On Wed, Nov 16, 2016 at 08:19:23AM -0500, Frediano Ziglio wrote:
> >
> > On Wed, Nov 16, 2016 at 06:21:24AM -0500, Frediano Ziglio wrote:
> > > >
> > > > On Wed, Nov 16, 2016 at 10:10:01AM +0100, Pavel Grunt wrote:
> > > > > ---
> > > > > server/reds.c | 2 +-
> > > > > server/tests/spice-codecs-parsing-test.c | 1 +
> > > > > 2 files changed, 2 insertions(+), 1 deletion(-)
> > > > >
> > > > > diff --git a/server/reds.c b/server/reds.c
> > > > > index 12a274c..eccc2ba 100644
> > > > > --- a/server/reds.c
> > > > > +++ b/server/reds.c
> > > > > @@ -3639,7 +3639,7 @@ static void reds_set_video_codecs(RedsState
> > > > > *reds,
> > > > > const char *codecs)
> > > > > {
> > > > > char *encoder_name, *codec_name;
> > > > >
> > > > > - if (strcmp(codecs, "auto") == 0) {
> > > > > + if (codecs == NULL || strcmp(codecs, "auto") == 0) {
> > > > > codecs = default_video_codecs;
> > > >
> > > > Or just if (g_strcmp0(codecs, "auto") == 0) {}
> > > >
> > > >
> > > > Acked-by: Christophe Fergeau <[email protected]>
> > > >
> > >
> > > What's the reason to allows NULL ?
> > > Personally I would nack this patch.
> >
> > Ah, we should not crash on NULL, I did not think too much about the
> > behaviour we want with NULL though. Using the default codecs when none
> > was specified does not sound too bad to me. But I'm fine with
> > g_return_if_fail(codecs != NULL); too.
> >
> > Christophe
> >
>
> I personally would use a "undefined behavior". Just don't use NULL,
> there is already a default value which is "auto".Then it's g_return_if_fail(codecs != NULL); That's fine with me. Christophe
signature.asc
Description: PGP signature
_______________________________________________ Spice-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/spice-devel
