dll"
   Delete /rebootok "$INSTDIR\bin\libglib-2.0-0.dll"
   Delete /rebootok "$INSTDIR\bin\libgio-2.0-0.dll"
@@ -298,11 +308,13 @@ Section "Uninstall"
   Delete /rebootok "$INSTDIR\bin\libgdk-win32-2.0-0.dll"
   Delete /rebootok "$INSTDIR\bin\libfreetype-6.dll"
   Delete /rebootok "$INSTDIR\bin\libfontconfig-1.dll"
+  Delete /rebootok "$INSTDIR\bin\libexpat-1.dll"
   Delete /rebootok "$INSTDIR\bin\libcrypto-10.dll"
   Delete /rebootok "$INSTDIR\bin\libcelt051-0.dll"
   Delete /rebootok "$INSTDIR\bin\libcairo-2.dll"
   Delete /rebootok "$INSTDIR\bin\libatk-1.0-0.dll"
   RMDir "$INSTDIR\bin"
+  RMDir "$INSTDIR\lib"
   RMDir "$INSTDIR"
 SectionEnd
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 24 12:23:10 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 24 Jul 2012 14:23:10 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 3/4] nsis: add deps.txt file
	in installer
In-Reply-To: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1343132591-1039-3-git-send-email-marcandre.lureau@gmail.com>

This helps track package version that were used during the
build of Windows installer. It's not ideal, but make up the
lack of package management on windows
---
 Makefile.am              |    6 ++++++
 data/Makefile.am         |   13 +++++++++++++
 data/virt-viewer.nsis.in |    5 +++++
 3 files changed, 24 insertions(+)

diff --git a/Makefile.am b/Makefile.am
index f307c06..2407b1b 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -57,4 +57,10 @@ gen-ChangeLog:
 	  mv $(distdir)/cl-t $(distdir)/ChangeLog;		\
 	fi
 
+if OS_WIN32
+.PHONY: nsis
+nsis:
+	make -C $(builddir)/data virt-viewer-$(VERSION).exe
+endif
+
 -include $(top_srcdir)/git.mk
diff --git a/data/Makefile.am b/data/Makefile.am
index ea2bb0a..38ffdb2 100644
--- a/data/Makefile.am
+++ b/data/Makefile.am
@@ -5,4 +5,17 @@ EXTRA_DIST =					\
 	virt-viewer.nsis.in			\
 	$(NULL)
 
+if OS_WIN32
+deps.txt:
+	$(AM_V_GEN)rpm -qa | grep $(host_os) | sort | unix2dos > $@
+
+virt-viewer-$(VERSION).exe: virt-viewer.nsis deps.txt
+	$(AM_V_GEN)DESTDIR=`mktemp -d` && \
+	make -C $(top_builddir) install DESTDIR=$$DESTDIR >/dev/null && \
+	makensis -NOCD -DDESTDIR=$$DESTDIR $< >/dev/null && \
+	rm -rf $$DESTDIR
+
+CLEANFILES = deps.txt virt-viewer-$(VERSION).exe
+endif
+
 -include $(top_srcdir)/git.mk
diff --git a/data/virt-viewer.nsis.in b/data/virt-viewer.nsis.in
index f221a20..98ce83a 100644
--- a/data/virt-viewer.nsis.in
+++ b/data/virt-viewer.nsis.in
@@ -49,6 +49,9 @@ Section "VirtViewer"
   SectionIn RO
   WriteRegStr HKCU "Software\virt-viewer" "" $INSTDIR
 
+  SetOutPath "$INSTDIR"
+  File "deps.txt"
+
   SetOutPath "$INSTDIR\bin"
   File "@prefix@/bin/gst-discoverer-0.10.exe"
   File "@prefix@/bin/gst-feedback.exe"
@@ -315,6 +318,8 @@ Section "Uninstall"
   Delete /rebootok "$INSTDIR\bin\libatk-1.0-0.dll"
   RMDir "$INSTDIR\bin"
   RMDir "$INSTDIR\lib"
+
+  Delete /rebootok "$INSTDIR\deps.txt"
   RMDir "$INSTDIR"
 SectionEnd
 
-- 
1.7.10.4



From marcandre.lureau at gmail.com  Tue Jul 24 12:23:11 2012
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Tue, 24 Jul 2012 14:23:11 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 4/4] windows: add debug-helper
In-Reply-To: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <1343132591-1039-4-git-send-email-marcandre.lureau@gmail.com>

This is a simple program that will set some debug variable, and run
gdb and wait until it finished. This makes it possible to debug
"remote-viewer --spice-controller" easily, by setting the necessary
variables and keeping the parent process running (the activex whatches
its death)
---
 src/Makefile.am    |    5 ++++
 src/debug-helper.c |   70 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 75 insertions(+)
 create mode 100644 src/debug-helper.c

diff --git a/src/Makefile.am b/src/Makefile.am
index 8f9339c..a3c1b4d 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -155,6 +155,11 @@ virt-viewer_rc.$(OBJEXT): $(VIRT_VIEWER_RES) $(ICONDIR)/virt-viewer.ico
 		-i $< -o $@
 LDADD += virt-viewer_rc.$(OBJEXT)
 MAINTAINERCLEANFILES += virt-viewer_rc.$(OBJEXT)
+
+bin_PROGRAMS += debug-helper
+debug_helper_SOURCES = debug-helper.c
+debug_helper_LDFLAGS = $(GLIB2_LIBS) -Wl,--subsystem,windows
+debug_helper_CFLAGS = $(GLIB2_CFLAGS)
 endif
 
 -include $(top_srcdir)/git.mk
diff --git a/src/debug-helper.c b/src/debug-helper.c
new file mode 100644
index 0000000..754f7cf
--- /dev/null
+++ b/src/debug-helper.c
@@ -0,0 +1,70 @@
+#include <windows.h>
+#include <stdio.h>
+#include <conio.h>
+#include <process.h>
+#include <glib.h>
+
+static gchar*
+get_program_path(void)
+{
+    gchar *utf8_buf, *path = NULL;
+    wchar_t buf[MAX_PATH+1];
+
+    if (GetModuleFileNameW(GetModuleHandle (NULL), buf, G_N_ELEMENTS (buf)) > 0) {
+        utf8_buf = g_utf16_to_utf8 (buf, -1, NULL, NULL, NULL);
+        path = g_path_get_dirname (utf8_buf);
+        g_free (utf8_buf);
+    }
+
+    return path;
+}
+
+int
+main(int argc, char *argv[])
+{
+    char pipe[2048];
+    STARTUPINFO si = { 0, };
+    PROCESS_INFORMATION pi = { 0, };
+    gchar *program_path = get_program_path();
+    gchar *command;
+    int rv = 0;
+
+    argv[0] = "gdb -ex run --args";
+    command = g_strjoinv(" ", argv);
+
+    snprintf(pipe, sizeof(pipe), "\\\\.\\pipe\\SpiceController-%lu", GetCurrentProcessId());
+
+    SetEnvironmentVariable("SPICE_DEBUG", "1");
+    SetEnvironmentVariable("G_MESSAGES_DEBUG", "all");
+    SetEnvironmentVariable("SPICE_XPI_NAMEDPIPE", pipe);
+
+    si.cb = sizeof(si);
+    if (!CreateProcess(NULL,
+                       command,
+                       NULL,
+                       NULL,
+                       FALSE,
+                       0,
+                       NULL,
+                       program_path,
+                       &si,
+                       &pi)) {
+        printf("CreateProcess failed (%ld).\n", GetLastError());
+        rv = 1;
+        goto end;
+    }
+
+    // Wait until child process exits
+    WaitForSingleObject(pi.hProcess, INFINITE);
+
+    // Close process and thread handles
+    CloseHandle(pi.hProcess);
+    CloseHandle(pi.hThread);
+
+end:
+    g_free(program_path);
+    g_free(command);
+
+    exit(rv);
+    return rv;
+}
-- 
1.7.10.4



From eacosta at yahoo.com  Tue Jul 24 00:07:12 2012
From: eacosta at yahoo.com (Erasmo Acosta)
Date: Mon, 23 Jul 2012 17:07:12 -0700
Subject: [virt-tools-list] MacVTap questions
Message-ID: <9589B56C-4A32-4B6C-BDD0-D0767126B95F@yahoo.com>

Hello,

	I'm new to VMM. Mostly trying to get away from the VMWare player because it has horrible support for the latest Linux Kernels on Fedora. What I'm trying to do is to have the Guest VM, running Windows XP, mount SAMBA Shares from the HOST and then offer Video transcoding and streaming to iOS devices in my house. Both the Default NAT Interface and the MacvTap interface solve part of the problem, but none solves the entire problem. Can someone please comment?

Thank You,
Erasmo.



From crobinso at redhat.com  Tue Jul 24 15:53:34 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Tue, 24 Jul 2012 11:53:34 -0400
Subject: [virt-tools-list] MacVTap questions
In-Reply-To: <9589B56C-4A32-4B6C-BDD0-D0767126B95F@yahoo.com>
References: <9589B56C-4A32-4B6C-BDD0-D0767126B95F@yahoo.com>
Message-ID: <500EC4FE.90909@redhat.com>

On 07/23/2012 08:07 PM, Erasmo Acosta wrote:
> Hello,
> 
> 	I'm new to VMM. Mostly trying to get away from the VMWare player because it has horrible support for the latest Linux Kernels on Fedora. What I'm trying to do is to have the Guest VM, running Windows XP, mount SAMBA Shares from the HOST and then offer Video transcoding and streaming to iOS devices in my house. Both the Default NAT Interface and the MacvTap interface solve part of the problem, but none solves the entire problem. Can someone please comment?
> 


If you are talking solely about the network issues, you want to set up a
bridge on your host machine and share that with the guest. The guest will get
an IP address like any other machine on your network.

http://wiki.libvirt.org/page/Networking#Bridged_networking_.28aka_.22shared_physical_device.22.29

- Cole




From cfergeau at redhat.com  Tue Jul 24 16:43:46 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 24 Jul 2012 18:43:46 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 1/4] win32: there is no
 CONERR$, only CONOUT$
In-Reply-To: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120724164345.GP2552@teriyaki.cdg.redhat.com>

ACK
On Tue, Jul 24, 2012 at 02:23:08PM +0200, Marc-Andr? Lureau wrote:
> ---
>  src/remote-viewer-main.c      |    2 +-
>  src/windows-cmdline-wrapper.c |    2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/src/remote-viewer-main.c b/src/remote-viewer-main.c
> index 156bd0b..aee90ea 100644
> --- a/src/remote-viewer-main.c
> +++ b/src/remote-viewer-main.c
> @@ -231,7 +231,7 @@ main(int argc, char **argv)
>      if (AttachConsole(ATTACH_PARENT_PROCESS) != 0) {
>          freopen("CONIN$", "r", stdin);
>          freopen("CONOUT$", "w", stdout);
> -        freopen("CONERR$", "w", stderr);
> +        freopen("CONOUT$", "w", stderr);
>          dup2(fileno(stdin), STDIN_FILENO);
>          dup2(fileno(stdout), STDOUT_FILENO);
>          dup2(fileno(stderr), STDERR_FILENO);
> diff --git a/src/windows-cmdline-wrapper.c b/src/windows-cmdline-wrapper.c
> index 510763b..704e2d7 100644
> --- a/src/windows-cmdline-wrapper.c
> +++ b/src/windows-cmdline-wrapper.c
> @@ -30,7 +30,7 @@
>   *   if (AttachConsole(ATTACH_PARENT_PROCESS) != 0) {
>   *       freopen("CONIN$", "r", stdin);
>   *       freopen("CONOUT$", "w", stdout);
> - *       freopen("CONERR$", "w", stderr);
> + *       freopen("CONOUT$", "w", stderr);
>   *       dup2(fileno(stdin), STDIN_FILENO);
>   *       dup2(fileno(stdout), STDOUT_FILENO);
>   *       dup2(fileno(stderr), STDERR_FILENO);
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120724/6e4edcbb/attachment.sig>

From cfergeau at redhat.com  Tue Jul 24 16:44:05 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 24 Jul 2012 18:44:05 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 2/4] nsis: update required
 files and libraries
In-Reply-To: <1343132591-1039-2-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
	<1343132591-1039-2-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120724164404.GQ2552@teriyaki.cdg.redhat.com>

ACK

On Tue, Jul 24, 2012 at 02:23:09PM +0200, Marc-Andr? Lureau wrote:
> ---
>  data/virt-viewer.nsis.in |   46 +++++++++++++++++++++++++++++-----------------
>  1 file changed, 29 insertions(+), 17 deletions(-)
> 
> diff --git a/data/virt-viewer.nsis.in b/data/virt-viewer.nsis.in
> index e438159..f221a20 100644
> --- a/data/virt-viewer.nsis.in
> +++ b/data/virt-viewer.nsis.in
> @@ -66,6 +66,7 @@ Section "VirtViewer"
>    File "@prefix@/bin/libcairo-2.dll"
>    File "@prefix@/bin/libcelt051-0.dll"
>    File "@prefix@/bin/libcrypto-10.dll"
> +  File "@prefix@/bin/libexpat-1.dll"
>    File "@prefix@/bin/libffi-6.dll"
>    File "@prefix@/bin/libfontconfig-1.dll"
>    File "@prefix@/bin/libfreetype-6.dll"
> @@ -99,8 +100,8 @@ Section "VirtViewer"
>    File "@prefix@/bin/libpangowin32-1.0-0.dll"
>    File "@prefix@/bin/libpixman-1-0.dll"
>    File "@prefix@/bin/libpng15-15.dll"
> -  File "@prefix@/bin/libspice-client-glib-2.0-1.dll"
> -  File "@prefix@/bin/libspice-client-gtk-2.0-1.dll"
> +  File "@prefix@/bin/libspice-client-glib-2.0-8.dll"
> +  File "@prefix@/bin/libspice-client-gtk-2.0-4.dll"
>    File "@prefix@/bin/libspice-controller-0.dll"
>    File "@prefix@/bin/libssl-10.dll"
>    File "@prefix@/bin/libtasn1-3.dll"
> @@ -128,6 +129,7 @@ Section "VirtViewer"
>    File /oname=remote-viewer.com "${DESTDIR}@prefix@/bin/windows-cmdline-wrapper.exe"
>    File "${DESTDIR}@prefix@/bin/remote-viewer.exe"
>    File "${DESTDIR}@prefix@/bin/virt-viewer.exe"
> +  File "@prefix@/bin/readline.dll"
>    File "@prefix@/bin/zlib1.dll"
>  
>    SetOutPath "$INSTDIR\etc"
> @@ -212,15 +214,11 @@ Section "Uninstall"
>    DeleteRegKey HKCU "${REG_UNINSTALL}"
>    DeleteRegKey HKCU "${REG_IEPOLICY}"
>  
> -  RMDir /r "$INSTDIR\lib\pango"
> -
> -  RMDir /r "$INSTDIR\lib\gtk-2.0"
> -
> -  RMDir /r "$INSTDIR\share\icons\hicolor"
>  
> -  Delete /rebootok "$INSTDIR\share\virt-viewer\ui\*.xml"
> -  Delete /rebootok "$INSTDIR\share\themes\MS-Windows\gtk-2.0\gtkrc"
> +  RMDir /r "$INSTDIR\share"
>  
> +  RMDir /r "$INSTDIR\lib\pango"
> +  RMDir /r "$INSTDIR\lib\gtk-2.0"
>    Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstdirectsoundsrc.dll"
>    Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstdirectsoundsink.dll"
>    Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstcoreelements.dll"
> @@ -231,18 +229,19 @@ Section "Uninstall"
>    Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstaudioconvert.dll"
>    Delete /rebootok "$INSTDIR\lib\gstreamer-0.10\libgstapp.dll"
>    RMDir "$INSTDIR\lib\gstreamer-0.10"
> -
>    RMDir /r "$INSTDIR\lib\gdk-pixbuf-2.0"
>  
>    RMDir /r "$INSTDIR\etc\pango"
> -
>    RMDir /r "$INSTDIR\etc\gtk-2.0"
> -
>    RMDir /r "$INSTDIR\etc\fonts"
>  
>    Delete /rebootok "$INSTDIR\bin\remote-viewer.com"
>    Delete /rebootok "$INSTDIR\bin\remote-viewer.exe"
> +  Delete /rebootok "$INSTDIR\bin\virt-viewer.exe"
> +  Delete /rebootok "$INSTDIR\bin\virsh.exe"
>    Delete /rebootok "$INSTDIR\bin\pango-querymodules.exe"
> +  Delete /rebootok "$INSTDIR\bin\virt-pki-validate"
> +  Delete /rebootok "$INSTDIR\bin\virt-xml-validate"
>    Delete /rebootok "$INSTDIR\bin\libpangoft2-1.0-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libgdk_pixbuf-2.0-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libgdk-win32-2.0-0.dll"
> @@ -258,16 +257,15 @@ Section "Uninstall"
>    Delete /rebootok "$INSTDIR\bin\gst-inspect-0.10.exe"
>    Delete /rebootok "$INSTDIR\bin\gst-feedback.exe"
>    Delete /rebootok "$INSTDIR\bin\gst-discoverer-0.10.exe"
> -  RMDir "$INSTDIR\bin"
> -
> -  Delete /rebootok "$INSTDIR\bin\libffi-6.dll"
>    Delete /rebootok "$INSTDIR\bin\zlib1.dll"
> +  Delete /rebootok "$INSTDIR\bin\readline.dll"
> +  Delete /rebootok "$INSTDIR\bin\libffi-6.dll"
>    Delete /rebootok "$INSTDIR\bin\libp11-kit-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libxml2-2.dll"
>    Delete /rebootok "$INSTDIR\bin\libtasn1-3.dll"
>    Delete /rebootok "$INSTDIR\bin\libssl-10.dll"
> -  Delete /rebootok "$INSTDIR\bin\libspice-client-gtk-2.0-1.dll"
> -  Delete /rebootok "$INSTDIR\bin\libspice-client-glib-2.0-1.dll"
> +  Delete /rebootok "$INSTDIR\bin\libspice-client-gtk-2.0-4.dll"
> +  Delete /rebootok "$INSTDIR\bin\libspice-client-glib-2.0-8.dll"
>    Delete /rebootok "$INSTDIR\bin\libspice-controller-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libpng15-15.dll"
>    Delete /rebootok "$INSTDIR\bin\libpixman-1-0.dll"
> @@ -278,6 +276,16 @@ Section "Uninstall"
>    Delete /rebootok "$INSTDIR\bin\liborc-0.4-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libjpeg-62.dll"
>    Delete /rebootok "$INSTDIR\bin\libintl-8.dll"
> +  Delete /rebootok "$INSTDIR\bin\libvirt-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libcurl-4.dll"
> +  Delete /rebootok "$INSTDIR\bin\libidn-11.dll"
> +  Delete /rebootok "$INSTDIR\bin\libssh2-1.dll"
> +  Delete /rebootok "$INSTDIR\bin\libgvnc-1.0-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libgtk-vnc-1.0-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libgpg-error-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libportablexdr-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libvirt-qemu-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libssp-0.dll"
>    Delete /rebootok "$INSTDIR\bin\iconv.dll"
>    Delete /rebootok "$INSTDIR\bin\libgtk-win32-2.0-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libgthread-2.0-0.dll"
> @@ -291,6 +299,8 @@ Section "Uninstall"
>    Delete /rebootok "$INSTDIR\bin\libgstaudio-0.10-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libgstapp-0.10-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libgobject-2.0-0.dll"
> +  Delete /rebootok "$INSTDIR\bin\libgnutls-26.dll"
> +  Delete /rebootok "$INSTDIR\bin\libgcrypt-11.dll"
>    Delete /rebootok "$INSTDIR\bin\libgmodule-2.0-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libglib-2.0-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libgio-2.0-0.dll"
> @@ -298,11 +308,13 @@ Section "Uninstall"
>    Delete /rebootok "$INSTDIR\bin\libgdk-win32-2.0-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libfreetype-6.dll"
>    Delete /rebootok "$INSTDIR\bin\libfontconfig-1.dll"
> +  Delete /rebootok "$INSTDIR\bin\libexpat-1.dll"
>    Delete /rebootok "$INSTDIR\bin\libcrypto-10.dll"
>    Delete /rebootok "$INSTDIR\bin\libcelt051-0.dll"
>    Delete /rebootok "$INSTDIR\bin\libcairo-2.dll"
>    Delete /rebootok "$INSTDIR\bin\libatk-1.0-0.dll"
>    RMDir "$INSTDIR\bin"
> +  RMDir "$INSTDIR\lib"
>    RMDir "$INSTDIR"
>  SectionEnd
>  
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120724/84f0c5e5/attachment.sig>

