Hi,

On Wed, Apr 6, 2011 at 1:19 AM, Anton Khirnov <[email protected]> wrote:
> On Mon, Apr 04, 2011 at 09:38:32PM +0100, Måns Rullgård wrote:
>> Anton Khirnov <[email protected]> writes:
>>
>> > ---
>> >  libavformat/matroskaenc.c |    9 ++++++---
>> >  1 files changed, 6 insertions(+), 3 deletions(-)
>> >
>> > diff --git a/libavformat/matroskaenc.c b/libavformat/matroskaenc.c
>> > index e28ac42..c0427f9 100644
>> > --- a/libavformat/matroskaenc.c
>> > +++ b/libavformat/matroskaenc.c
>> > @@ -1140,10 +1140,13 @@ static int mkv_write_trailer(AVFormatContext *s)
>> >      }
>> >
>> >      if (pb->seekable) {
>> > -        cuespos = mkv_write_cues(pb, mkv->cues, s->nb_streams);
>> > +        if (mkv->cues->num_entries) {
>> > +            cuespos = mkv_write_cues(pb, mkv->cues, s->nb_streams);
>> > +
>> > +            ret = mkv_add_seekhead_entry(mkv->main_seekhead, 
>> > MATROSKA_ID_CUES, cuespos);
>> > +            if (ret < 0) return ret;
>> > +        }
>> >
>> > -        ret = mkv_add_seekhead_entry(mkv->main_seekhead, MATROSKA_ID_CUES 
>> >    , cuespos);
>> > -        if (ret < 0) return ret;
>> >          mkv_write_seekhead(pb, mkv->main_seekhead);
>> >
>> >          // update the duration
>> > --
>> > 1.7.4.1
>>
>> Looks OK.  Do I infer correctly from the lack of checksum updates that
>> this does not affect the regression tests?
>>
>
> No, that was an evil subversive attempt to break FATE on my way to world
> domination. Unfortunately you saw right through it, so I have no choice,
> but to attach a proper patch.

This means we don't write idnexes (or write empty indexes) for
audio-only mkv files? Adding a FIXME at the very least would be
useful, I suppose. I admit audio-only matroska files aren't used much,
but still...

Otherwise OK (and unrelated to this patch, also).

Ronald
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to