Hi,

2011/8/17 Måns Rullgård <m...@mansr.com>:
> Kostya <kostya.shish...@gmail.com> writes:
>
>> On Mon, Aug 15, 2011 at 09:47:54PM +0100, Måns Rullgård wrote:
>>> Kostya Shishkov <kostya.shish...@gmail.com> writes:
>>>
>>> > ---
>>> >  avconv.c |    2 +-
>>> >  1 files changed, 1 insertions(+), 1 deletions(-)
>>> >
>>> > diff --git a/avconv.c b/avconv.c
>>> > index 18c5f07..74db59f 100644
>>> > --- a/avconv.c
>>> > +++ b/avconv.c
>>> > @@ -1186,7 +1186,7 @@ static void do_video_out(AVFormatContext *s,
>>> >                 ost->resample_width, ost->resample_height, 
>>> > av_get_pix_fmt_name(ost->resample_pix_fmt),
>>> >                 dec->width         , dec->height         , 
>>> > av_get_pix_fmt_name(dec->pix_fmt));
>>> >          if(!ost->video_resample)
>>> > -            exit_program(1);
>>> > +            ost->video_resample = 1;
>>> >      }
>>> >
>>> >  #if !CONFIG_AVFILTER
>>> > --
>>>
>>> Is this a good idea?  It seems to me the correct thing to do would be to
>>> reconfigure the encoder if supported and only if that fails start
>>> scaling.
>>
>> I don't know many formats that allow that.
>
> Whatever input formats would trigger this action obviously support it.

H264 for example. We have videos on roundup that feature this.
Presumably, in the ideal infinite-time case, we need flags for things
like size change just as much as we do for container-supports-vfr.

I guess for now this hack is OK though, until someone feels like
implementing this properly.

Ronald
_______________________________________________
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to