From cfergeau at redhat.com  Tue Jul 24 19:04:49 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 24 Jul 2012 21:04:49 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 3/4] nsis: add deps.txt
 file in installer
In-Reply-To: <1343132591-1039-3-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
	<1343132591-1039-3-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120724190449.GS2552@teriyaki.cdg.redhat.com>

On Tue, Jul 24, 2012 at 02:23:10PM +0200, Marc-Andr? Lureau wrote:
> This helps track package version that were used during the
> build of Windows installer. It's not ideal, but make up the
> lack of package management on windows
> ---
>  Makefile.am              |    6 ++++++
>  data/Makefile.am         |   13 +++++++++++++
>  data/virt-viewer.nsis.in |    5 +++++
>  3 files changed, 24 insertions(+)
> 
> diff --git a/Makefile.am b/Makefile.am
> index f307c06..2407b1b 100644
> --- a/Makefile.am
> +++ b/Makefile.am
> @@ -57,4 +57,10 @@ gen-ChangeLog:
>  	  mv $(distdir)/cl-t $(distdir)/ChangeLog;		\
>  	fi
>  
> +if OS_WIN32
> +.PHONY: nsis
> +nsis:
> +	make -C $(builddir)/data virt-viewer-$(VERSION).exe
> +endif
> +
>  -include $(top_srcdir)/git.mk
> diff --git a/data/Makefile.am b/data/Makefile.am
> index ea2bb0a..38ffdb2 100644
> --- a/data/Makefile.am
> +++ b/data/Makefile.am
> @@ -5,4 +5,17 @@ EXTRA_DIST =					\
>  	virt-viewer.nsis.in			\
>  	$(NULL)
>  
> +if OS_WIN32
> +deps.txt:
> +	$(AM_V_GEN)rpm -qa | grep $(host_os) | sort | unix2dos > $@

If $(host_os) is the prefix of the package names you want to match, rpm -qa
"$(host_os)*" is more efficient iirc.

Looks good otherwise even if the deps.txt part is indeed a bit ugly ;)

Christophe

> +
> +virt-viewer-$(VERSION).exe: virt-viewer.nsis deps.txt
> +	$(AM_V_GEN)DESTDIR=`mktemp -d` && \
> +	make -C $(top_builddir) install DESTDIR=$$DESTDIR >/dev/null && \
> +	makensis -NOCD -DDESTDIR=$$DESTDIR $< >/dev/null && \
> +	rm -rf $$DESTDIR
> +
> +CLEANFILES = deps.txt virt-viewer-$(VERSION).exe
> +endif
> +
>  -include $(top_srcdir)/git.mk
> diff --git a/data/virt-viewer.nsis.in b/data/virt-viewer.nsis.in
> index f221a20..98ce83a 100644
> --- a/data/virt-viewer.nsis.in
> +++ b/data/virt-viewer.nsis.in
> @@ -49,6 +49,9 @@ Section "VirtViewer"
>    SectionIn RO
>    WriteRegStr HKCU "Software\virt-viewer" "" $INSTDIR
>  
> +  SetOutPath "$INSTDIR"
> +  File "deps.txt"
> +
>    SetOutPath "$INSTDIR\bin"
>    File "@prefix@/bin/gst-discoverer-0.10.exe"
>    File "@prefix@/bin/gst-feedback.exe"
> @@ -315,6 +318,8 @@ Section "Uninstall"
>    Delete /rebootok "$INSTDIR\bin\libatk-1.0-0.dll"
>    RMDir "$INSTDIR\bin"
>    RMDir "$INSTDIR\lib"
> +
> +  Delete /rebootok "$INSTDIR\deps.txt"
>    RMDir "$INSTDIR"
>  SectionEnd
>  
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120724/6d0ed832/attachment.sig>

From cfergeau at redhat.com  Tue Jul 24 19:59:39 2012
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 24 Jul 2012 21:59:39 +0200
Subject: [virt-tools-list] [PATCH virt-viewer 4/4] windows: add
 debug-helper
In-Reply-To: <1343132591-1039-4-git-send-email-marcandre.lureau@gmail.com>
References: <1343132591-1039-1-git-send-email-marcandre.lureau@gmail.com>
	<1343132591-1039-4-git-send-email-marcandre.lureau@gmail.com>
Message-ID: <20120724195939.GT2552@teriyaki.cdg.redhat.com>

On Tue, Jul 24, 2012 at 02:23:11PM +0200, Marc-Andr? Lureau wrote:
> This is a simple program that will set some debug variable, and run
> gdb and wait until it finished. This makes it possible to debug
> "remote-viewer --spice-controller" easily, by setting the necessary
> variables and keeping the parent process running (the activex whatches
Typo:                                                           ^^^^^^^^
                                                                watches

Can you explain in the commit log how it will be launched, and with which
arguments it will get started?
Looks good otherwise

Christophe


> ---
>  src/Makefile.am    |    5 ++++
>  src/debug-helper.c |   70 ++++++++++++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 75 insertions(+)
>  create mode 100644 src/debug-helper.c
> 
> diff --git a/src/Makefile.am b/src/Makefile.am
> index 8f9339c..a3c1b4d 100644
> --- a/src/Makefile.am
> +++ b/src/Makefile.am
> @@ -155,6 +155,11 @@ virt-viewer_rc.$(OBJEXT): $(VIRT_VIEWER_RES) $(ICONDIR)/virt-viewer.ico
>  		-i $< -o $@
>  LDADD += virt-viewer_rc.$(OBJEXT)
>  MAINTAINERCLEANFILES += virt-viewer_rc.$(OBJEXT)
> +
> +bin_PROGRAMS += debug-helper
> +debug_helper_SOURCES = debug-helper.c
> +debug_helper_LDFLAGS = $(GLIB2_LIBS) -Wl,--subsystem,windows
> +debug_helper_CFLAGS = $(GLIB2_CFLAGS)
>  endif
>  
>  -include $(top_srcdir)/git.mk
> diff --git a/src/debug-helper.c b/src/debug-helper.c
> new file mode 100644
> index 0000000..754f7cf
> --- /dev/null
> +++ b/src/debug-helper.c
> @@ -0,0 +1,70 @@
> +#include <windows.h>
> +#include <stdio.h>
> +#include <conio.h>
> +#include <process.h>
> +#include <glib.h>
> +
> +static gchar*
> +get_program_path(void)
> +{
> +    gchar *utf8_buf, *path = NULL;
> +    wchar_t buf[MAX_PATH+1];
> +
> +    if (GetModuleFileNameW(GetModuleHandle (NULL), buf, G_N_ELEMENTS (buf)) > 0) {
> +        utf8_buf = g_utf16_to_utf8 (buf, -1, NULL, NULL, NULL);
> +        path = g_path_get_dirname (utf8_buf);
> +        g_free (utf8_buf);
> +    }
> +
> +    return path;
> +}
> +
> +int
> +main(int argc, char *argv[])
> +{
> +    char pipe[2048];
> +    STARTUPINFO si = { 0, };
> +    PROCESS_INFORMATION pi = { 0, };
> +    gchar *program_path = get_program_path();
> +    gchar *command;
> +    int rv = 0;
> +
> +    argv[0] = "gdb -ex run --args";
> +    command = g_strjoinv(" ", argv);
> +
> +    snprintf(pipe, sizeof(pipe), "\\\\.\\pipe\\SpiceController-%lu", GetCurrentProcessId());
> +
> +    SetEnvironmentVariable("SPICE_DEBUG", "1");
> +    SetEnvironmentVariable("G_MESSAGES_DEBUG", "all");
> +    SetEnvironmentVariable("SPICE_XPI_NAMEDPIPE", pipe);
> +
> +    si.cb = sizeof(si);
> +    if (!CreateProcess(NULL,
> +                       command,
> +                       NULL,
> +                       NULL,
> +                       FALSE,
> +                       0,
> +                       NULL,
> +                       program_path,
> +                       &si,
> +                       &pi)) {
> +        printf("CreateProcess failed (%ld).\n", GetLastError());
> +        rv = 1;
> +        goto end;
> +    }
> +
> +    // Wait until child process exits
> +    WaitForSingleObject(pi.hProcess, INFINITE);
> +
> +    // Close process and thread handles
> +    CloseHandle(pi.hProcess);
> +    CloseHandle(pi.hThread);
> +
> +end:
> +    g_free(program_path);
> +    g_free(command);
> +
> +    exit(rv);
> +    return rv;
> +}
> -- 
> 1.7.10.4
> 
> _______________________________________________
> virt-tools-list mailing list
> virt-tools-list at redhat.com
> https://www.redhat.com/mailman/listinfo/virt-tools-list
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120724/293be429/attachment.sig>

From fabiano at fidencio.org  Tue Jul 24 20:05:12 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:12 +0200
Subject: [virt-tools-list] [PATCH 0/6] Set of patches to install-script
	stuffs
Message-ID: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>

Here is a set of patches to add support to:
*generate output files
*desktop profile for fedora

It was tested and is working on Boxes.
Patches to add windows support will be sent soon.

Fabiano Fid?ncio (6):
  Add "hostname" prop to install-script-config
  Don't return a gboolean when a gchar * is expected
  Change install script output (tool)
  Add functions to know if a config can/must be set
  Add function to fetch console keyboard layout
  Add desktop profile for Fedora 15 and newer

 data/install-scripts/fedora.xml           |  165 ++++++++++++-
 data/install-scripts/windows-sif.xml      |    7 +-
 data/install-scripts/windows-unattend.xml |   11 +-
 data/oses/fedora.xml                      |    4 +
 data/schemas/libosinfo.rng                |   13 ++
 osinfo/libosinfo.syms                     |    8 +
 osinfo/osinfo_install_config.c            |   13 ++
 osinfo/osinfo_install_config.h            |    6 +
 osinfo/osinfo_install_script.c            |  361 ++++++++++++++++++++++++++++-
 osinfo/osinfo_install_script.h            |   35 +++
 osinfo/osinfo_install_scriptlist.c        |   15 ++
 osinfo/osinfo_install_scriptlist.h        |    1 +
 osinfo/osinfo_loader.c                    |   28 +++
 tools/osinfo-install-script.c             |   54 +++--
 14 files changed, 688 insertions(+), 33 deletions(-)

-- 
1.7.10.4



From fabiano at fidencio.org  Tue Jul 24 20:05:13 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:13 +0200
Subject: [virt-tools-list] [PATCH 1/6] Add "hostname" prop to
	install-script-config
In-Reply-To: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1343160318-20796-2-git-send-email-fabiano@fidencio.org>

It is desirable to configure the hostname of the new virtual
machine during initial automated provisioning
---
 osinfo/libosinfo.syms          |    1 +
 osinfo/osinfo_install_config.c |   13 +++++++++++++
 osinfo/osinfo_install_config.h |    6 ++++++
 3 files changed, 20 insertions(+)

diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index f22ad37..2f90183 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -263,6 +263,7 @@ LIBOSINFO_0.2.0 {
 	osinfo_install_config_set_user_realname;
 	osinfo_install_config_set_user_administrator;
 	osinfo_install_config_set_user_autologin;
+	osinfo_install_config_set_hostname;
 	osinfo_install_script_get_type;
 	osinfo_install_script_new;
 	osinfo_install_script_new_data;
diff --git a/osinfo/osinfo_install_config.c b/osinfo/osinfo_install_config.c
index 80643b1..eb50680 100644
--- a/osinfo/osinfo_install_config.c
+++ b/osinfo/osinfo_install_config.c
@@ -314,6 +314,19 @@ const gchar *osinfo_install_config_get_reg_product_key(OsinfoInstallConfig *conf
                                          OSINFO_INSTALL_CONFIG_PROP_REG_PRODUCTKEY);
 }
 
+void osinfo_install_config_set_hostname(OsinfoInstallConfig *config,
+                                        const gchar *hostname)
+{
+    osinfo_entity_set_param(OSINFO_ENTITY(config),
+                            OSINFO_INSTALL_CONFIG_PROP_HOSTNAME,
+                            hostname);
+}
+
+const gchar *osinfo_install_config_get_hostname(OsinfoInstallConfig *config)
+{
+    return osinfo_entity_get_param_value(OSINFO_ENTITY(config),
+                                         OSINFO_INSTALL_CONFIG_PROP_HOSTNAME);
+}
 
 
 /*
diff --git a/osinfo/osinfo_install_config.h b/osinfo/osinfo_install_config.h
index 2d1c065..32fe370 100644
--- a/osinfo/osinfo_install_config.h
+++ b/osinfo/osinfo_install_config.h
@@ -54,6 +54,8 @@
 #define OSINFO_INSTALL_CONFIG_PROP_REG_PASSWORD   "reg-password"
 #define OSINFO_INSTALL_CONFIG_PROP_REG_PRODUCTKEY "reg-product-key"
 
+#define OSINFO_INSTALL_CONFIG_PROP_HOSTNAME "hostname"
+
 
 typedef struct _OsinfoInstallConfig        OsinfoInstallConfig;
 typedef struct _OsinfoInstallConfigClass   OsinfoInstallConfigClass;
@@ -146,6 +148,10 @@ void osinfo_install_config_set_reg_product_key(OsinfoInstallConfig *config,
                                                const gchar *key);
 const gchar *osinfo_install_config_get_reg_product_key(OsinfoInstallConfig *config);
 
+void osinfo_install_config_set_hostname(OsinfoInstallConfig *config,
+                                        const gchar *hostname);
+const gchar *osinfo_install_config_get_hostname(OsinfoInstallConfig *config);
+
 
 #endif /* __OSINFO_INSTALL_CONFIG_H__ */
 /*
-- 
1.7.10.4



From fabiano at fidencio.org  Tue Jul 24 20:05:14 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:14 +0200
Subject: [virt-tools-list] [PATCH 2/6] Don't return a gboolean when a gchar
	* is expected
In-Reply-To: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1343160318-20796-3-git-send-email-fabiano@fidencio.org>

---
 osinfo/osinfo_install_script.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 7bf1e2e..1f437a1 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -627,10 +627,10 @@ gchar *osinfo_install_script_generate_finish(OsinfoInstallScript *script,
 {
     GSimpleAsyncResult *simple = G_SIMPLE_ASYNC_RESULT(res);
 
-    g_return_val_if_fail(error == NULL || *error == NULL, FALSE);
+    g_return_val_if_fail(error == NULL || *error == NULL, NULL);
 
     if (g_simple_async_result_propagate_error(simple, error))
-        return FALSE;
+        return NULL;
 
     return g_simple_async_result_get_op_res_gpointer(simple);
 }
-- 
1.7.10.4



From fabiano at fidencio.org  Tue Jul 24 20:05:15 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:15 +0200
Subject: [virt-tools-list] [PATCH 3/6] Change install script output (tool)
In-Reply-To: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1343160318-20796-4-git-send-email-fabiano@fidencio.org>

Now, instead of return a string, the osinfo-install-script tool writes
that string in a file, passed as argument, that will be into a dir,
also passed as argument.

If the filename argument is NULL, the output files will be written as
set in data/install-scripts/*.xml (in filename attribute from template
element):
    - Linuxes: fedora.ks
    - Windows 2k3r2 and older: windows.sif
    - Windows 2k8 and newer: windows.xml

If the dirname argument is NULL, the output files will be written in the
current directory.

