Thanks for the quick response. matroska_resync() is currently written to only return AVERROR_EOF, so using that error code directly wouldn't be any more informative. But your suggestion would work if matroska_resync() kept the error code returned by avio_seek(). Would such a change warrant splitting into a separate patch?
On Sun, Jul 31, 2016 at 10:26 AM, Nicolas George <geo...@nsup.org> wrote: > Le quartidi 14 thermidor, an CCXXIV, Sophia Wang a écrit : > > Since matroska->done is only set to 1 in matroska_resync(), the choice > > of error is made by checking the return value of matroska_resync() > > rather than checking matroska->done directly on the next > > while-iteration. > > This is not what your code do: > > > + if (matroska_resync(matroska, pos) < 0) > > + return avio_feof(s->pb) ? AVERROR_EOF : AVERROR(EIO); > > It checks the return value of matroska_resync() and then invents a > completely unrelated error code. > > What it should do is use the error code from matroska_resync(). As simple > as > that. > > Regards, > > -- > Nicolas George > _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
