The MOV_TRUN_DATA_OFFSET flag was always getting set, so data_offset is
being expected and read. The offset was computed correctly anyway during
fragment-flush. Alternatively, I could have selectively set the flag
only when base-data-offset is defined, but certain players (Chrome)
seem to fail when no data-offset is provided in the trun.
---
libavformat/movenc.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/libavformat/movenc.c b/libavformat/movenc.c
index a43752a..923cf92 100644
--- a/libavformat/movenc.c
+++ b/libavformat/movenc.c
@@ -3187,13 +3187,8 @@ static int mov_write_trun_tag(AVIOContext *pb,
MOVMuxContext *mov,
avio_wb24(pb, flags);
avio_wb32(pb, track->entry); /* sample count */
- if (mov->flags & FF_MOV_FLAG_OMIT_TFHD_OFFSET &&
- !(mov->flags & FF_MOV_FLAG_SEPARATE_MOOF) &&
- track->track_id != 1)
- avio_wb32(pb, 0); /* Later tracks follow immediately after the
previous one */
- else
- avio_wb32(pb, moof_size + 8 + track->data_offset +
- track->cluster[0].pos); /* data offset */
+ avio_wb32(pb, moof_size + 8 + track->data_offset +
+ track->cluster[0].pos); /* data offset */
if (flags & MOV_TRUN_FIRST_SAMPLE_FLAGS)
avio_wb32(pb, get_sample_flags(track, &track->cluster[0]));
--
1.9.3 (Apple Git-50)
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