It will be used to create, easily, multiple scripts, as used in:
http://bugzilla-attachments.gnome.org/attachment.cgi?id=214681
---
 data/install-scripts/fedora.xml           |    2 +-
 data/install-scripts/windows-sif.xml      |    2 +-
 data/install-scripts/windows-unattend.xml |    2 +-
 data/schemas/libosinfo.rng                |    1 +
 osinfo/libosinfo.syms                     |    4 +
 osinfo/osinfo_install_script.c            |  189 ++++++++++++++++++++++++++++-
 osinfo/osinfo_install_script.h            |   25 ++++
 osinfo/osinfo_install_scriptlist.c        |   15 +++
 osinfo/osinfo_install_scriptlist.h        |    1 +
 osinfo/osinfo_loader.c                    |    9 ++
 tools/osinfo-install-script.c             |   54 +++++----
 11 files changed, 273 insertions(+), 31 deletions(-)

diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
index b4ad72d..338a570 100644
--- a/data/install-scripts/fedora.xml
+++ b/data/install-scripts/fedora.xml
@@ -1,7 +1,7 @@
 <libosinfo version="0.0.1">
   <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
     <profile>jeos</profile>
-    <template>
+    <template filename="fedora.ks">
       <xsl:stylesheet
         xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
         version="1.0">
diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
index 29a0eae..46b24ae 100644
--- a/data/install-scripts/windows-sif.xml
+++ b/data/install-scripts/windows-sif.xml
@@ -1,7 +1,7 @@
 <libosinfo version="0.0.1">
   <install-script id='http://microsoft.com/windows/sif'>
     <profile>jeos</profile>
-    <template>
+    <template filename="windows.sif">
       <xsl:stylesheet
         xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
         version="1.0">
diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
index 9828b34..20e3e23 100644
--- a/data/install-scripts/windows-unattend.xml
+++ b/data/install-scripts/windows-unattend.xml
@@ -1,7 +1,7 @@
 <libosinfo version="0.0.1">
   <install-script id='http://microsoft.com/windows/unattend'>
     <profile>jeos</profile>
-    <template>
+    <template filename="windows.xml">
       <xsl:stylesheet
         xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
         version="1.0">
diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
index a5f527d..7c8d7f7 100644
--- a/data/schemas/libosinfo.rng
+++ b/data/schemas/libosinfo.rng
@@ -414,6 +414,7 @@
         <text/>
       </element>
       <element name='template'>
+        <attribute name="filename"/>
         <choice>
           <group>
             <attribute name="uri"/>
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 2f90183..5a7de7b 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -264,6 +264,7 @@ LIBOSINFO_0.2.0 {
 	osinfo_install_config_set_user_administrator;
 	osinfo_install_config_set_user_autologin;
 	osinfo_install_config_set_hostname;
+	osinfo_install_script_set_output_filename;
 	osinfo_install_script_get_type;
 	osinfo_install_script_new;
 	osinfo_install_script_new_data;
@@ -271,6 +272,8 @@ LIBOSINFO_0.2.0 {
 	osinfo_install_script_generate;
 	osinfo_install_script_generate_async;
 	osinfo_install_script_generate_finish;
+	osinfo_install_script_generate_output;
+	osinfo_install_script_generate_output_async;
 	osinfo_install_script_get_profile;
 	osinfo_install_script_get_uri;
 	osinfo_install_scriptlist_new;
@@ -279,6 +282,7 @@ LIBOSINFO_0.2.0 {
 	osinfo_install_scriptlist_new_intersection;
 	osinfo_install_scriptlist_new_copy;
 	osinfo_install_scriptlist_get_type;
+	osinfo_install_scriptlist_get_by_profile;
 	osinfo_db_get_install_script;
 	osinfo_db_add_install_script;
 	osinfo_db_get_install_script_list;
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 1f437a1..b2d7351 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -56,6 +56,7 @@ enum {
     PROP_TEMPLATE_DATA,
     PROP_PROFILE,
     PROP_PRODUCT_KEY_FORMAT,
+    PROP_OUTPUT_FILENAME,
 };
 
 
@@ -93,6 +94,14 @@ osinfo_os_set_property(GObject    *object,
                                     data);
         break;
 
+    case PROP_OUTPUT_FILENAME:
+        data = g_value_get_string(value);
+        if (data)
+            osinfo_entity_set_param(OSINFO_ENTITY(script),
+                                    OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME,
+                                    data);
+        break;
+
     default:
         /* We don't have any other property... */
         G_OBJECT_WARN_INVALID_PROPERTY_ID (object, property_id, pspec);
@@ -129,6 +138,11 @@ osinfo_os_get_property(GObject    *object,
                            osinfo_install_script_get_product_key_format(script));
         break;
 
+    case PROP_OUTPUT_FILENAME:
+        g_value_set_string(value,
+                           osinfo_install_script_get_output_filename(script));
+        break;
+
     default:
         /* We don't have any other property... */
         G_OBJECT_WARN_INVALID_PROPERTY_ID (object, property_id, pspec);
@@ -210,6 +224,19 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
                                     PROP_PROFILE,
                                     pspec);
 
+    pspec = g_param_spec_string("output-filename",
+                                "Output Filename",
+                                "Output filename for the script",
+                                NULL /* default value */,
+                                G_PARAM_READABLE |
+                                G_PARAM_WRITABLE |
+                                G_PARAM_STATIC_NAME |
+                                G_PARAM_STATIC_NICK |
+                                G_PARAM_STATIC_BLURB);
+    g_object_class_install_property(g_klass,
+                                    PROP_OUTPUT_FILENAME,
+                                    pspec);
+
     g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
 }
 
@@ -218,7 +245,6 @@ osinfo_install_script_init (OsinfoInstallScript *list)
 {
     OsinfoInstallScriptPrivate *priv;
     list->priv = priv = OSINFO_INSTALL_SCRIPT_GET_PRIVATE(list);
-
 }
 
 
@@ -291,6 +317,25 @@ const gchar *osinfo_install_script_get_product_key_format(OsinfoInstallScript *s
                                          OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT);
 }
 
+void osinfo_install_script_set_output_filename(OsinfoInstallScript *script,
+                                               const gchar *filename)
+{
+    GValue value = G_VALUE_INIT;
+    g_value_init(&value, G_TYPE_STRING);
+    g_value_set_string(&value, filename);
+
+    osinfo_os_set_property(G_OBJECT (script),
+                           PROP_OUTPUT_FILENAME,
+                           &value,
+                           NULL);
+}
+
+const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *script)
+{
+    return osinfo_entity_get_param_value(OSINFO_ENTITY(script),
+                                         OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
+}
+
 struct OsinfoInstallScriptGenerate {
     GSimpleAsyncResult *res;
     OsinfoOs *os;
@@ -635,11 +680,18 @@ gchar *osinfo_install_script_generate_finish(OsinfoInstallScript *script,
     return g_simple_async_result_get_op_res_gpointer(simple);
 }
 
-
 struct OsinfoInstallScriptGenerateSync {
     GMainLoop *loop;
+    GCancellable *cancellable;
     GError *error;
+    OsinfoInstallScript *script;
+    OsinfoOs *os;
+    GFile *dir;
+    GFile *file;
     gchar *output;
+    gssize output_len;
+    gssize output_pos;
+    GFileOutputStream *stream;
 };
 
 static void osinfo_install_script_generate_done(GObject *src,
@@ -655,7 +707,6 @@ static void osinfo_install_script_generate_done(GObject *src,
     g_main_loop_quit(data->loop);
 }
 
-
 gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
                                       OsinfoOs *os,
                                       OsinfoInstallConfig *config,
@@ -665,7 +716,7 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
     GMainLoop *loop = g_main_loop_new(g_main_context_get_thread_default(),
                                       TRUE);
     struct OsinfoInstallScriptGenerateSync data = {
-        loop, NULL, NULL
+        loop, NULL, NULL, NULL, NULL, NULL, NULL, NULL, 0, 0, NULL
     };
 
     osinfo_install_script_generate_async(script,
@@ -686,6 +737,136 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
     return data.output;
 }
 
+void osinfo_install_script_generate_output_async(OsinfoInstallScript *script,
+                                                 OsinfoOs *os,
+                                                 OsinfoInstallConfig *config,
+                                                 GFile *output_dir,
+                                                 GCancellable *cancellable,
+                                                 GAsyncReadyCallback callback,
+                                                 gpointer user_data)
+{
+    osinfo_install_script_generate_async(script,
+                                         os,
+                                         config,
+                                         cancellable,
+                                         callback,
+                                         user_data);
+}
+
+gboolean osinfo_install_script_generate_output_finish(GFileOutputStream *stream,
+                                                      GAsyncResult *res,
+                                                      GError **error)
+{
+    return g_output_stream_close_finish (G_OUTPUT_STREAM (stream), res, error);
+}
+
+static void osinfo_install_script_generate_output_done(GObject *src,
+                                                       GAsyncResult *res,
+                                                       gpointer user_data)
+{
+    struct OsinfoInstallScriptGenerateSync *data = user_data;
+
+    osinfo_install_script_generate_output_finish(data->stream,
+                                                 res,
+                                                 &data->error);
+    g_main_loop_quit(data->loop);
+}
+
+static void osinfo_install_script_generate_output_write_file(GObject *src,
+                                                             GAsyncResult *res,
+                                                             gpointer user_data)
+{
+    struct OsinfoInstallScriptGenerateSync *data = user_data;
+    gssize size = 0;
+
+    if (data->stream == NULL)
+        data->stream = g_file_replace_finish(G_FILE (src), res, &data->error);
+    else
+        size = g_output_stream_write_finish(G_OUTPUT_STREAM (data->stream),
+                                                             res,
+                                                             &data->error);
+
+    g_assert_no_error(data->error);
+
+    data->output_pos += size;
+    if (data->output_pos < data->output_len) {
+        g_output_stream_write_async(G_OUTPUT_STREAM (data->stream),
+                                    data->output + data->output_pos,
+                                    data->output_len - data->output_pos,
+                                    G_PRIORITY_DEFAULT,
+                                    data->cancellable,
+                                    osinfo_install_script_generate_output_write_file,
+                                    data);
+
+    } else {
+        g_output_stream_close_async(G_OUTPUT_STREAM (data->stream),
+                                    G_PRIORITY_DEFAULT,
+                                    data->cancellable,
+                                    osinfo_install_script_generate_output_done,
+                                    data);
+    }
+}
+
+static void osinfo_install_script_generate_create_output_file(GObject *src,
+                                                              GAsyncResult *res,
+                                                              gpointer user_data)
+{
+    struct OsinfoInstallScriptGenerateSync *data = user_data;
+    const gchar *filename;
+
+    data->output =
+        osinfo_install_script_generate_finish(OSINFO_INSTALL_SCRIPT(src),
+                                              res,
+                                              &data->error);
+    data->output_len = strlen(data->output);
+
+    filename =
+        osinfo_install_script_get_output_filename(OSINFO_INSTALL_SCRIPT(src));
+
+    data->file = g_file_get_child(data->dir, filename);
+
+    g_file_replace_async(data->file,
+                         NULL,
+                         TRUE,
+                         G_FILE_CREATE_NONE,
+                         G_PRIORITY_DEFAULT,
+                         data->cancellable,
+                         osinfo_install_script_generate_output_write_file,
+                         data);
+
+}
+
+const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
+                                                   OsinfoOs *os,
+                                                   OsinfoInstallConfig *config,
+                                                   GFile *output_dir,
+                                                   GCancellable *cancellable,
+                                                   GError **error)
+{
+    GMainLoop *loop = g_main_loop_new(g_main_context_get_thread_default(),
+                                      TRUE);
+    struct OsinfoInstallScriptGenerateSync data = {
+        loop, cancellable, NULL, script, os, output_dir, NULL, NULL, 0, 0, NULL
+    };
+
+    osinfo_install_script_generate_output_async(script,
+                                                os,
+                                                config,
+                                                output_dir,
+                                                cancellable,
+                                                osinfo_install_script_generate_create_output_file,
+                                                &data);
+
+    if (g_main_loop_is_running(loop))
+        g_main_loop_run(loop);
+
+    if (data.error)
+        g_propagate_error(error, data.error);
+
+    g_main_loop_unref(loop);
+
+    return data.file;
+}
 
 /*
  * Local variables:
diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
index 00740c2..d75ec93 100644
--- a/osinfo/osinfo_install_script.h
+++ b/osinfo/osinfo_install_script.h
@@ -49,6 +49,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
 #define OSINFO_INSTALL_SCRIPT_PROP_TEMPLATE_DATA      "template-data"
 #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
 #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
+#define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
+
 
 /* object */
 struct _OsinfoInstallScript
@@ -85,6 +87,11 @@ const gchar *osinfo_install_script_get_profile(OsinfoInstallScript *script);
 
 const gchar *osinfo_install_script_get_product_key_format(OsinfoInstallScript *script);
 
+void osinfo_install_script_set_output_filename(OsinfoInstallScript *script,
+                                               const gchar *filename);
+
+const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *script);
+
 void osinfo_install_script_generate_async(OsinfoInstallScript *script,
                                           OsinfoOs *os,
                                           OsinfoInstallConfig *config,
@@ -102,6 +109,24 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
                                       GCancellable *cancellable,
                                       GError **error);
 
+void osinfo_install_script_generate_output_async(OsinfoInstallScript *script,
+                                                 OsinfoOs *os,
+                                                 OsinfoInstallConfig *config,
+                                                 GFile *output_dir,
+                                                 GCancellable *cancellable,
+                                                 GAsyncReadyCallback callback,
+                                                 gpointer user_data);
+
+gboolean osinfo_install_script_generate_output_finish(GFileOutputStream *stream,
+                                                      GAsyncResult *res,
+                                                      GError **error);
+
+const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
+                                                   OsinfoOs *os,
+                                                   OsinfoInstallConfig *config,
+                                                   GFile *output_dir,
+                                                   GCancellable *cancellable,
+                                                   GError **error);
 
 #endif /* __OSINFO_INSTALL_SCRIPT_H__ */
 /*
diff --git a/osinfo/osinfo_install_scriptlist.c b/osinfo/osinfo_install_scriptlist.c
index e5309b5..dfc830b 100644
--- a/osinfo/osinfo_install_scriptlist.c
+++ b/osinfo/osinfo_install_scriptlist.c
@@ -160,6 +160,21 @@ OsinfoInstallScriptList *osinfo_install_scriptlist_new_union(OsinfoInstallScript
     return newList;
 }
 
+OsinfoInstallScriptList *osinfo_install_scriptlist_get_by_profile(OsinfoInstallScriptList *source,
+                                                                  const gchar *profile)
+{
+    OsinfoInstallScriptList *newList;
+    OsinfoFilter *filter;
+
+    filter = osinfo_filter_new();
+    osinfo_filter_add_constraint(filter,
+                                 OSINFO_INSTALL_SCRIPT_PROP_PROFILE,
+                                 profile);
+
+    newList = osinfo_install_scriptlist_new_filtered(source, filter);
+    return newList;
+}
+
 /*
  * Local variables:
  *  indent-tabs-mode: nil
diff --git a/osinfo/osinfo_install_scriptlist.h b/osinfo/osinfo_install_scriptlist.h
index 39544d2..6407304 100644
--- a/osinfo/osinfo_install_scriptlist.h
+++ b/osinfo/osinfo_install_scriptlist.h
@@ -71,6 +71,7 @@ OsinfoInstallScriptList *osinfo_install_scriptlist_new_copy(OsinfoInstallScriptL
 OsinfoInstallScriptList *osinfo_install_scriptlist_new_filtered(OsinfoInstallScriptList *source, OsinfoFilter *filter);
 OsinfoInstallScriptList *osinfo_install_scriptlist_new_intersection(OsinfoInstallScriptList *sourceOne, OsinfoInstallScriptList *sourceTwo);
 OsinfoInstallScriptList *osinfo_install_scriptlist_new_union(OsinfoInstallScriptList *sourceOne, OsinfoInstallScriptList *sourceTwo);
+OsinfoInstallScriptList *osinfo_install_scriptlist_get_by_profile(OsinfoInstallScriptList *source, const gchar *profile);
 
 #endif /* __OSINFO_INSTALL_SCRIPTLIST_H__ */
 /*
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index f89dee0..31e442e 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -585,6 +585,15 @@ static void osinfo_loader_install_script(OsinfoLoader *loader,
                                 value);
     g_free(value);
 
+    value = osinfo_loader_string("string(./template/@filename)", ctxt, err);
+    if (error_is_set(err))
+        goto error;
+    if (value)
+        osinfo_entity_set_param(OSINFO_ENTITY(installScript),
+                                OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME,
+                                value);
+    g_free(value);
+
     osinfo_db_add_install_script(loader->priv->db, installScript);
 
     return;
diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
index c6aeae6..8ea69c0 100644
--- a/tools/osinfo-install-script.c
+++ b/tools/osinfo-install-script.c
@@ -26,7 +26,9 @@
 #include <osinfo/osinfo.h>
 #include <string.h>
 
-static gchar *profile;
+static const gchar *profile;
+static const gchar *output_dir;
+static const gchar *output_file;
 
 static OsinfoInstallConfig *config;
 
@@ -60,6 +62,14 @@ static GOptionEntry entries[] =
 {
     { "profile", 'p', 0, G_OPTION_ARG_STRING, (void*)&profile,
       "Install script profile", NULL, },
+    { "dir", 'd', 0, G_OPTION_ARG_STRING, (void*)&output_dir,
+      "Install script output directory", NULL, },
+    { "file", 'f', 0, G_OPTION_ARG_STRING, (void*)&output_file,
+      "Install script output filename\n"
+      "\t\t\t     The extension expected for each type of OS is:\n"
+      "\t\t\t       .ks: for linuxes\n"
+      "\t\t\t       .sif: for older windows, up to win2k3r2\n"
+      "\t\t\t       .xml: for newer windows", NULL, },
     { "config", 'c', 0, G_OPTION_ARG_CALLBACK,
       handle_config,
       "Set configuration parameter", "key=value" },
@@ -103,48 +113,45 @@ static OsinfoOs *find_os(OsinfoDb *db,
 static gboolean generate_script(OsinfoOs *os)
 {
     OsinfoInstallScriptList *scripts = osinfo_os_get_install_script_list(os);
-    OsinfoInstallScriptList *jeosScripts;
-    OsinfoFilter *filter;
+    OsinfoInstallScriptList *profileScripts;
     OsinfoInstallScript *script;
     gboolean ret = FALSE;
     GError *error = NULL;
-    gchar *data;
+    GFile *dir = g_file_new_for_commandline_arg(output_dir ? output_dir : ".");
 
-    filter = osinfo_filter_new();
-    osinfo_filter_add_constraint(filter,
-                                 OSINFO_INSTALL_SCRIPT_PROP_PROFILE,
-                                 profile ? profile :
-                                 OSINFO_INSTALL_SCRIPT_PROFILE_JEOS);
+    profileScripts =
+        osinfo_install_scriptlist_get_by_profile(scripts,
+                                                 profile ? profile:
+                                                 OSINFO_INSTALL_SCRIPT_PROFILE_JEOS);
 
-    jeosScripts = osinfo_install_scriptlist_new_filtered(scripts,
-                                                         filter);
-    if (osinfo_list_get_length(OSINFO_LIST(jeosScripts)) != 1) {
+    if (osinfo_list_get_length(OSINFO_LIST(profileScripts)) != 1) {
         g_printerr("Cannot find any install script for profile '%s'\n",
                    profile ? profile :
                    OSINFO_INSTALL_SCRIPT_PROFILE_JEOS);
         goto cleanup;
     }
 
-    script = OSINFO_INSTALL_SCRIPT(osinfo_list_get_nth(OSINFO_LIST(jeosScripts), 0));
-    if (!(data = osinfo_install_script_generate(script,
-                                                os,
-                                                config,
-                                                NULL,
-                                                &error))) {
+    script = OSINFO_INSTALL_SCRIPT(osinfo_list_get_nth(OSINFO_LIST(profileScripts), 0));
+
+    if (output_file)
+        osinfo_install_script_set_output_filename(script, output_file);
+
+    if (osinfo_install_script_generate_output(script,
+                                              os,
+                                              config,
+                                              dir,
+                                              NULL,
+                                              &error) == NULL) {
         g_printerr("Unable to generate install script: %s\n",
                    error ? error->message : "unknown");
         goto cleanup;
     }
 
-    g_print("%s\n", data);
-
     ret = TRUE;
 
  cleanup:
-    g_free(data);
     g_object_unref(scripts);
-    g_object_unref(jeosScripts);
-    g_object_unref(filter);
+    g_object_unref(profileScripts);
     return ret;
 }
 
@@ -197,7 +204,6 @@ gint main(gint argc, gchar **argv)
         goto EXIT;
     }
 
-
     if (!generate_script(os)) {
         ret = -5;
         goto EXIT;
-- 
1.7.10.4



From fabiano at fidencio.org  Tue Jul 24 20:05:16 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:16 +0200
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
	can/must be set
In-Reply-To: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1343160318-20796-5-git-send-email-fabiano@fidencio.org>

Now we need to set what are the configs that will be used in each
script. To set it, just add, in the .xml's script file:
<config>
  <param name="..." policy="mandatory"|"optional"/>
</config>

With these configs loaded on our db, applications that uses/will use
libosinfo can check if some config *can* be set using:
osinfo_install_script_is_config(OsinfoInstallScript *script,
                                const gchar* config);

