[libvirt] [PATCH 09/11] virsh: convert command line parsing to use GOptionContext

[Pavel Hrdina]

On Tue, Oct 01, 2019 at 10:48:50AM +0100, Daniel P. Berrangé wrote:
> On Tue, Oct 01, 2019 at 11:37:17AM +0200, Pavel Hrdina wrote:
> > On Fri, Sep 27, 2019 at 06:17:31PM +0100, Daniel P. Berrangé wrote:
> > > The GOptionContext API has the benefit over getopt_long that it will
> > > automatically handle --help output formatting.
> > > 
> > > Signed-off-by: Daniel P. Berrangé <berrange at redhat.com>
> > > ---
> > >  tools/virsh.c | 303 ++++++++++++++++++++++----------------------------
> > >  1 file changed, 135 insertions(+), 168 deletions(-)
> > > 
> > > diff --git a/tools/virsh.c b/tools/virsh.c
> > > index ec20f35a77..6c469ff576 100644
> > > --- a/tools/virsh.c
> > > +++ b/tools/virsh.c
> > > @@ -23,7 +23,6 @@
> > >  
> > >  #include <stdarg.h>
> > >  #include <unistd.h>
> > > -#include <getopt.h>
> > >  #include <sys/time.h>
> > >  #include <fcntl.h>
> > >  #include <time.h>
> > > @@ -445,53 +444,36 @@ virshDeinit(vshControl *ctl)
> > >  }
> 
> > > +        { "version", 'v', G_OPTION_FLAG_OPTIONAL_ARG,
> > > +          G_OPTION_ARG_CALLBACK, virshVersion,
> > > +          _("print short version"), "[short]" },
> > > +        { "version", 'V', 0,
> > > +          G_OPTION_ARG_NONE, &version,
> > > +          _("print long version"), "long" },
> > 
> > We should be able to have both -v and -V call virshVersion if the
> > functions will look like this:
> > 
> > static gboolean
> > virshVersion(const gchar *option_name,
> >              const gchar *value,
> >              gpointer data,
> >              GError **error G_GNUC_UNUSED)
> > {
> >     vshControl *ctl = data;
> > 
> >     if (STREQ(option_name, "-V") || STREQ_NULLABLE(value, "long"))
> >         virshShowVersion(ctl);
> >     else
> >         puts(VERSION);
> > 
> >     exit(EXIT_SUCCESS);
> > }
> > 
> > That way we will have only a single place where the version printing
> > code is.
> 
> Hmm, so "option_name" in the callback is the option that the user
> actually passed ? I was thinking it was the option name from the
> static array - ie it would always be "version". If you're right
> though, this is definitely my preference.

It is the option actually used by the user, from the glib docs:

option_name     The name of the option being parsed. This will be either
                a single dash followed by a single letter (for a short
                name) or two dashes followed by a long option name.

Pavel
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20191001/7e9ed8a1/attachment-0001.sig>