[FFmpeg-devel] [PATCH v7 2/3] avdevice/lavfi: support the dumpgraph with options

lance.lmwang at gmail.com lance.lmwang at gmail.com
Mon Jun 8 17:20:36 EEST 2020


On Fri, Jun 05, 2020 at 07:10:08PM +0200, Marton Balint wrote:
> 
> 
> On Fri, 5 Jun 2020, Nicolas George wrote:
> 
> > lance.lmwang at gmail.com (12020-05-25):
> > > From: Limin Wang <lance.lmwang at gmail.com>
> > > 
> > > Signed-off-by: Limin Wang <lance.lmwang at gmail.com>
> > > ---
> > >  doc/indevs.texi     | 19 +++++++++++++++++--
> > >  libavdevice/lavfi.c |  8 +++++---
> > >  2 files changed, 22 insertions(+), 5 deletions(-)
> > 
> > Either we case about backwards compatibility, and you cannot change the
> > type of the existing option at all, or we do not, and it is not
> > necessary to add another option.
> 
> I think it always depended on the fallout, we were never blindly strict
> about this.
> 
> > 
> > I move we do not care much about compatibility: the syntax was never
> > documented, except "1" should work. Then accept "1" and any valid
> > key-value string.
> 
> I suggested the separate option for the options because otherwise you can't
> do dumpgraph without options specified, so you cant make the code call
> avfilter_dump_graph(graph, NULL).
> 
> "1" as a special string might be OK until some deprecation period but not
> for all eternity if the options string wants to be an options list and if it
> must be parseable as an AVDictionary. That would be the same inconsistency
> which we tried to avoid all over the codebase. E.g. this option looks like a
> dictionary, except "1" is also accepted...

Nicolas, are you agree with Marton's opinion? If you're agree with that,
then I'll update with the other patch by your comments.


> 
> Regards,
> Marton
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request at ffmpeg.org with subject "unsubscribe".

-- 
Thanks,
Limin Wang


More information about the ffmpeg-devel mailing list