And if some config *must* be set using:
osinfo_install_script_is_config_required(OsinfoInstallScript *script,
                                         const gchar* config);
---
 data/install-scripts/fedora.xml           |    6 ++++
 data/install-scripts/windows-sif.xml      |    5 +++
 data/install-scripts/windows-unattend.xml |    9 ++++++
 data/schemas/libosinfo.rng                |   12 ++++++++
 osinfo/libosinfo.syms                     |    2 ++
 osinfo/osinfo_install_script.c            |   47 +++++++++++++++++++++++++++++
 osinfo/osinfo_install_script.h            |    9 +++++-
 osinfo/osinfo_loader.c                    |   19 ++++++++++++
 8 files changed, 108 insertions(+), 1 deletion(-)

diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
index 338a570..a608cdb 100644
--- a/data/install-scripts/fedora.xml
+++ b/data/install-scripts/fedora.xml
@@ -1,6 +1,12 @@
 <libosinfo version="0.0.1">
   <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
     <profile>jeos</profile>
+    <config>
+      <param name="admin-password" policy="optional"/>
+      <param name="l10n-keyboard" policy="optional"/>
+      <param name="l10n-language" policy="optional"/>
+      <param name="l10n-timezone" policy="optional"/>
+    </config>
     <template filename="fedora.ks">
       <xsl:stylesheet
         xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
index 46b24ae..77fc2c5 100644
--- a/data/install-scripts/windows-sif.xml
+++ b/data/install-scripts/windows-sif.xml
@@ -1,6 +1,11 @@
 <libosinfo version="0.0.1">
   <install-script id='http://microsoft.com/windows/sif'>
     <profile>jeos</profile>
+    <config>
+      <param name="admin-password" policy="optional"/>
+      <param name="reg-product-key" policy="required"/>
+      <param name="user-realname" policy="required"/>
+    </config>
     <template filename="windows.sif">
       <xsl:stylesheet
         xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
index 20e3e23..5698bc8 100644
--- a/data/install-scripts/windows-unattend.xml
+++ b/data/install-scripts/windows-unattend.xml
@@ -1,6 +1,15 @@
 <libosinfo version="0.0.1">
   <install-script id='http://microsoft.com/windows/unattend'>
     <profile>jeos</profile>
+    <config>
+      <param name="admin-password" policy="optional"/>
+      <param name="hardware-arch" policy="optional"/>
+      <param name="l10n-language" policy="optional"/>
+      <param name="user-login" policy="optional"/>
+      <param name="user-password" policy="optional"/>
+      <param name="user-realname" policy="optional"/>
+      <param name="reg-product-key" policy="required"/>
+    </config>
     <template filename="windows.xml">
       <xsl:stylesheet
         xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
index 7c8d7f7..1392f74 100644
--- a/data/schemas/libosinfo.rng
+++ b/data/schemas/libosinfo.rng
@@ -410,6 +410,12 @@
       <element name='profile'>
         <text/>
       </element>
+      <element name='config'>
+        <attribute name="name"/>
+        <attribute name="policy">
+          <ref name='policies'/>
+        </attribute>
+      </element>
       <element name='product-key-format'>
         <text/>
       </element>
@@ -479,4 +485,10 @@
       <param name="pattern">\w+://.*</param>
     </data>
   </define>
+
+  <define name='policies'>
+    <data type="string">
+      <param name="pattern">required|optional</param>
+    </data>
+  </define>
 </grammar>
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 5a7de7b..2d64ebb 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -276,6 +276,8 @@ LIBOSINFO_0.2.0 {
 	osinfo_install_script_generate_output_async;
 	osinfo_install_script_get_profile;
 	osinfo_install_script_get_uri;
+	osinfo_install_script_is_config;
+	osinfo_install_script_is_config_required;
 	osinfo_install_scriptlist_new;
 	osinfo_install_scriptlist_new_filtered;
 	osinfo_install_scriptlist_new_union;
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index b2d7351..6ce1c55 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -240,6 +240,26 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
     g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
 }
 
+static GList *osinfo_install_script_get_config_required(OsinfoInstallScript *script)
+{
+    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
+                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED);
+}
+
+static GList *osinfo_install_script_get_config_optional(OsinfoInstallScript *script)
+{
+    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
+                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL);
+}
+
+static GList *osinfo_install_script_get_config(OsinfoInstallScript *script)
+{
+    GList *required = osinfo_install_script_get_config_required(script);
+    GList *optional = osinfo_install_script_get_config_optional(script);
+
+    return g_list_concat(required, optional);
+}
+
 static void
 osinfo_install_script_init (OsinfoInstallScript *list)
 {
@@ -336,6 +356,33 @@ const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *scri
                                          OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
 }
 
+static gint osinfo_install_script_compare_configs(gconstpointer a, gconstpointer b)
+{
+    return (g_strcmp0((const gchar *)a, (const gchar *)b));
+}
+
+gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
+                                         const gchar *config)
+{
+    if (g_list_find_custom(osinfo_install_script_get_config(script),
+                           config,
+                           osinfo_install_script_compare_configs) == NULL)
+        return FALSE;
+
+    return TRUE;
+}
+
+gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
+                                                  const gchar *config)
+{
+    if (g_list_find_custom(osinfo_install_script_get_config_required(script),
+                           config,
+                           osinfo_install_script_compare_configs) == NULL)
+        return FALSE;
+
+    return TRUE;
+}
+
 struct OsinfoInstallScriptGenerate {
     GSimpleAsyncResult *res;
     OsinfoOs *os;
diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
index d75ec93..cca257a 100644
--- a/osinfo/osinfo_install_script.h
+++ b/osinfo/osinfo_install_script.h
@@ -50,7 +50,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
 #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
 #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
 #define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
-
+#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED    "required"
+#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL    "optional"
 
 /* object */
 struct _OsinfoInstallScript
@@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
                                                    GCancellable *cancellable,
                                                    GError **error);
 
+gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
+                                         const gchar *config);
+
+gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
+                                                  const gchar *config);
+
 #endif /* __OSINFO_INSTALL_SCRIPT_H__ */
 /*
  * Local variables:
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index 31e442e..2e53d14 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -546,6 +546,11 @@ static void osinfo_loader_install_script(OsinfoLoader *loader,
                                          xmlNodePtr root,
                                          GError **err)
 {
+
+    xmlNodePtr *nodes = NULL;
+    gint nnodes;
+    guint i;
+
     gchar *id = (gchar *)xmlGetProp(root, BAD_CAST "id");
     const gchar *const keys[] = {
         OSINFO_INSTALL_SCRIPT_PROP_PROFILE,
@@ -594,6 +599,20 @@ static void osinfo_loader_install_script(OsinfoLoader *loader,
                                 value);
     g_free(value);
 
+    nnodes = osinfo_loader_nodeset("./config/*", ctxt, &nodes, err);
+    if (error_is_set(err))
+        return;
+
+    for (i = 0; i < nnodes; i++) {
+        gchar *policy = (gchar *)xmlGetProp(nodes[i], BAD_CAST "policy");
+        gchar *name = (gchar *)xmlGetProp(nodes[i], BAD_CAST "name");
+
+        osinfo_entity_add_param(OSINFO_ENTITY(installScript), policy, name);
+
+        g_free(policy);
+        g_free(name);
+    }
+
     osinfo_db_add_install_script(loader->priv->db, installScript);
 
     return;
-- 
1.7.10.4



From fabiano at fidencio.org  Tue Jul 24 20:05:17 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:17 +0200
Subject: [virt-tools-list] [PATCH 5/6] Add function to fetch console
	keyboard layout
In-Reply-To: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1343160318-20796-6-git-send-email-fabiano@fidencio.org>

Similar as found on Boxes, this function can be useful for all linuxes,
so, is better put it into our API
---
 osinfo/libosinfo.syms          |    1 +
 osinfo/osinfo_install_script.c |  121 ++++++++++++++++++++++++++++++++++++++++
 osinfo/osinfo_install_script.h |    3 +
 3 files changed, 125 insertions(+)

diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 2d64ebb..ba8f045 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -278,6 +278,7 @@ LIBOSINFO_0.2.0 {
 	osinfo_install_script_get_uri;
 	osinfo_install_script_is_config;
 	osinfo_install_script_is_config_required;
+	osinfo_install_script_get_console_keyboard_layout;
 	osinfo_install_scriptlist_new;
 	osinfo_install_scriptlist_new_filtered;
 	osinfo_install_scriptlist_new_union;
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 6ce1c55..37467aa 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -383,6 +383,127 @@ gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
     return TRUE;
 }
 
+struct KbdLayout {
+    const gchar *xkb_layout;
+    const gchar *xkb_variant;
+    const gchar *console_layout;
+};
+
+struct KbdLayout kbd_layouts[] = {
+    { "ara", NULL, "ar-azerty" },
+    { "ara", "azerty", "ar-azerty" },
+    { "ara", "azerty_digits", "ar-azerty-digits" },
+    { "ara", "digits", "ar-digits" },
+    { "ara", "qwerty", "ar-qwerty" },
+    { "ara", "qwerty_digits", "ar-qwerty-digits" },
+    { "be", NULL, "be-latin1" },
+    { "bg", NULL, "bg_bds-utf8" },
+    { "bg", "phonetic", "bg_pho-utf8" },
+    { "bg", "bas_phonetic", "bg_pho-utf8" },
+    { "br", NULL, "br-abnt2" },
+    { "ca(fr)", NULL, "cf" },
+    { "hr", NULL, "croat" },
+    { "cz", NULL, "cz-us-qwertz" },
+    { "cz", "qwerty", "cz-lat2" },
+    { "cz", "", "cz-us-qwertz" },
+    { "de", NULL, "de" },
+    { "de", "nodeadkeys", "de-latin1-nodeadkeys" },
+    { "dev", NULL, "dev" },
+    { "dk", NULL, "dk" },
+    { "dk", "dvorak", "dk-dvorak" },
+    { "es", NULL, "es" },
+    { "ee", NULL, "et" },
+    { "fi", NULL, "fi" },
+    { "fr", NULL, "fr" },
+    { "fr", "latin9", "fr-latin9" },
+    { "gr", NULL, "gr" },
+    { "gur", NULL, "gur" },
+    { "hu", NULL, "hu" },
+    { "hu", "qwerty", "hu101" },
+    { "ie", NULL, "ie" },
+    { "in", NULL, "us" },
+    { "in", "ben", "ben" },
+    { "in", "ben-probhat", "ben_probhat" },
+    { "in", "guj", "guj" },
+    { "in", "tam", "tml-inscript" },
+    { "in", "tam_TAB", "tml-uni" },
+    { "is", NULL, "is-latin1" },
+    { "it", NULL, "it" },
+    { "jp", NULL, "jp106" },
+    { "kr", NULL, "ko" },
+    { "latam", NULL, "la-latin1" },
+    { "mkd", NULL, "mk-utf" },
+    { "nl", NULL, "nl" },
+    { "no", NULL, "no" },
+    { "pl", NULL, "pl2" },
+    { "pt", NULL, "pt-latin1" },
+    { "ro", NULL, "ro" },
+    { "ro", "std", "ro-std" },
+    { "ro", "cedilla", "ro-cedilla" },
+    { "ro", "std_cedilla", "ro-std-cedilla" },
+    { "ru", NULL, "ru" },
+    { "rs", NULL, "sr-cy" },
+    { "rs", "latin", "sr-latin"},
+    { "se", NULL, "sv-latin1" },
+    { "ch", "de_nodeadkeys", "sg" },
+    { "ch", "fr", "fr_CH" },
+    { "sk", NULL, "sk-qwerty" },
+    { "si", NULL, "slovene" },
+    { "tj", NULL, "tj" },
+    { "tr", NULL, "trq" },
+    { "gb", NULL, "uk" },
+    { "ua", NULL, "ua-utf" },
+    { "us", NULL, "us" },
+    { "us", "intl", "us-acentos" }
+};
+
+const gchar *osinfo_install_script_get_console_keyboard_layout (OsinfoInstallScript *script)
+{
+    GSettings *settings = g_settings_new("org.gnome.libgnomekbd.keyboard");
+    gchar **tokens, **layouts;
+    gchar *layout_str, *xkb_layout, *xkb_variant;
+    const gchar *console_layout = NULL;
+    gint i;
+
+    layouts = g_settings_get_strv(settings, "layouts");
+    layout_str = layouts[0];
+
+    if (layout_str == NULL || g_strcmp0(layout_str, "") == 0) {
+        g_printerr ("Failed to fetch prefered keyboard layout from user settings, falling back to 'us'..");
+
+        console_layout = "us";
+        goto finish;
+    }
+
+    tokens = g_strsplit(layout_str, "\t", 0);
+    xkb_layout = tokens[0];
+    xkb_variant = tokens[1];
+
+    for (i = 0; i < G_N_ELEMENTS(kbd_layouts); i++) {
+        if (xkb_layout == kbd_layouts[i].xkb_layout) {
+            if (xkb_variant == kbd_layouts[i].xkb_variant) {
+                console_layout = kbd_layouts[i].console_layout;
+                break;
+            } else if (kbd_layouts[i].xkb_variant == NULL)
+                console_layout = kbd_layouts[i].console_layout;
+        }
+    }
+
+    if (console_layout == NULL)
+        console_layout = "us";
+
+    for (i = 0; i < G_N_ELEMENTS(tokens); i++)
+        g_free(tokens[i]);
+    g_free(tokens);
+
+finish:
+    for (i = 0; i < G_N_ELEMENTS(layouts); i++)
+        g_free(layouts[i]);
+    g_free(layouts);
+
+    return console_layout;
+}
+
 struct OsinfoInstallScriptGenerate {
     GSimpleAsyncResult *res;
     OsinfoOs *os;
diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
index cca257a..d141e06 100644
--- a/osinfo/osinfo_install_script.h
+++ b/osinfo/osinfo_install_script.h
@@ -135,6 +135,9 @@ gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
 gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
                                                   const gchar *config);
 
+const gchar *osinfo_install_script_get_console_keyboard_layout (OsinfoInstallScript *script);
+
+
 #endif /* __OSINFO_INSTALL_SCRIPT_H__ */
 /*
  * Local variables:
-- 
1.7.10.4



From fabiano at fidencio.org  Tue Jul 24 20:05:18 2012
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 24 Jul 2012 22:05:18 +0200
Subject: [virt-tools-list] [PATCH 6/6] Add desktop profile for Fedora 15 and
	newer
In-Reply-To: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1343160318-20796-7-git-send-email-fabiano@fidencio.org>

As talked on IRC, was decided to not install QXL drivers for older
Fedora (16 or lesser). Why? We are keeping some hacks only to avoid
broken QXL driver and, at least for now, we are removing these.
---
 data/install-scripts/fedora.xml |  157 +++++++++++++++++++++++++++++++++++++++
 data/oses/fedora.xml            |    4 +
 2 files changed, 161 insertions(+)

diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
index a608cdb..04e970d 100644
--- a/data/install-scripts/fedora.xml
+++ b/data/install-scripts/fedora.xml
@@ -1,4 +1,5 @@
 <libosinfo version="0.0.1">
+  <!-- JEOS PROFILE -->
   <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
     <profile>jeos</profile>
     <config>
@@ -100,4 +101,160 @@ reboot
     </template>
   </install-script>
 
+  <!-- DESKTOP PROFILE -->
+  <install-script id='http://fedoraproject.org/scripts/fedora/desktop'>
+    <profile>desktop</profile>
+    <config>
+        <param name="l10n-keyboard" policy="optional"/>
+        <param name="l10n-language" policy="optional"/>
+        <param name="l10n-timezone" policy="optional"/>
+        <param name="hostname" policy="optional"/>
+        <param name="user-login" policy="required"/>
+        <param name="user-password" policy="required"/>
+        <param name="admin-password" policy="required"/>
+    </config>
+    <template filename="fedora.ks">
+      <xsl:stylesheet
+        xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+        version="1.0">
+
+        <xsl:output method="text"/>
+
+	<xsl:template name="disk">
+	  <xsl:choose>
+	    <xsl:when test="os/version &gt; 9">
+	      <!-- virtio -->
+	      <xsl:text>vda</xsl:text>
+	    </xsl:when>
+	    <xsl:when test="os/version &gt; 6">
+	      <!-- libata IDE -->
+	      <xsl:text>sda</xsl:text>
+	    </xsl:when>
+	    <xsl:otherwise>
+	      <!-- IDE -->
+	      <xsl:text>hda</xsl:text>
+	    </xsl:otherwise>
+	  </xsl:choose>
+	</xsl:template>
+
+	<xsl:template name="avatar-disk">
+	  <xsl:choose>
+	    <xsl:when test="os/version &gt; 9">
+	      <!-- virtio -->
+	      <xsl:text>sda</xsl:text>
+	    </xsl:when>
+	    <xsl:when test="os/version &gt; 6">
+	      <!-- libata IDE -->
+	      <xsl:text>hda</xsl:text>
+	    </xsl:when>
+	    <xsl:otherwise>
+	      <!-- IDE -->
+	      <xsl:text>hdb</xsl:text>
+	    </xsl:otherwise>
+	  </xsl:choose>
+	</xsl:template>
+
+	<xsl:template name="rootfs">
+	  <xsl:choose>
+	    <xsl:when test="os/version &gt; 10">
+	      <xsl:text>ext4</xsl:text>
+	    </xsl:when>
+	    <xsl:otherwise>
+	      <xsl:text>ext3</xsl:text>
+	    </xsl:otherwise>
+	  </xsl:choose>
+	</xsl:template>
+
+	<xsl:template name="bootfs">
+	  <xsl:choose>
+	    <xsl:when test="os/version &gt; 11">
+	      <xsl:text>ext4</xsl:text>
+	    </xsl:when>
+	    <xsl:otherwise>
+	      <xsl:text>ext3</xsl:text>
+	    </xsl:otherwise>
+	  </xsl:choose>
+	</xsl:template>
+
+        <xsl:template match="/install-script-config">
+# Install script for <xsl:value-of select="os/short-id"/> profile <xsl:value-of select="script/profile"/>
+install
+keyboard <xsl:value-of select="config/l10n-keyboard"/>
+lang <xsl:value-of select="config/l10n-language"/>
+network --onboot yes --device eth0 --bootproto dhcp --noipv6 --hostname=<xsl:value-of select="config/hostname"/> --activate
+rootpw <xsl:value-of select="config/admin-password"/>
+firewall --disabled
+authconfig --enableshadow --enablemd5
+timezone --utc <xsl:value-of select="config/l10n-timezone"/>
+bootloader --location=mbr
+zerombr
+
+clearpart --all --drives=<xsl:call-template name="disk"/>
+
+firstboot --disable
+
+part biosboot --fstype=biosboot --size=1
+part /boot --fstype <xsl:call-template name="bootfs"/> --recommended --ondisk=<xsl:call-template name="disk"/>
+part pv.2 --size=1 --grow --ondisk=<xsl:call-template name="disk"/>
+volgroup VolGroup00 --pesize=32768 pv.2
+logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=768 --grow --maxsize=1536
+logvol / --fstype <xsl:call-template name="rootfs"/> --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow
+reboot
+
+user --name=<xsl:value-of select="config/user-login"/> --password=<xsl:value-of select="config/user-password"/>
+
+%packages
+ at base
+ at core
+ at hardware-support
+ at base-x
+ at gnome-desktop
+ at graphical-internet
+ at sound-and-video
+
+# QXL driver and SPICE vdagent
+xorg-x11-drv-qxl
+spice-vdagent
+
+%end
+
+%post --erroronfail
+
+# Add user to admin group
+usermod -a -G wheel <xsl:value-of select="config/user-login"/>
+
+# Set user avatar
+mkdir /mnt/unattended-media
+mount /dev/<xsl:call-template name='avatar-disk'/> /mnt/unattended-media
+cp /mnt/unattended-media/<xsl:value-of select="config/user-login"/> /var/lib/AccountsService/icons/
+umount /mnt/unattended-media
+echo "
+[User]
+Language=
+XSession=
+Icon=/var/lib/AccountsService/icons/<xsl:value-of select="config/user-login"/>
+" >> /var/lib/AccountsService/users/<xsl:value-of select="config/user-login"/>
+
+# Enable autologin
+echo "[daemon]
+AutomaticLoginEnable=true
+AutomaticLogin=<xsl:value-of select="config/user-login"/>
+
+[security]
+
+[xdmcp]
+
+[greeter]
+
+[chooser]
+
+[debug]
+" > /etc/gdm/custom.conf
+
+%end
+	</xsl:template>
+      </xsl:stylesheet>
+    </template>
+  </install-script>
+
 </libosinfo>
diff --git a/data/oses/fedora.xml b/data/oses/fedora.xml
index 662d8af..c62a25e 100644
--- a/data/oses/fedora.xml
+++ b/data/oses/fedora.xml
@@ -1024,6 +1024,7 @@
 
