[virt-tools-list] [libosinfo 2/2] win8: Add one more volume ID pattern

[Richard W.M. Jones]

On Tue, Nov 27, 2012 at 05:07:59PM +0200, Zeeshan Ali (Khattak) wrote:
> On Tue, Nov 27, 2012 at 5:04 PM, Richard W.M. Jones <rjones at redhat.com> wrote:
> > On Tue, Nov 27, 2012 at 04:12:10PM +0200, Zeeshan Ali (Khattak) wrote:
> >> From: "Zeeshan Ali (Khattak)" <zeeshanak at gnome.org>
> >>
> >> ---
> >>  data/oses/windows.xml.in                           |  2 +-
> >>  .../en_windows_8_enterprise_x86_dvd_917587.iso.txt | 29 ++++++++++++++++++++++
> >>  2 files changed, 30 insertions(+), 1 deletion(-)
> >>  create mode 100644 test/isodata/windows/win8/en_windows_8_enterprise_x86_dvd_917587.iso.txt
> >>
> >> diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
> >> index 2a58d8d..851c1b7 100644
> >> --- a/data/oses/windows.xml.in
> >> +++ b/data/oses/windows.xml.in
> >> @@ -737,7 +737,7 @@
> >>
> >>      <media arch="i386">
> >>        <iso>
> >> -        <volume-id>HB1_CCPA_X86FRE</volume-id>
> >> +        <volume-id>(HB1_CCPA_X86FRE)|(HRM_CENA_X86)</volume-id>
> >
> > Right .. as you guessed I downloaded the 32 bit ISO by accident :-(
> >
> > It's a bit unfortunate that libosinfo isn't distinguishing between
> > different editions of the base OS (ie. Desktop vs Server vs

BTW I meant to write "Client" instead of "Desktop" here.
Windows calls all the consumer editions "Client".
RHEL has a variant called "Desktop".

> > Enterprise).
> 
> We know of this issue and have plans to address it (I even have
> half-baked patch). Just don't know whats the best approach: Separate
> media entries for each variant or separate OS entries for each
> variant?

It's definitely more like a different OS than just different media.

Presumably using <derives-from> it should be easy to express
this?  ie:

  <os id="http://redhat.com/rhel/6.0">
    <variant>Server</variant>
     |
     V
  <os id="???">
    <derives-from id="http://redhat.com/rhel/6.0"/>
    <variant>Desktop</variant>

Note also that many Windows OSes you already have will simply need
<variant>Client</variant> or <variant>Server</variant> added to the
existing definitions.  They simply don't exist in a "null-variant"
form.  (The same is true for RHEL.  I've no idea about SLES).

Here is the relevant libguestfs documentation:

http://libguestfs.org/guestfs.3.html#guestfs_inspect_get_product_variant

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