     <installer>
       <script id='http://fedoraproject.org/scripts/fedora/jeos'/>
+      <script id='http://fedoraproject.org/scripts/fedora/desktop'/>
     </installer>
   </os>
 
@@ -1111,6 +1112,7 @@
 
     <installer>
       <script id='http://fedoraproject.org/scripts/fedora/jeos'/>
+      <script id='http://fedoraproject.org/scripts/fedora/desktop'/>
     </installer>
   </os>
 
@@ -1198,6 +1200,7 @@
 
     <installer>
       <script id='http://fedoraproject.org/scripts/fedora/jeos'/>
+      <script id='http://fedoraproject.org/scripts/fedora/desktop'/>
     </installer>
   </os>
 
@@ -1283,6 +1286,7 @@
 
     <installer>
       <script id='http://fedoraproject.org/scripts/fedora/jeos'/>
+      <script id='http://fedoraproject.org/scripts/fedora/desktop'/>
     </installer>
   </os>
 
-- 
1.7.10.4



From list-virt at legroom.net  Wed Jul 25 01:52:15 2012
From: list-virt at legroom.net (Jared)
Date: Tue, 24 Jul 2012 20:52:15 -0500
Subject: [virt-tools-list] virt-install cpuset=auto error
In-Reply-To: <500F47FB.8000407@legroom.net>
References: <500A6AFB.8030500@legroom.net> <500D0180.4040508@redhat.com>
	<500D0426.7090509@legroom.net> <500D4EBC.8000902@redhat.com>
	<500D5AF2.8060504@legroom.net> <500ED7B9.9050509@redhat.com>
	<500F47FB.8000407@legroom.net>
Message-ID: <500F514F.4080001@legroom.net>

On 07/24/2012 08:12 PM, Jared wrote:
> On 07/24/2012 12:13 PM, Cole Robinson wrote:
>> Hmm. Just to make sure it isn't an install issue:
>> 
>> git clone git://git.fedorahosted.org/git/python-virtinst.git
>> cd python-virtinst
>> python setup.py build
>> ./virt-install --debug <the rest of your command>
>> 
>> and reply with the output attached.
>
> looks like doing it that way actually worked:
> 
> [Tue, 24 Jul 2012 20:09:48 virt-install 11669] DEBUG (cli:838) Auto cpuset
> is: 0,2,4,6,8,10,12,14,16,18,20,22


I did a diff of the git version and my installed version and found what you
see below.  Looks like this setting was shuffled around a bit so it could be
properly initialized.  You sure this wasn't added after the 0.600.2 release?

Either way, I'm good now.  I can patch my system version easily enough, and
as this is already fixed in the code I won't need to worry about it for the
next version.

Thanks a bunch for helping me track this down.

# diff -u /usr/bin/virt-install-2.7 ~/python-virtinst/virt-install
--- /usr/bin/virt-install-2.7   2012-07-21 20:30:53.000000000 -0500
+++ ~/python-virtinst/virt-install 2012-07-24 19:58:41.293789541 -0500
@@ -1,4 +1,4 @@
-#!/usr/bin/python2.7 -tt
+#!/usr/bin/python -tt
 #
 # Script to set up a Xen guest and kick off an install
 #
@@ -478,7 +478,6 @@
     # Guest configuration
     cli.get_uuid(options.uuid, guest)
     cli.get_vcpus(guest, options.vcpus, options.check_cpu)
-    cli.get_cpuset(guest, options.cpuset)
     cli.parse_numatune(guest, options.numatune)
     cli.parse_cpu(guest, options.cpu)
     cli.parse_security(guest, options.security)
@@ -534,6 +533,9 @@
     # this after setting guest.installer at least
     check_option_collisions(options, guest)

+    # Needs to come after setting memory
+    cli.get_cpuset(guest, options.cpuset, guest.memory)
+
     # Warnings
     if options.pxe and not supports_pxe(guest):
         logging.warn(_("The guest's network configuration does not support "

-- 
Jared



From berrange at redhat.com  Wed Jul 25 13:04:02 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 14:04:02 +0100
Subject: [virt-tools-list] [PATCH 1/6] Add "hostname" prop to
 install-script-config
In-Reply-To: <1343160318-20796-2-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-2-git-send-email-fabiano@fidencio.org>
Message-ID: <20120725130401.GE6596@redhat.com>

On Tue, Jul 24, 2012 at 10:05:13PM +0200, Fabiano Fid?ncio wrote:
> It is desirable to configure the hostname of the new virtual
> machine during initial automated provisioning
> ---
>  osinfo/libosinfo.syms          |    1 +
>  osinfo/osinfo_install_config.c |   13 +++++++++++++
>  osinfo/osinfo_install_config.h |    6 ++++++
>  3 files changed, 20 insertions(+)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Wed Jul 25 13:04:24 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 14:04:24 +0100
Subject: [virt-tools-list] [PATCH 2/6] Don't return a gboolean when a
 gchar * is expected
In-Reply-To: <1343160318-20796-3-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-3-git-send-email-fabiano@fidencio.org>
Message-ID: <20120725130424.GF6596@redhat.com>

On Tue, Jul 24, 2012 at 10:05:14PM +0200, Fabiano Fid?ncio wrote:
> ---
>  osinfo/osinfo_install_script.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
> index 7bf1e2e..1f437a1 100644
> --- a/osinfo/osinfo_install_script.c
> +++ b/osinfo/osinfo_install_script.c
> @@ -627,10 +627,10 @@ gchar *osinfo_install_script_generate_finish(OsinfoInstallScript *script,
>  {
>      GSimpleAsyncResult *simple = G_SIMPLE_ASYNC_RESULT(res);
>  
> -    g_return_val_if_fail(error == NULL || *error == NULL, FALSE);
> +    g_return_val_if_fail(error == NULL || *error == NULL, NULL);
>  
>      if (g_simple_async_result_propagate_error(simple, error))
> -        return FALSE;
> +        return NULL;
>  
>      return g_simple_async_result_get_op_res_gpointer(simple);
>  }

ACK


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Wed Jul 25 13:14:52 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 14:14:52 +0100
Subject: [virt-tools-list] [PATCH 3/6] Change install script output
	(tool)
In-Reply-To: <1343160318-20796-4-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-4-git-send-email-fabiano@fidencio.org>
Message-ID: <20120725131452.GG6596@redhat.com>

On Tue, Jul 24, 2012 at 10:05:15PM +0200, Fabiano Fid?ncio wrote:
> Now, instead of return a string, the osinfo-install-script tool writes
> that string in a file, passed as argument, that will be into a dir,
> also passed as argument.
> 
> If the filename argument is NULL, the output files will be written as
> set in data/install-scripts/*.xml (in filename attribute from template
> element):
>     - Linuxes: fedora.ks
>     - Windows 2k3r2 and older: windows.sif
>     - Windows 2k8 and newer: windows.xml

I think we should omit the filename argument entirely on the
command line, because in the Windows case we're going to need
to support generation of multiple output files. In such a case
passing a filename on the ARGV doesn't make sense.

I could see value in having an optional arg to specify a
filename prefix. eg

   # osinfo-install-script --prefix foo   win2k3r2

would result in a file  'foo-windows.sif' being created.

> If the dirname argument is NULL, the output files will be written in the
> current directory.

I think we could probably just make that an parameter flag
instead eg.

   # osinfo-install-script --output-dir /some/dir  win2k3r2


> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
> index 1f437a1..b2d7351 100644
> --- a/osinfo/osinfo_install_script.c
> +++ b/osinfo/osinfo_install_script.c
> @@ -56,6 +56,7 @@ enum {
>      PROP_TEMPLATE_DATA,
>      PROP_PROFILE,
>      PROP_PRODUCT_KEY_FORMAT,
> +    PROP_OUTPUT_FILENAME,
>  };

Given an an <install-script> will allow multiple <template> args
each with their own filename, I don't think it really makes sense
to have an 'output-filename' property.  I'd support having an
"output-prefix" property to support the usage I desribed earlier

> @@ -686,6 +737,136 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
>      return data.output;
>  }
>  
> +void osinfo_install_script_generate_output_async(OsinfoInstallScript *script,
> +                                                 OsinfoOs *os,
> +                                                 OsinfoInstallConfig *config,
> +                                                 GFile *output_dir,

This parameter appears to be ignored in this method which is strange ?

> +                                                 GCancellable *cancellable,
> +                                                 GAsyncReadyCallback callback,
> +                                                 gpointer user_data)
> +{
> +    osinfo_install_script_generate_async(script,
> +                                         os,
> +                                         config,
> +                                         cancellable,
> +                                         callback,
> +                                         user_data);
> +}

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Wed Jul 25 13:18:57 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 14:18:57 +0100
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <1343160318-20796-5-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
Message-ID: <20120725131857.GH6596@redhat.com>

On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
> Now we need to set what are the configs that will be used in each
> script. To set it, just add, in the .xml's script file:
> <config>
>   <param name="..." policy="mandatory"|"optional"/>
> </config>
> 
> With these configs loaded on our db, applications that uses/will use
> libosinfo can check if some config *can* be set using:
> osinfo_install_script_is_config(OsinfoInstallScript *script,
>                                 const gchar* config);
> 
> And if some config *must* be set using:
> osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>                                          const gchar* config);
> ---
>  data/install-scripts/fedora.xml           |    6 ++++
>  data/install-scripts/windows-sif.xml      |    5 +++
>  data/install-scripts/windows-unattend.xml |    9 ++++++
>  data/schemas/libosinfo.rng                |   12 ++++++++
>  osinfo/libosinfo.syms                     |    2 ++
>  osinfo/osinfo_install_script.c            |   47 +++++++++++++++++++++++++++++
>  osinfo/osinfo_install_script.h            |    9 +++++-
>  osinfo/osinfo_loader.c                    |   19 ++++++++++++
>  8 files changed, 108 insertions(+), 1 deletion(-)
> 
> diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
> index 338a570..a608cdb 100644
> --- a/data/install-scripts/fedora.xml
> +++ b/data/install-scripts/fedora.xml
> @@ -1,6 +1,12 @@
>  <libosinfo version="0.0.1">
>    <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
>      <profile>jeos</profile>
> +    <config>
> +      <param name="admin-password" policy="optional"/>
> +      <param name="l10n-keyboard" policy="optional"/>
> +      <param name="l10n-language" policy="optional"/>
> +      <param name="l10n-timezone" policy="optional"/>
> +    </config>
>      <template filename="fedora.ks">
>        <xsl:stylesheet
>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
> index 46b24ae..77fc2c5 100644
> --- a/data/install-scripts/windows-sif.xml
> +++ b/data/install-scripts/windows-sif.xml
> @@ -1,6 +1,11 @@
>  <libosinfo version="0.0.1">
>    <install-script id='http://microsoft.com/windows/sif'>
>      <profile>jeos</profile>
> +    <config>
> +      <param name="admin-password" policy="optional"/>
> +      <param name="reg-product-key" policy="required"/>
> +      <param name="user-realname" policy="required"/>
> +    </config>
>      <template filename="windows.sif">
>        <xsl:stylesheet
>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
> index 20e3e23..5698bc8 100644
> --- a/data/install-scripts/windows-unattend.xml
> +++ b/data/install-scripts/windows-unattend.xml
> @@ -1,6 +1,15 @@
>  <libosinfo version="0.0.1">
>    <install-script id='http://microsoft.com/windows/unattend'>
>      <profile>jeos</profile>
> +    <config>
> +      <param name="admin-password" policy="optional"/>
> +      <param name="hardware-arch" policy="optional"/>
> +      <param name="l10n-language" policy="optional"/>
> +      <param name="user-login" policy="optional"/>
> +      <param name="user-password" policy="optional"/>
> +      <param name="user-realname" policy="optional"/>
> +      <param name="reg-product-key" policy="required"/>
> +    </config>
>      <template filename="windows.xml">
>        <xsl:stylesheet
>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
> index 7c8d7f7..1392f74 100644
> --- a/data/schemas/libosinfo.rng
> +++ b/data/schemas/libosinfo.rng
> @@ -410,6 +410,12 @@
>        <element name='profile'>
>          <text/>
>        </element>
> +      <element name='config'>
> +        <attribute name="name"/>
> +        <attribute name="policy">
> +          <ref name='policies'/>
> +        </attribute>
> +      </element>
>        <element name='product-key-format'>
>          <text/>
>        </element>
> @@ -479,4 +485,10 @@
>        <param name="pattern">\w+://.*</param>
>      </data>
>    </define>
> +
> +  <define name='policies'>
> +    <data type="string">
> +      <param name="pattern">required|optional</param>
> +    </data>
> +  </define>
>  </grammar>
> diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
> index 5a7de7b..2d64ebb 100644
> --- a/osinfo/libosinfo.syms
> +++ b/osinfo/libosinfo.syms
> @@ -276,6 +276,8 @@ LIBOSINFO_0.2.0 {
>  	osinfo_install_script_generate_output_async;
>  	osinfo_install_script_get_profile;
>  	osinfo_install_script_get_uri;
> +	osinfo_install_script_is_config;
> +	osinfo_install_script_is_config_required;
>  	osinfo_install_scriptlist_new;
>  	osinfo_install_scriptlist_new_filtered;
>  	osinfo_install_scriptlist_new_union;
> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
> index b2d7351..6ce1c55 100644
> --- a/osinfo/osinfo_install_script.c
> +++ b/osinfo/osinfo_install_script.c
> @@ -240,6 +240,26 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
>      g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
>  }
>  
> +static GList *osinfo_install_script_get_config_required(OsinfoInstallScript *script)
> +{
> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED);
> +}
> +
> +static GList *osinfo_install_script_get_config_optional(OsinfoInstallScript *script)
> +{
> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL);
> +}
> +
> +static GList *osinfo_install_script_get_config(OsinfoInstallScript *script)
> +{
> +    GList *required = osinfo_install_script_get_config_required(script);
> +    GList *optional = osinfo_install_script_get_config_optional(script);
> +
> +    return g_list_concat(required, optional);
> +}

This leaks memory associated with 'optional'

> +
>  static void
>  osinfo_install_script_init (OsinfoInstallScript *list)
>  {
> @@ -336,6 +356,33 @@ const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *scri
>                                           OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
>  }
>  
> +static gint osinfo_install_script_compare_configs(gconstpointer a, gconstpointer b)
> +{
> +    return (g_strcmp0((const gchar *)a, (const gchar *)b));
> +}
> +
> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
> +                                         const gchar *config)
> +{
> +    if (g_list_find_custom(osinfo_install_script_get_config(script),

This leaks the GList you're using

> +                           config,
> +                           osinfo_install_script_compare_configs) == NULL)
> +        return FALSE;
> +
> +    return TRUE;
> +}
> +
> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
> +                                                  const gchar *config)
> +{
> +    if (g_list_find_custom(osinfo_install_script_get_config_required(script),

This leaks the GList you're using

> +                           config,
> +                           osinfo_install_script_compare_configs) == NULL)
> +        return FALSE;
> +
> +    return TRUE;
> +}
> +
>  struct OsinfoInstallScriptGenerate {
>      GSimpleAsyncResult *res;
>      OsinfoOs *os;
> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
> index d75ec93..cca257a 100644
> --- a/osinfo/osinfo_install_script.h
> +++ b/osinfo/osinfo_install_script.h
> @@ -50,7 +50,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
>  #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
>  #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
>  #define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
> -
> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED    "required"
> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL    "optional"
>  
>  /* object */
>  struct _OsinfoInstallScript
> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
>                                                     GCancellable *cancellable,
>                                                     GError **error);
>  
> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
> +                                         const gchar *config);
> +
> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
> +                                                  const gchar *config);

How about having an enum

   typedef enum {
     OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
     OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
     OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
   } OsinfoInstallScriptConfigPolicy;


  OsinfoInstallScriptConfigPolicy;
  osinfo_install_script_get_config_parameter(OsinfoInstallScript *script,
                                            const gchar *config);

Also I'd probably add an API to list all parameters

  GList *
  osinfo_install_script_get_config_parameters(OsinfoInstallScript *script);



Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Wed Jul 25 13:20:50 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 14:20:50 +0100
Subject: [virt-tools-list] [PATCH 5/6] Add function to fetch console
 keyboard layout
In-Reply-To: <1343160318-20796-6-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-6-git-send-email-fabiano@fidencio.org>
Message-ID: <20120725132050.GI6596@redhat.com>

On Tue, Jul 24, 2012 at 10:05:17PM +0200, Fabiano Fid?ncio wrote:
> Similar as found on Boxes, this function can be useful for all linuxes,
> so, is better put it into our API

While I agree its a useful function, I don't think it should go in
the libosinfo library. I'd like libosinfo to remain as a metadata
driven database.

This is keymap lookup & translation is something that I think
would be useful in a "libvirt installer" library, which is
something we're about to start working on real soon. So I'll
keep this code in mind for that.


Regards,
Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From berrange at redhat.com  Wed Jul 25 13:21:13 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 14:21:13 +0100
Subject: [virt-tools-list] [PATCH 6/6] Add desktop profile for Fedora 15
 and newer
In-Reply-To: <1343160318-20796-7-git-send-email-fabiano@fidencio.org>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-7-git-send-email-fabiano@fidencio.org>
Message-ID: <20120725132113.GJ6596@redhat.com>

On Tue, Jul 24, 2012 at 10:05:18PM +0200, Fabiano Fid?ncio wrote:
> As talked on IRC, was decided to not install QXL drivers for older
> Fedora (16 or lesser). Why? We are keeping some hacks only to avoid
> broken QXL driver and, at least for now, we are removing these.
> ---
>  data/install-scripts/fedora.xml |  157 +++++++++++++++++++++++++++++++++++++++
>  data/oses/fedora.xml            |    4 +
>  2 files changed, 161 insertions(+)

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From crobinso at redhat.com  Wed Jul 25 14:00:47 2012
From: crobinso at redhat.com (Cole Robinson)
Date: Wed, 25 Jul 2012 10:00:47 -0400
Subject: [virt-tools-list] virt-install cpuset=auto error
In-Reply-To: <500F514F.4080001@legroom.net>
References: <500A6AFB.8030500@legroom.net> <500D0180.4040508@redhat.com>
	<500D0426.7090509@legroom.net> <500D4EBC.8000902@redhat.com>
	<500D5AF2.8060504@legroom.net> <500ED7B9.9050509@redhat.com>
	<500F47FB.8000407@legroom.net> <500F514F.4080001@legroom.net>
Message-ID: <500FFC0F.2080809@redhat.com>

On 07/24/2012 09:52 PM, Jared wrote:
> On 07/24/2012 08:12 PM, Jared wrote:
>> On 07/24/2012 12:13 PM, Cole Robinson wrote:
>>> Hmm. Just to make sure it isn't an install issue:
>>>
>>> git clone git://git.fedorahosted.org/git/python-virtinst.git
>>> cd python-virtinst
>>> python setup.py build
>>> ./virt-install --debug <the rest of your command>
>>>
>>> and reply with the output attached.
>>
>> looks like doing it that way actually worked:
>>
>> [Tue, 24 Jul 2012 20:09:48 virt-install 11669] DEBUG (cli:838) Auto cpuset
>> is: 0,2,4,6,8,10,12,14,16,18,20,22
> 
> 
> I did a diff of the git version and my installed version and found what you
> see below.  Looks like this setting was shuffled around a bit so it could be
> properly initialized.  You sure this wasn't added after the 0.600.2 release?
> 
> Either way, I'm good now.  I can patch my system version easily enough, and
> as this is already fixed in the code I won't need to worry about it for the
> next version.
> 
> Thanks a bunch for helping me track this down.
> 

Argh, my bad, it was indeed added after the release :( I though I hadn't made
any changes after the release but I am forgetful. Sorry for the run around,
but glad it's fixed for you in some capacity.

Thanks,
Cole




From fabiano at fidencio.org  Wed Jul 25 14:27:01 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Wed, 25 Jul 2012 16:27:01 +0200
Subject: [virt-tools-list] [PATCH 3/6] Change install script output
	(tool)
In-Reply-To: <20120725131452.GG6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-4-git-send-email-fabiano@fidencio.org>
	<20120725131452.GG6596@redhat.com>
Message-ID: <CAK9pz9+r2r1Cq33Xa3k=qAQNCsNZs-Mr-xRL-wFGCs=FTjG1ew@mail.gmail.com>

On Wed, Jul 25, 2012 at 3:14 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Tue, Jul 24, 2012 at 10:05:15PM +0200, Fabiano Fid?ncio wrote:
>> Now, instead of return a string, the osinfo-install-script tool writes
>> that string in a file, passed as argument, that will be into a dir,
>> also passed as argument.
>>
>> If the filename argument is NULL, the output files will be written as
>> set in data/install-scripts/*.xml (in filename attribute from template
>> element):
>>     - Linuxes: fedora.ks
>>     - Windows 2k3r2 and older: windows.sif
>>     - Windows 2k8 and newer: windows.xml
>
> I think we should omit the filename argument entirely on the
> command line, because in the Windows case we're going to need
> to support generation of multiple output files. In such a case
> passing a filename on the ARGV doesn't make sense.
>
> I could see value in having an optional arg to specify a
> filename prefix. eg
>
>    # osinfo-install-script --prefix foo   win2k3r2
>
> would result in a file  'foo-windows.sif' being created.

agreed

>
>> If the dirname argument is NULL, the output files will be written in the
>> current directory.
>
> I think we could probably just make that an parameter flag
> instead eg.
>
>    # osinfo-install-script --output-dir /some/dir  win2k3r2
>

eorry, i didn't understand your point.
could you try to clarify it a bit more for me?

>
>> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
>> index 1f437a1..b2d7351 100644
>> --- a/osinfo/osinfo_install_script.c
>> +++ b/osinfo/osinfo_install_script.c
>> @@ -56,6 +56,7 @@ enum {
>>      PROP_TEMPLATE_DATA,
>>      PROP_PROFILE,
>>      PROP_PRODUCT_KEY_FORMAT,
>> +    PROP_OUTPUT_FILENAME,
>>  };
>
> Given an an <install-script> will allow multiple <template> args
> each with their own filename, I don't think it really makes sense
> to have an 'output-filename' property.  I'd support having an
> "output-prefix" property to support the usage I desribed earlier

agreed

>
>> @@ -686,6 +737,136 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
>>      return data.output;
>>  }
>>
>> +void osinfo_install_script_generate_output_async(OsinfoInstallScript *script,
>> +                                                 OsinfoOs *os,
>> +                                                 OsinfoInstallConfig *config,
>> +                                                 GFile *output_dir,
>
> This parameter appears to be ignored in this method which is strange ?

it's not so strange and i can explain why.
GOutputFile is used in the callback. Here, I just call the old
function to generate the script and, in the callback, I'll write the
script in the file. Makes sense? Still not clear?

>
>> +                                                 GCancellable *cancellable,
>> +                                                 GAsyncReadyCallback callback,
>> +                                                 gpointer user_data)
>> +{
>> +    osinfo_install_script_generate_async(script,
>> +                                         os,
>> +                                         config,
>> +                                         cancellable,
>> +                                         callback,
>> +                                         user_data);
>> +}
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From fabiano at fidencio.org  Wed Jul 25 14:28:54 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Wed, 25 Jul 2012 16:28:54 +0200
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <20120725131857.GH6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
	<20120725131857.GH6596@redhat.com>
Message-ID: <CAK9pz9J0Zs1qJyKftZzE0mrqLNukkuVT+EvkTe060cfh1k7qrg@mail.gmail.com>

On Wed, Jul 25, 2012 at 3:18 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
>> Now we need to set what are the configs that will be used in each
>> script. To set it, just add, in the .xml's script file:
>> <config>
>>   <param name="..." policy="mandatory"|"optional"/>
>> </config>
>>
>> With these configs loaded on our db, applications that uses/will use
>> libosinfo can check if some config *can* be set using:
>> osinfo_install_script_is_config(OsinfoInstallScript *script,
>>                                 const gchar* config);
>>
>> And if some config *must* be set using:
>> osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>>                                          const gchar* config);
>> ---
>>  data/install-scripts/fedora.xml           |    6 ++++
>>  data/install-scripts/windows-sif.xml      |    5 +++
>>  data/install-scripts/windows-unattend.xml |    9 ++++++
>>  data/schemas/libosinfo.rng                |   12 ++++++++
>>  osinfo/libosinfo.syms                     |    2 ++
>>  osinfo/osinfo_install_script.c            |   47 +++++++++++++++++++++++++++++
>>  osinfo/osinfo_install_script.h            |    9 +++++-
>>  osinfo/osinfo_loader.c                    |   19 ++++++++++++
>>  8 files changed, 108 insertions(+), 1 deletion(-)
>>
>> diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
>> index 338a570..a608cdb 100644
>> --- a/data/install-scripts/fedora.xml
>> +++ b/data/install-scripts/fedora.xml
>> @@ -1,6 +1,12 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="l10n-keyboard" policy="optional"/>
>> +      <param name="l10n-language" policy="optional"/>
>> +      <param name="l10n-timezone" policy="optional"/>
>> +    </config>
>>      <template filename="fedora.ks">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
>> index 46b24ae..77fc2c5 100644
>> --- a/data/install-scripts/windows-sif.xml
>> +++ b/data/install-scripts/windows-sif.xml
>> @@ -1,6 +1,11 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://microsoft.com/windows/sif'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="reg-product-key" policy="required"/>
>> +      <param name="user-realname" policy="required"/>
>> +    </config>
>>      <template filename="windows.sif">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
>> index 20e3e23..5698bc8 100644
>> --- a/data/install-scripts/windows-unattend.xml
>> +++ b/data/install-scripts/windows-unattend.xml
>> @@ -1,6 +1,15 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://microsoft.com/windows/unattend'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="hardware-arch" policy="optional"/>
>> +      <param name="l10n-language" policy="optional"/>
>> +      <param name="user-login" policy="optional"/>
>> +      <param name="user-password" policy="optional"/>
>> +      <param name="user-realname" policy="optional"/>
>> +      <param name="reg-product-key" policy="required"/>
>> +    </config>
>>      <template filename="windows.xml">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
>> index 7c8d7f7..1392f74 100644
>> --- a/data/schemas/libosinfo.rng
>> +++ b/data/schemas/libosinfo.rng
>> @@ -410,6 +410,12 @@
>>        <element name='profile'>
>>          <text/>
>>        </element>
>> +      <element name='config'>
>> +        <attribute name="name"/>
>> +        <attribute name="policy">
>> +          <ref name='policies'/>
>> +        </attribute>
>> +      </element>
>>        <element name='product-key-format'>
>>          <text/>
>>        </element>
>> @@ -479,4 +485,10 @@
>>        <param name="pattern">\w+://.*</param>
>>      </data>
>>    </define>
>> +
>> +  <define name='policies'>
>> +    <data type="string">
>> +      <param name="pattern">required|optional</param>
>> +    </data>
>> +  </define>
>>  </grammar>
>> diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
>> index 5a7de7b..2d64ebb 100644
>> --- a/osinfo/libosinfo.syms
>> +++ b/osinfo/libosinfo.syms
>> @@ -276,6 +276,8 @@ LIBOSINFO_0.2.0 {
>>       osinfo_install_script_generate_output_async;
>>       osinfo_install_script_get_profile;
>>       osinfo_install_script_get_uri;
>> +     osinfo_install_script_is_config;
>> +     osinfo_install_script_is_config_required;
>>       osinfo_install_scriptlist_new;
>>       osinfo_install_scriptlist_new_filtered;
>>       osinfo_install_scriptlist_new_union;
>> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
>> index b2d7351..6ce1c55 100644
>> --- a/osinfo/osinfo_install_script.c
>> +++ b/osinfo/osinfo_install_script.c
>> @@ -240,6 +240,26 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
>>      g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
>>  }
>>
>> +static GList *osinfo_install_script_get_config_required(OsinfoInstallScript *script)
>> +{
>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED);
>> +}
>> +
>> +static GList *osinfo_install_script_get_config_optional(OsinfoInstallScript *script)
>> +{
>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL);
>> +}
>> +
>> +static GList *osinfo_install_script_get_config(OsinfoInstallScript *script)
>> +{
>> +    GList *required = osinfo_install_script_get_config_required(script);
>> +    GList *optional = osinfo_install_script_get_config_optional(script);
>> +
>> +    return g_list_concat(required, optional);
>> +}
>
> This leaks memory associated with 'optional'
>
>> +
>>  static void
>>  osinfo_install_script_init (OsinfoInstallScript *list)
>>  {
>> @@ -336,6 +356,33 @@ const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *scri
>>                                           OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
>>  }
>>
>> +static gint osinfo_install_script_compare_configs(gconstpointer a, gconstpointer b)
>> +{
>> +    return (g_strcmp0((const gchar *)a, (const gchar *)b));
>> +}
>> +
>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> +                                         const gchar *config)
>> +{
>> +    if (g_list_find_custom(osinfo_install_script_get_config(script),
>
> This leaks the GList you're using
>
>> +                           config,
>> +                           osinfo_install_script_compare_configs) == NULL)
>> +        return FALSE;
>> +
>> +    return TRUE;
>> +}
>> +
>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> +                                                  const gchar *config)
>> +{
>> +    if (g_list_find_custom(osinfo_install_script_get_config_required(script),
>
> This leaks the GList you're using
>
>> +                           config,
>> +                           osinfo_install_script_compare_configs) == NULL)
>> +        return FALSE;
>> +
>> +    return TRUE;
>> +}
>> +
>>  struct OsinfoInstallScriptGenerate {
>>      GSimpleAsyncResult *res;
>>      OsinfoOs *os;
>> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
>> index d75ec93..cca257a 100644
>> --- a/osinfo/osinfo_install_script.h
>> +++ b/osinfo/osinfo_install_script.h
>> @@ -50,7 +50,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
>>  #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
>>  #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
>>  #define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
>> -
>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED    "required"
>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL    "optional"
>>
>>  /* object */
>>  struct _OsinfoInstallScript
>> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
>>                                                     GCancellable *cancellable,
>>                                                     GError **error);
>>
>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> +                                         const gchar *config);
>> +
>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> +                                                  const gchar *config);
>
> How about having an enum
>
>    typedef enum {
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
>    } OsinfoInstallScriptConfigPolicy;
>

it makes sense, but how xml will know about this enum?
should I set as 0, 1, 2 in the xml?

>
>   OsinfoInstallScriptConfigPolicy;
>   osinfo_install_script_get_config_parameter(OsinfoInstallScript *script,
>                                             const gchar *config);
>
> Also I'd probably add an API to list all parameters
>
>   GList *
>   osinfo_install_script_get_config_parameters(OsinfoInstallScript *script);
>

agreed

>
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From fabiano at fidencio.org  Wed Jul 25 14:29:36 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Wed, 25 Jul 2012 16:29:36 +0200
Subject: [virt-tools-list] [PATCH 5/6] Add function to fetch console
 keyboard layout
In-Reply-To: <20120725132050.GI6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-6-git-send-email-fabiano@fidencio.org>
	<20120725132050.GI6596@redhat.com>
Message-ID: <CAK9pz9JRHWoHn=261DfRp96dyi2Qh+ySGhHzaA4SUG0LqOKpTQ@mail.gmail.com>

On Wed, Jul 25, 2012 at 3:20 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Tue, Jul 24, 2012 at 10:05:17PM +0200, Fabiano Fid?ncio wrote:
>> Similar as found on Boxes, this function can be useful for all linuxes,
>> so, is better put it into our API
>
> While I agree its a useful function, I don't think it should go in
> the libosinfo library. I'd like libosinfo to remain as a metadata
> driven database.
>
> This is keymap lookup & translation is something that I think
> would be useful in a "libvirt installer" library, which is
> something we're about to start working on real soon. So I'll
> keep this code in mind for that.

okay, dropped for now :)

>
>
> Regards,
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From berrange at redhat.com  Wed Jul 25 14:34:21 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 15:34:21 +0100
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <CAK9pz9J0Zs1qJyKftZzE0mrqLNukkuVT+EvkTe060cfh1k7qrg@mail.gmail.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
	<20120725131857.GH6596@redhat.com>
	<CAK9pz9J0Zs1qJyKftZzE0mrqLNukkuVT+EvkTe060cfh1k7qrg@mail.gmail.com>
Message-ID: <20120725143421.GK6596@redhat.com>

On Wed, Jul 25, 2012 at 04:28:54PM +0200, Fabiano Fid?ncio wrote:
> On Wed, Jul 25, 2012 at 3:18 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> > On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
> >> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
> >>                                                     GCancellable *cancellable,
> >>                                                     GError **error);
> >>
> >> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
> >> +                                         const gchar *config);
> >> +
> >> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
> >> +                                                  const gchar *config);
> >
> > How about having an enum
> >
> >    typedef enum {
> >      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
> >      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
> >      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
> >    } OsinfoInstallScriptConfigPolicy;
> >
> 
> it makes sense, but how xml will know about this enum?
> should I set as 0, 1, 2 in the xml?

No, you only need deal with this is in the 'get_config_parameter' method.
At the XML level just use the strings as you already have.

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From fabiano at fidencio.org  Wed Jul 25 14:36:03 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Wed, 25 Jul 2012 16:36:03 +0200
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <20120725143421.GK6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
	<20120725131857.GH6596@redhat.com>
	<CAK9pz9J0Zs1qJyKftZzE0mrqLNukkuVT+EvkTe060cfh1k7qrg@mail.gmail.com>
	<20120725143421.GK6596@redhat.com>
Message-ID: <CAK9pz9+Mu-AKjUUwwxEKPP0JUi955+qRR5ePMY3rk_xhjRj92A@mail.gmail.com>

On Wed, Jul 25, 2012 at 4:34 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Wed, Jul 25, 2012 at 04:28:54PM +0200, Fabiano Fid?ncio wrote:
>> On Wed, Jul 25, 2012 at 3:18 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
>> > On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
>> >> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
>> >>                                                     GCancellable *cancellable,
>> >>                                                     GError **error);
>> >>
>> >> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> >> +                                         const gchar *config);
>> >> +
>> >> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> >> +                                                  const gchar *config);
>> >
>> > How about having an enum
>> >
>> >    typedef enum {
>> >      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
>> >      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
>> >      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
>> >    } OsinfoInstallScriptConfigPolicy;
>> >
>>
>> it makes sense, but how xml will know about this enum?
>> should I set as 0, 1, 2 in the xml?
>
> No, you only need deal with this is in the 'get_config_parameter' method.
> At the XML level just use the strings as you already have.

So, agreed!

>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From berrange at redhat.com  Wed Jul 25 14:37:29 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 15:37:29 +0100
Subject: [virt-tools-list] [PATCH 3/6] Change install script output
	(tool)
In-Reply-To: <CAK9pz9+r2r1Cq33Xa3k=qAQNCsNZs-Mr-xRL-wFGCs=FTjG1ew@mail.gmail.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-4-git-send-email-fabiano@fidencio.org>
	<20120725131452.GG6596@redhat.com>
	<CAK9pz9+r2r1Cq33Xa3k=qAQNCsNZs-Mr-xRL-wFGCs=FTjG1ew@mail.gmail.com>
Message-ID: <20120725143729.GL6596@redhat.com>

On Wed, Jul 25, 2012 at 04:27:01PM +0200, Fabiano Fid?ncio wrote:
> On Wed, Jul 25, 2012 at 3:14 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> >
> >> If the dirname argument is NULL, the output files will be written in the
> >> current directory.
> >
> > I think we could probably just make that an parameter flag
> > instead eg.
> >
> >    # osinfo-install-script --output-dir /some/dir  win2k3r2
> >
> 
> eorry, i didn't understand your point.
> could you try to clarify it a bit more for me?

IIUC, currently you have  output dir as a parameter

   # osinfo-install-script  OSNAME FILENAME  OUTPUT-DIR

I'm suggesting using a flag

   # osinfo-install-script --output-dir OUTPUT-DIR OSNAME

> >> @@ -686,6 +737,136 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
> >>      return data.output;
> >>  }
> >>
> >> +void osinfo_install_script_generate_output_async(OsinfoInstallScript *script,
> >> +                                                 OsinfoOs *os,
> >> +                                                 OsinfoInstallConfig *config,
> >> +                                                 GFile *output_dir,
> >
> > This parameter appears to be ignored in this method which is strange ?
> 
> it's not so strange and i can explain why.
> GOutputFile is used in the callback. Here, I just call the old
> function to generate the script and, in the callback, I'll write the
> script in the file. Makes sense? Still not clear?

What you describe cannot work, because nothing in this method ever
references the 'output_dir' parameter address.

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From fabiano at fidencio.org  Wed Jul 25 14:43:34 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Wed, 25 Jul 2012 16:43:34 +0200
Subject: [virt-tools-list] [PATCH 3/6] Change install script output
	(tool)
In-Reply-To: <20120725143729.GL6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-4-git-send-email-fabiano@fidencio.org>
	<20120725131452.GG6596@redhat.com>
	<CAK9pz9+r2r1Cq33Xa3k=qAQNCsNZs-Mr-xRL-wFGCs=FTjG1ew@mail.gmail.com>
	<20120725143729.GL6596@redhat.com>
Message-ID: <CAK9pz9J-bfDM8knP=_eBujS6=c1Z48LntrmrAtucugvu=ouPNw@mail.gmail.com>

On Wed, Jul 25, 2012 at 4:37 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Wed, Jul 25, 2012 at 04:27:01PM +0200, Fabiano Fid?ncio wrote:
>> On Wed, Jul 25, 2012 at 3:14 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
>> >
>> >> If the dirname argument is NULL, the output files will be written in the
>> >> current directory.
>> >
>> > I think we could probably just make that an parameter flag
>> > instead eg.
>> >
>> >    # osinfo-install-script --output-dir /some/dir  win2k3r2
>> >
>>
>> eorry, i didn't understand your point.
>> could you try to clarify it a bit more for me?
>
> IIUC, currently you have  output dir as a parameter
>
>    # osinfo-install-script  OSNAME FILENAME  OUTPUT-DIR
>
> I'm suggesting using a flag
>
>    # osinfo-install-script --output-dir OUTPUT-DIR OSNAME

As I did, we have:

# osinfo-install-script --output-dir OUTPUT-DIR --output-file FILENAME OSNAME

we agreed in remove --output-file, right?
about output dir i'm working in this way:
if OUTPUT-DIR is passed, the path will be OUTPUT-DIR, else, the path will be "."
do you agree with this behavior?

>
>> >> @@ -686,6 +737,136 @@ gchar *osinfo_install_script_generate(OsinfoInstallScript *script,
>> >>      return data.output;
>> >>  }
>> >>
>> >> +void osinfo_install_script_generate_output_async(OsinfoInstallScript *script,
>> >> +                                                 OsinfoOs *os,
>> >> +                                                 OsinfoInstallConfig *config,
>> >> +                                                 GFile *output_dir,
>> >
>> > This parameter appears to be ignored in this method which is strange ?
>>
>> it's not so strange and i can explain why.
>> GOutputFile is used in the callback. Here, I just call the old
>> function to generate the script and, in the callback, I'll write the
>> script in the file. Makes sense? Still not clear?
>
> What you describe cannot work, because nothing in this method ever
> references the 'output_dir' parameter address.

yeah, you're right.
I was supposing that callback passed to this function should use the output-dir

I'll think in a correct way to implement it.


>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From berrange at redhat.com  Wed Jul 25 14:48:00 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 25 Jul 2012 15:48:00 +0100
Subject: [virt-tools-list] [PATCH 3/6] Change install script output
	(tool)
In-Reply-To: <CAK9pz9J-bfDM8knP=_eBujS6=c1Z48LntrmrAtucugvu=ouPNw@mail.gmail.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-4-git-send-email-fabiano@fidencio.org>
	<20120725131452.GG6596@redhat.com>
	<CAK9pz9+r2r1Cq33Xa3k=qAQNCsNZs-Mr-xRL-wFGCs=FTjG1ew@mail.gmail.com>
	<20120725143729.GL6596@redhat.com>
	<CAK9pz9J-bfDM8knP=_eBujS6=c1Z48LntrmrAtucugvu=ouPNw@mail.gmail.com>
Message-ID: <20120725144800.GM6596@redhat.com>

On Wed, Jul 25, 2012 at 04:43:34PM +0200, Fabiano Fid?ncio wrote:
> On Wed, Jul 25, 2012 at 4:37 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> > On Wed, Jul 25, 2012 at 04:27:01PM +0200, Fabiano Fid?ncio wrote:
> >> On Wed, Jul 25, 2012 at 3:14 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> >> >
> >> >> If the dirname argument is NULL, the output files will be written in the
> >> >> current directory.
> >> >
> >> > I think we could probably just make that an parameter flag
> >> > instead eg.
> >> >
> >> >    # osinfo-install-script --output-dir /some/dir  win2k3r2
> >> >
> >>
> >> eorry, i didn't understand your point.
> >> could you try to clarify it a bit more for me?
> >
> > IIUC, currently you have  output dir as a parameter
> >
> >    # osinfo-install-script  OSNAME FILENAME  OUTPUT-DIR
> >
> > I'm suggesting using a flag
> >
> >    # osinfo-install-script --output-dir OUTPUT-DIR OSNAME
> 
> As I did, we have:
> 
> # osinfo-install-script --output-dir OUTPUT-DIR --output-file FILENAME OSNAME
> 
> we agreed in remove --output-file, right?

Yes.

> about output dir i'm working in this way:
> if OUTPUT-DIR is passed, the path will be OUTPUT-DIR, else, the path will be "."
> do you agree with this behavior?

Ok, yes, that's good. I was mis-reading the options you had
configured. Sorry for the confusion


Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From fabiano at fidencio.org  Thu Jul 26 10:27:35 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Thu, 26 Jul 2012 12:27:35 +0200
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <20120725131857.GH6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
	<20120725131857.GH6596@redhat.com>
Message-ID: <CAK9pz9LwyJ_ths+cGSSv2y96ZV9yNAfFupT53h0FSyewo4AKwQ@mail.gmail.com>

On Wed, Jul 25, 2012 at 3:18 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
>> Now we need to set what are the configs that will be used in each
>> script. To set it, just add, in the .xml's script file:
>> <config>
>>   <param name="..." policy="mandatory"|"optional"/>
>> </config>
>>
>> With these configs loaded on our db, applications that uses/will use
>> libosinfo can check if some config *can* be set using:
>> osinfo_install_script_is_config(OsinfoInstallScript *script,
>>                                 const gchar* config);
>>
>> And if some config *must* be set using:
>> osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>>                                          const gchar* config);
>> ---
>>  data/install-scripts/fedora.xml           |    6 ++++
>>  data/install-scripts/windows-sif.xml      |    5 +++
>>  data/install-scripts/windows-unattend.xml |    9 ++++++
>>  data/schemas/libosinfo.rng                |   12 ++++++++
>>  osinfo/libosinfo.syms                     |    2 ++
>>  osinfo/osinfo_install_script.c            |   47 +++++++++++++++++++++++++++++
>>  osinfo/osinfo_install_script.h            |    9 +++++-
>>  osinfo/osinfo_loader.c                    |   19 ++++++++++++
>>  8 files changed, 108 insertions(+), 1 deletion(-)
>>
>> diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
>> index 338a570..a608cdb 100644
>> --- a/data/install-scripts/fedora.xml
>> +++ b/data/install-scripts/fedora.xml
>> @@ -1,6 +1,12 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="l10n-keyboard" policy="optional"/>
>> +      <param name="l10n-language" policy="optional"/>
>> +      <param name="l10n-timezone" policy="optional"/>
>> +    </config>
>>      <template filename="fedora.ks">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
>> index 46b24ae..77fc2c5 100644
>> --- a/data/install-scripts/windows-sif.xml
>> +++ b/data/install-scripts/windows-sif.xml
>> @@ -1,6 +1,11 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://microsoft.com/windows/sif'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="reg-product-key" policy="required"/>
>> +      <param name="user-realname" policy="required"/>
>> +    </config>
>>      <template filename="windows.sif">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
>> index 20e3e23..5698bc8 100644
>> --- a/data/install-scripts/windows-unattend.xml
>> +++ b/data/install-scripts/windows-unattend.xml
>> @@ -1,6 +1,15 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://microsoft.com/windows/unattend'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="hardware-arch" policy="optional"/>
>> +      <param name="l10n-language" policy="optional"/>
>> +      <param name="user-login" policy="optional"/>
>> +      <param name="user-password" policy="optional"/>
>> +      <param name="user-realname" policy="optional"/>
>> +      <param name="reg-product-key" policy="required"/>
>> +    </config>
>>      <template filename="windows.xml">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
>> index 7c8d7f7..1392f74 100644
>> --- a/data/schemas/libosinfo.rng
>> +++ b/data/schemas/libosinfo.rng
>> @@ -410,6 +410,12 @@
>>        <element name='profile'>
>>          <text/>
>>        </element>
>> +      <element name='config'>
>> +        <attribute name="name"/>
>> +        <attribute name="policy">
>> +          <ref name='policies'/>
>> +        </attribute>
>> +      </element>
>>        <element name='product-key-format'>
>>          <text/>
>>        </element>
>> @@ -479,4 +485,10 @@
>>        <param name="pattern">\w+://.*</param>
>>      </data>
>>    </define>
>> +
>> +  <define name='policies'>
>> +    <data type="string">
>> +      <param name="pattern">required|optional</param>
>> +    </data>
>> +  </define>
>>  </grammar>
>> diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
>> index 5a7de7b..2d64ebb 100644
>> --- a/osinfo/libosinfo.syms
>> +++ b/osinfo/libosinfo.syms
>> @@ -276,6 +276,8 @@ LIBOSINFO_0.2.0 {
>>       osinfo_install_script_generate_output_async;
>>       osinfo_install_script_get_profile;
>>       osinfo_install_script_get_uri;
>> +     osinfo_install_script_is_config;
>> +     osinfo_install_script_is_config_required;
>>       osinfo_install_scriptlist_new;
>>       osinfo_install_scriptlist_new_filtered;
>>       osinfo_install_scriptlist_new_union;
>> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
>> index b2d7351..6ce1c55 100644
>> --- a/osinfo/osinfo_install_script.c
>> +++ b/osinfo/osinfo_install_script.c
>> @@ -240,6 +240,26 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
>>      g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
>>  }
>>
>> +static GList *osinfo_install_script_get_config_required(OsinfoInstallScript *script)
>> +{
>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED);
>> +}
>> +
>> +static GList *osinfo_install_script_get_config_optional(OsinfoInstallScript *script)
>> +{
>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL);
>> +}
>> +
>> +static GList *osinfo_install_script_get_config(OsinfoInstallScript *script)
>> +{
>> +    GList *required = osinfo_install_script_get_config_required(script);
>> +    GList *optional = osinfo_install_script_get_config_optional(script);
>> +
>> +    return g_list_concat(required, optional);
>> +}
>
> This leaks memory associated with 'optional'
>
>> +
>>  static void
>>  osinfo_install_script_init (OsinfoInstallScript *list)
>>  {
>> @@ -336,6 +356,33 @@ const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *scri
>>                                           OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
>>  }
>>
>> +static gint osinfo_install_script_compare_configs(gconstpointer a, gconstpointer b)
>> +{
>> +    return (g_strcmp0((const gchar *)a, (const gchar *)b));
>> +}
>> +
>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> +                                         const gchar *config)
>> +{
>> +    if (g_list_find_custom(osinfo_install_script_get_config(script),
>
> This leaks the GList you're using


Just to understand, why is this leaking the GList I'm using?

>
>> +                           config,
>> +                           osinfo_install_script_compare_configs) == NULL)
>> +        return FALSE;
>> +
>> +    return TRUE;
>> +}
>> +
>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> +                                                  const gchar *config)
>> +{
>> +    if (g_list_find_custom(osinfo_install_script_get_config_required(script),
>
> This leaks the GList you're using

Same here.

>
>> +                           config,
>> +                           osinfo_install_script_compare_configs) == NULL)
>> +        return FALSE;
>> +
>> +    return TRUE;
>> +}
>> +
>>  struct OsinfoInstallScriptGenerate {
>>      GSimpleAsyncResult *res;
>>      OsinfoOs *os;
>> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
>> index d75ec93..cca257a 100644
>> --- a/osinfo/osinfo_install_script.h
>> +++ b/osinfo/osinfo_install_script.h
>> @@ -50,7 +50,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
>>  #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
>>  #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
>>  #define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
>> -
>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED    "required"
>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL    "optional"
>>
>>  /* object */
>>  struct _OsinfoInstallScript
>> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
>>                                                     GCancellable *cancellable,
>>                                                     GError **error);
>>
>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> +                                         const gchar *config);
>> +
>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> +                                                  const gchar *config);
>
> How about having an enum
>
>    typedef enum {
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
>    } OsinfoInstallScriptConfigPolicy;
>
>
>   OsinfoInstallScriptConfigPolicy;
>   osinfo_install_script_get_config_parameter(OsinfoInstallScript *script,
>                                             const gchar *config);
>
> Also I'd probably add an API to list all parameters
>
>   GList *
>   osinfo_install_script_get_config_parameters(OsinfoInstallScript *script);
>
>
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From fabiano at fidencio.org  Thu Jul 26 10:51:42 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Thu, 26 Jul 2012 12:51:42 +0200
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <20120725131857.GH6596@redhat.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
	<20120725131857.GH6596@redhat.com>
Message-ID: <CAK9pz9+hnuQAJ2fr4ooneVN1DrTaRj=0PCFc86-SAh+EwjsShQ@mail.gmail.com>

On Wed, Jul 25, 2012 at 3:18 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
> On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
>> Now we need to set what are the configs that will be used in each
>> script. To set it, just add, in the .xml's script file:
>> <config>
>>   <param name="..." policy="mandatory"|"optional"/>
>> </config>
>>
>> With these configs loaded on our db, applications that uses/will use
>> libosinfo can check if some config *can* be set using:
>> osinfo_install_script_is_config(OsinfoInstallScript *script,
>>                                 const gchar* config);
>>
>> And if some config *must* be set using:
>> osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>>                                          const gchar* config);
>> ---
>>  data/install-scripts/fedora.xml           |    6 ++++
>>  data/install-scripts/windows-sif.xml      |    5 +++
>>  data/install-scripts/windows-unattend.xml |    9 ++++++
>>  data/schemas/libosinfo.rng                |   12 ++++++++
>>  osinfo/libosinfo.syms                     |    2 ++
>>  osinfo/osinfo_install_script.c            |   47 +++++++++++++++++++++++++++++
>>  osinfo/osinfo_install_script.h            |    9 +++++-
>>  osinfo/osinfo_loader.c                    |   19 ++++++++++++
>>  8 files changed, 108 insertions(+), 1 deletion(-)
>>
>> diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
>> index 338a570..a608cdb 100644
>> --- a/data/install-scripts/fedora.xml
>> +++ b/data/install-scripts/fedora.xml
>> @@ -1,6 +1,12 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="l10n-keyboard" policy="optional"/>
>> +      <param name="l10n-language" policy="optional"/>
>> +      <param name="l10n-timezone" policy="optional"/>
>> +    </config>
>>      <template filename="fedora.ks">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
>> index 46b24ae..77fc2c5 100644
>> --- a/data/install-scripts/windows-sif.xml
>> +++ b/data/install-scripts/windows-sif.xml
>> @@ -1,6 +1,11 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://microsoft.com/windows/sif'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="reg-product-key" policy="required"/>
>> +      <param name="user-realname" policy="required"/>
>> +    </config>
>>      <template filename="windows.sif">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
>> index 20e3e23..5698bc8 100644
>> --- a/data/install-scripts/windows-unattend.xml
>> +++ b/data/install-scripts/windows-unattend.xml
>> @@ -1,6 +1,15 @@
>>  <libosinfo version="0.0.1">
>>    <install-script id='http://microsoft.com/windows/unattend'>
>>      <profile>jeos</profile>
>> +    <config>
>> +      <param name="admin-password" policy="optional"/>
>> +      <param name="hardware-arch" policy="optional"/>
>> +      <param name="l10n-language" policy="optional"/>
>> +      <param name="user-login" policy="optional"/>
>> +      <param name="user-password" policy="optional"/>
>> +      <param name="user-realname" policy="optional"/>
>> +      <param name="reg-product-key" policy="required"/>
>> +    </config>
>>      <template filename="windows.xml">
>>        <xsl:stylesheet
>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
>> index 7c8d7f7..1392f74 100644
>> --- a/data/schemas/libosinfo.rng
>> +++ b/data/schemas/libosinfo.rng
>> @@ -410,6 +410,12 @@
>>        <element name='profile'>
>>          <text/>
>>        </element>
>> +      <element name='config'>
>> +        <attribute name="name"/>
>> +        <attribute name="policy">
>> +          <ref name='policies'/>
>> +        </attribute>
>> +      </element>
>>        <element name='product-key-format'>
>>          <text/>
>>        </element>
>> @@ -479,4 +485,10 @@
>>        <param name="pattern">\w+://.*</param>
>>      </data>
>>    </define>
>> +
>> +  <define name='policies'>
>> +    <data type="string">
>> +      <param name="pattern">required|optional</param>
>> +    </data>
>> +  </define>
>>  </grammar>
>> diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
>> index 5a7de7b..2d64ebb 100644
>> --- a/osinfo/libosinfo.syms
>> +++ b/osinfo/libosinfo.syms
>> @@ -276,6 +276,8 @@ LIBOSINFO_0.2.0 {
>>       osinfo_install_script_generate_output_async;
>>       osinfo_install_script_get_profile;
>>       osinfo_install_script_get_uri;
>> +     osinfo_install_script_is_config;
>> +     osinfo_install_script_is_config_required;
>>       osinfo_install_scriptlist_new;
>>       osinfo_install_scriptlist_new_filtered;
>>       osinfo_install_scriptlist_new_union;
>> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
>> index b2d7351..6ce1c55 100644
>> --- a/osinfo/osinfo_install_script.c
>> +++ b/osinfo/osinfo_install_script.c
>> @@ -240,6 +240,26 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
>>      g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
>>  }
>>
>> +static GList *osinfo_install_script_get_config_required(OsinfoInstallScript *script)
>> +{
>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED);
>> +}
>> +
>> +static GList *osinfo_install_script_get_config_optional(OsinfoInstallScript *script)
>> +{
>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL);
>> +}
>> +
>> +static GList *osinfo_install_script_get_config(OsinfoInstallScript *script)
>> +{
>> +    GList *required = osinfo_install_script_get_config_required(script);
>> +    GList *optional = osinfo_install_script_get_config_optional(script);
>> +
>> +    return g_list_concat(required, optional);
>> +}
>
> This leaks memory associated with 'optional'
>
>> +
>>  static void
>>  osinfo_install_script_init (OsinfoInstallScript *list)
>>  {
>> @@ -336,6 +356,33 @@ const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *scri
>>                                           OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
>>  }
>>
>> +static gint osinfo_install_script_compare_configs(gconstpointer a, gconstpointer b)
>> +{
>> +    return (g_strcmp0((const gchar *)a, (const gchar *)b));
>> +}
>> +
>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> +                                         const gchar *config)
>> +{
>> +    if (g_list_find_custom(osinfo_install_script_get_config(script),
>
> This leaks the GList you're using
>
>> +                           config,
>> +                           osinfo_install_script_compare_configs) == NULL)
>> +        return FALSE;
>> +
>> +    return TRUE;
>> +}
>> +
>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> +                                                  const gchar *config)
>> +{
>> +    if (g_list_find_custom(osinfo_install_script_get_config_required(script),
>
> This leaks the GList you're using
>
>> +                           config,
>> +                           osinfo_install_script_compare_configs) == NULL)
>> +        return FALSE;
>> +
>> +    return TRUE;
>> +}
>> +
>>  struct OsinfoInstallScriptGenerate {
>>      GSimpleAsyncResult *res;
>>      OsinfoOs *os;
>> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
>> index d75ec93..cca257a 100644
>> --- a/osinfo/osinfo_install_script.h
>> +++ b/osinfo/osinfo_install_script.h
>> @@ -50,7 +50,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
>>  #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
>>  #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
>>  #define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
>> -
>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED    "required"
>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL    "optional"
>>
>>  /* object */
>>  struct _OsinfoInstallScript
>> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
>>                                                     GCancellable *cancellable,
>>                                                     GError **error);
>>
>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>> +                                         const gchar *config);
>> +
>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>> +                                                  const gchar *config);
>
> How about having an enum
>
>    typedef enum {
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
>    } OsinfoInstallScriptConfigPolicy;

Thinking a bit about it.
In my implementation I'm keeping 2 lists of strings. One of required
configs and another one of optional configs,
So, I'm not keeping the policy on my lists, I'm keeping the config
property name. With this in mind, how this enum could help, so?
If you are thinking about a more efficient way to keep the lists,
please, let me know.


>
>
>   OsinfoInstallScriptConfigPolicy;
>   osinfo_install_script_get_config_parameter(OsinfoInstallScript *script,
>                                             const gchar *config);
>
> Also I'd probably add an API to list all parameters
>
>   GList *
>   osinfo_install_script_get_config_parameters(OsinfoInstallScript *script);
>
>
>
> Daniel
> --
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



-- 
Fabiano Fid?ncio



From stevelo at za.ibm.com  Thu Jul 26 19:03:06 2012
From: stevelo at za.ibm.com (Stephen Louw)
Date: Thu, 26 Jul 2012 21:03:06 +0200
Subject: [virt-tools-list] AUTO: Stephen Louw is out of the office
	(returning 07/31/2012)
Message-ID: <OF4DAE2AFC.E05DFA70-ON42257A47.0068A79C-42257A47.0068A79D@za.ibm.com>


I am out of the office until 07/31/2012.

For SCM related issues pls contact Anton van Niekerk. For AntiVirus related
issues please contact Brian Germishuys. Other isues, please contact Manoj
Sooka. Alternatively, I will respond to your message when I return.


Note: This is an automated response to your message  "virt-tools-list
Digest, Vol 37, Issue 19" sent on 26/7/2012 18:00:02.

This is the only notification you will receive while this person is away.



From hdegoede at redhat.com  Sat Jul 28 09:23:08 2012
From: hdegoede at redhat.com (Hans de Goede)
Date: Sat, 28 Jul 2012 11:23:08 +0200
Subject: [virt-tools-list] [PATCH virt-viewer] Improve USB redirection error
	dialogs
Message-ID: <1343467388-31841-1-git-send-email-hdegoede@redhat.com>

The USB errors passed into the error signal handlers are somewhat lowlevel
and quite detailed. Also they usually already contain a colon. so instead of
prefixing them with "USB redirection error", use USB Redirection Error as the
primary dialog text, and make the detailed message the secondary dialog text.

Signed-off-by: Hans de Goede <hdegoede at redhat.com>
---
 src/virt-viewer-app.c |   13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
index def52c5..47949d9 100644
--- a/src/virt-viewer-app.c
+++ b/src/virt-viewer-app.c
@@ -1153,7 +1153,18 @@ static void virt_viewer_app_usb_failed(VirtViewerSession *session G_GNUC_UNUSED,
                                        const gchar *msg,
                                        VirtViewerApp *self)
 {
-    virt_viewer_app_simple_message_dialog(self, _("USB redirection error: %s"), msg);
+    GtkWindow *window = GTK_WINDOW(virt_viewer_window_get_window(self->priv->main_window));
+    GtkWidget *dialog = gtk_message_dialog_new(window,
+                                    GTK_DIALOG_MODAL |
+                                    GTK_DIALOG_DESTROY_WITH_PARENT,
+                                    GTK_MESSAGE_ERROR,
+                                    GTK_BUTTONS_OK,
+                                    _("USB Redirection Error"));
+
+    gtk_message_dialog_format_secondary_text(GTK_MESSAGE_DIALOG(dialog),
+                                             "%s", msg);
+    gtk_dialog_run(GTK_DIALOG(dialog));
+    gtk_widget_destroy(dialog);
 }
 
 static void
-- 
1.7.10.4



From fabiano at fidencio.org  Sat Jul 28 16:20:24 2012
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Sat, 28 Jul 2012 18:20:24 +0200
Subject: [virt-tools-list] [PATCH 4/6] Add functions to know if a config
 can/must be set
In-Reply-To: <CAK9pz9+hnuQAJ2fr4ooneVN1DrTaRj=0PCFc86-SAh+EwjsShQ@mail.gmail.com>
References: <1343160318-20796-1-git-send-email-fabiano@fidencio.org>
	<1343160318-20796-5-git-send-email-fabiano@fidencio.org>
	<20120725131857.GH6596@redhat.com>
	<CAK9pz9+hnuQAJ2fr4ooneVN1DrTaRj=0PCFc86-SAh+EwjsShQ@mail.gmail.com>
Message-ID: <CAK9pz9JMGiQXrm4Btpe6KYfN4K5-ojhReF7Q38Q7cSHHKn68DA@mail.gmail.com>

On Thu, Jul 26, 2012 at 12:51 PM, Fabiano Fid?ncio <fabiano at fidencio.org> wrote:
> On Wed, Jul 25, 2012 at 3:18 PM, Daniel P. Berrange <berrange at redhat.com> wrote:
>> On Tue, Jul 24, 2012 at 10:05:16PM +0200, Fabiano Fid?ncio wrote:
>>> Now we need to set what are the configs that will be used in each
>>> script. To set it, just add, in the .xml's script file:
>>> <config>
>>>   <param name="..." policy="mandatory"|"optional"/>
>>> </config>
>>>
>>> With these configs loaded on our db, applications that uses/will use
>>> libosinfo can check if some config *can* be set using:
>>> osinfo_install_script_is_config(OsinfoInstallScript *script,
>>>                                 const gchar* config);
>>>
>>> And if some config *must* be set using:
>>> osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>>>                                          const gchar* config);
>>> ---
>>>  data/install-scripts/fedora.xml           |    6 ++++
>>>  data/install-scripts/windows-sif.xml      |    5 +++
>>>  data/install-scripts/windows-unattend.xml |    9 ++++++
>>>  data/schemas/libosinfo.rng                |   12 ++++++++
>>>  osinfo/libosinfo.syms                     |    2 ++
>>>  osinfo/osinfo_install_script.c            |   47 +++++++++++++++++++++++++++++
>>>  osinfo/osinfo_install_script.h            |    9 +++++-
>>>  osinfo/osinfo_loader.c                    |   19 ++++++++++++
>>>  8 files changed, 108 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
>>> index 338a570..a608cdb 100644
>>> --- a/data/install-scripts/fedora.xml
>>> +++ b/data/install-scripts/fedora.xml
>>> @@ -1,6 +1,12 @@
>>>  <libosinfo version="0.0.1">
>>>    <install-script id='http://fedoraproject.org/scripts/fedora/jeos'>
>>>      <profile>jeos</profile>
>>> +    <config>
>>> +      <param name="admin-password" policy="optional"/>
>>> +      <param name="l10n-keyboard" policy="optional"/>
>>> +      <param name="l10n-language" policy="optional"/>
>>> +      <param name="l10n-timezone" policy="optional"/>
>>> +    </config>
>>>      <template filename="fedora.ks">
>>>        <xsl:stylesheet
>>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>>> diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
>>> index 46b24ae..77fc2c5 100644
>>> --- a/data/install-scripts/windows-sif.xml
>>> +++ b/data/install-scripts/windows-sif.xml
>>> @@ -1,6 +1,11 @@
>>>  <libosinfo version="0.0.1">
>>>    <install-script id='http://microsoft.com/windows/sif'>
>>>      <profile>jeos</profile>
>>> +    <config>
>>> +      <param name="admin-password" policy="optional"/>
>>> +      <param name="reg-product-key" policy="required"/>
>>> +      <param name="user-realname" policy="required"/>
>>> +    </config>
>>>      <template filename="windows.sif">
>>>        <xsl:stylesheet
>>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>>> diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
>>> index 20e3e23..5698bc8 100644
>>> --- a/data/install-scripts/windows-unattend.xml
>>> +++ b/data/install-scripts/windows-unattend.xml
>>> @@ -1,6 +1,15 @@
>>>  <libosinfo version="0.0.1">
>>>    <install-script id='http://microsoft.com/windows/unattend'>
>>>      <profile>jeos</profile>
>>> +    <config>
>>> +      <param name="admin-password" policy="optional"/>
>>> +      <param name="hardware-arch" policy="optional"/>
>>> +      <param name="l10n-language" policy="optional"/>
>>> +      <param name="user-login" policy="optional"/>
>>> +      <param name="user-password" policy="optional"/>
>>> +      <param name="user-realname" policy="optional"/>
>>> +      <param name="reg-product-key" policy="required"/>
>>> +    </config>
>>>      <template filename="windows.xml">
>>>        <xsl:stylesheet
>>>          xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>>> diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
>>> index 7c8d7f7..1392f74 100644
>>> --- a/data/schemas/libosinfo.rng
>>> +++ b/data/schemas/libosinfo.rng
>>> @@ -410,6 +410,12 @@
>>>        <element name='profile'>
>>>          <text/>
>>>        </element>
>>> +      <element name='config'>
>>> +        <attribute name="name"/>
>>> +        <attribute name="policy">
>>> +          <ref name='policies'/>
>>> +        </attribute>
>>> +      </element>
>>>        <element name='product-key-format'>
>>>          <text/>
>>>        </element>
>>> @@ -479,4 +485,10 @@
>>>        <param name="pattern">\w+://.*</param>
>>>      </data>
>>>    </define>
>>> +
>>> +  <define name='policies'>
>>> +    <data type="string">
>>> +      <param name="pattern">required|optional</param>
>>> +    </data>
>>> +  </define>
>>>  </grammar>
>>> diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
>>> index 5a7de7b..2d64ebb 100644
>>> --- a/osinfo/libosinfo.syms
>>> +++ b/osinfo/libosinfo.syms
>>> @@ -276,6 +276,8 @@ LIBOSINFO_0.2.0 {
>>>       osinfo_install_script_generate_output_async;
>>>       osinfo_install_script_get_profile;
>>>       osinfo_install_script_get_uri;
>>> +     osinfo_install_script_is_config;
>>> +     osinfo_install_script_is_config_required;
>>>       osinfo_install_scriptlist_new;
>>>       osinfo_install_scriptlist_new_filtered;
>>>       osinfo_install_scriptlist_new_union;
>>> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
>>> index b2d7351..6ce1c55 100644
>>> --- a/osinfo/osinfo_install_script.c
>>> +++ b/osinfo/osinfo_install_script.c
>>> @@ -240,6 +240,26 @@ osinfo_install_script_class_init (OsinfoInstallScriptClass *klass)
>>>      g_type_class_add_private (klass, sizeof (OsinfoInstallScriptPrivate));
>>>  }
>>>
>>> +static GList *osinfo_install_script_get_config_required(OsinfoInstallScript *script)
>>> +{
>>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED);
>>> +}
>>> +
>>> +static GList *osinfo_install_script_get_config_optional(OsinfoInstallScript *script)
>>> +{
>>> +    return osinfo_entity_get_param_value_list(OSINFO_ENTITY(script),
>>> +                                              OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL);
>>> +}
>>> +
>>> +static GList *osinfo_install_script_get_config(OsinfoInstallScript *script)
>>> +{
>>> +    GList *required = osinfo_install_script_get_config_required(script);
>>> +    GList *optional = osinfo_install_script_get_config_optional(script);
>>> +
>>> +    return g_list_concat(required, optional);
>>> +}
>>
>> This leaks memory associated with 'optional'
>>
>>> +
>>>  static void
>>>  osinfo_install_script_init (OsinfoInstallScript *list)
>>>  {
>>> @@ -336,6 +356,33 @@ const gchar *osinfo_install_script_get_output_filename(OsinfoInstallScript *scri
>>>                                           OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME);
>>>  }
>>>
>>> +static gint osinfo_install_script_compare_configs(gconstpointer a, gconstpointer b)
>>> +{
>>> +    return (g_strcmp0((const gchar *)a, (const gchar *)b));
>>> +}
>>> +
>>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>>> +                                         const gchar *config)
>>> +{
>>> +    if (g_list_find_custom(osinfo_install_script_get_config(script),
>>
>> This leaks the GList you're using
>>
>>> +                           config,
>>> +                           osinfo_install_script_compare_configs) == NULL)
>>> +        return FALSE;
>>> +
>>> +    return TRUE;
>>> +}
>>> +
>>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>>> +                                                  const gchar *config)
>>> +{
>>> +    if (g_list_find_custom(osinfo_install_script_get_config_required(script),
>>
>> This leaks the GList you're using
>>
>>> +                           config,
>>> +                           osinfo_install_script_compare_configs) == NULL)
>>> +        return FALSE;
>>> +
>>> +    return TRUE;
>>> +}
>>> +
>>>  struct OsinfoInstallScriptGenerate {
>>>      GSimpleAsyncResult *res;
>>>      OsinfoOs *os;
>>> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
>>> index d75ec93..cca257a 100644
>>> --- a/osinfo/osinfo_install_script.h
>>> +++ b/osinfo/osinfo_install_script.h
>>> @@ -50,7 +50,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
>>>  #define OSINFO_INSTALL_SCRIPT_PROP_PROFILE            "profile"
>>>  #define OSINFO_INSTALL_SCRIPT_PROP_PRODUCT_KEY_FORMAT "product-key-format"
>>>  #define OSINFO_INSTALL_SCRIPT_PROP_OUTPUT_FILENAME    "output-filename"
>>> -
>>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_REQUIRED    "required"
>>> +#define OSINFO_INSTALL_SCRIPT_PROP_CONFIG_OPTIONAL    "optional"
>>>
>>>  /* object */
>>>  struct _OsinfoInstallScript
>>> @@ -128,6 +129,12 @@ const GFile *osinfo_install_script_generate_output(OsinfoInstallScript *script,
>>>                                                     GCancellable *cancellable,
>>>                                                     GError **error);
>>>
>>> +gboolean osinfo_install_script_is_config(OsinfoInstallScript *script,
>>> +                                         const gchar *config);
>>> +
>>> +gboolean osinfo_install_script_is_config_required(OsinfoInstallScript *script,
>>> +                                                  const gchar *config);
>>
>> How about having an enum
>>
>>    typedef enum {
>>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_NONE,
>>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_REQUIRED,
>>      OSINFO_INSTALL_SCRIPT_CONFIG_POLICY_OPTIONAL,
>>    } OsinfoInstallScriptConfigPolicy;
>
> Thinking a bit about it.
> In my implementation I'm keeping 2 lists of strings. One of required
> configs and another one of optional configs,
> So, I'm not keeping the policy on my lists, I'm keeping the config
> property name. With this in mind, how this enum could help, so?
> If you are thinking about a more efficient way to keep the lists,
> please, let me know.

I think Zeeshan answered all questions personally.
Many thanks,

>
>
>>
>>
>>   OsinfoInstallScriptConfigPolicy;
>>   osinfo_install_script_get_config_parameter(OsinfoInstallScript *script,
>>                                             const gchar *config);
>>
>> Also I'd probably add an API to list all parameters
>>
>>   GList *
>>   osinfo_install_script_get_config_parameters(OsinfoInstallScript *script);
>>
>>
>>
>> Daniel
>> --
>> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
>> |: http://libvirt.org              -o-             http://virt-manager.org :|
>> |: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
>> |: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|
>
>
>
> --
> Fabiano Fid?ncio



-- 
Fabiano Fid?ncio



From berrange at redhat.com  Mon Jul 30 08:17:15 2012
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 30 Jul 2012 09:17:15 +0100
Subject: [virt-tools-list] [PATCH virt-viewer] Improve USB redirection
 error dialogs
In-Reply-To: <1343467388-31841-1-git-send-email-hdegoede@redhat.com>
References: <1343467388-31841-1-git-send-email-hdegoede@redhat.com>
Message-ID: <20120730081715.GE11894@redhat.com>

On Sat, Jul 28, 2012 at 11:23:08AM +0200, Hans de Goede wrote:
> The USB errors passed into the error signal handlers are somewhat lowlevel
> and quite detailed. Also they usually already contain a colon. so instead of
> prefixing them with "USB redirection error", use USB Redirection Error as the
> primary dialog text, and make the detailed message the secondary dialog text.
> 
> Signed-off-by: Hans de Goede <hdegoede at redhat.com>
> ---
>  src/virt-viewer-app.c |   13 ++++++++++++-
>  1 file changed, 12 insertions(+), 1 deletion(-)
> 
> diff --git a/src/virt-viewer-app.c b/src/virt-viewer-app.c
> index def52c5..47949d9 100644
> --- a/src/virt-viewer-app.c
> +++ b/src/virt-viewer-app.c
> @@ -1153,7 +1153,18 @@ static void virt_viewer_app_usb_failed(VirtViewerSession *session G_GNUC_UNUSED,
>                                         const gchar *msg,
>                                         VirtViewerApp *self)
>  {
> -    virt_viewer_app_simple_message_dialog(self, _("USB redirection error: %s"), msg);
> +    GtkWindow *window = GTK_WINDOW(virt_viewer_window_get_window(self->priv->main_window));
> +    GtkWidget *dialog = gtk_message_dialog_new(window,
> +                                    GTK_DIALOG_MODAL |
> +                                    GTK_DIALOG_DESTROY_WITH_PARENT,
> +                                    GTK_MESSAGE_ERROR,
> +                                    GTK_BUTTONS_OK,
> +                                    _("USB Redirection Error"));
> +
> +    gtk_message_dialog_format_secondary_text(GTK_MESSAGE_DIALOG(dialog),
> +                                             "%s", msg);
> +    gtk_dialog_run(GTK_DIALOG(dialog));
> +    gtk_widget_destroy(dialog);
>  }
>  
>  static void

ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|



From jpsaba at gmail.com  Tue Jul 31 19:28:38 2012
From: jpsaba at gmail.com (Juan Saba)
Date: Tue, 31 Jul 2012 16:28:38 -0300
Subject: [virt-tools-list] Using  Virt_What on windows application
Message-ID: <002a01cd6f52$b0db83f0$12928bd0$@gmail.com>

Dear People, I am software developer self-made that try to protect my work
from installing on virtual machines and then distribute. 

 

I use Visual Basic 6. 

 

Is there a way to use Virt_What to stop my program is it trying to use on
VM?

 

Is there someone who can guide me on this?

 

Thank You in advance

 

Juan Saba

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120731/4520075d/attachment.htm>

From eblake at redhat.com  Tue Jul 31 19:41:13 2012
From: eblake at redhat.com (Eric Blake)
Date: Tue, 31 Jul 2012 13:41:13 -0600
Subject: [virt-tools-list] Using  Virt_What on windows application
In-Reply-To: <002a01cd6f52$b0db83f0$12928bd0$@gmail.com>
References: <002a01cd6f52$b0db83f0$12928bd0$@gmail.com>
Message-ID: <501834D9.9080100@redhat.com>

On 07/31/2012 01:28 PM, Juan Saba wrote:
> Dear People, I am software developer self-made that try to protect my work
> from installing on virtual machines and then distribute. 

Unfortunately, the primary audience of this list believes in Free
Software (as in freedom, not beer).  The very notion of writing software
that refuses to run in certain situations is anti-free, so you probably
won't get much sympathy for your goal.

> Is there a way to use Virt_What to stop my program is it trying to use on
> VM?

Nope.  The ideal hypervisor is one in which the fact that you are
running in a VM is transparent; while virt-what can generally make good
guesses about whether it is in a VM, it is documented up front that
there is no 100% reliable way to prove this.

That said, the fastest VMs happen to be the ones that _are_ aware that
they are virtual, so they can take advantage of hypercalls to the host
for closer cooperation and better overall efficiency, so virt-what will
usually get the right answer in most practical VM setups, provided that
virt-what has been ported to that particular guest OS.

-- 
Eric Blake   eblake at redhat.com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 620 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/virt-tools-list/attachments/20120731/6122a950/attachment.sig>