FaSTTUBe/GST-Tensordecoder-ov_ep
37
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

asfdemux: Improve seek behaviour for audio-only with no index

Browse Source 
Instead of seeking to seek_time - 5s in the hope of hitting a keyframe
for video, we can just seek to seek_time instead.
This commit is contained in:
Robert Swain 2010-02-16 12:59:54 +01:00
parent bde03ffc01
commit ad45fd1827
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
gst/asfdemux/gstasfdemux.c
View File

@ -661,9 +661,12 @@ gst_asf_demux_handle_seek_event (GstASFDemux * demux, GstEvent * event)
GST_TIME_ARGS (seek_time), offset, demux->data_offset, GST_TIME_ARGS (seek_time), offset, demux->data_offset,
demux->packet_size, packet); demux->packet_size, packet);
} else { } else {
/* Hackety hack, this sucks. We just seek to an earlier position /* FIXME: For streams containing video, seek to an earlier position in
* and let the sinks throw away the stuff before the segment start */ * the hope of hitting a keyframe and let the sinks throw away the stuff
if (flush && (demux->accurate || keyunit_sync)) { * before the segment start. For audio-only this is unnecessary as every
* frame is 'key'. */
if (flush && (demux->accurate || keyunit_sync)
&& demux->num_video_streams > 0) {
seek_time -= 5 * GST_SECOND; seek_time -= 5 * GST_SECOND;
if (seek_time < 0) if (seek_time < 0)
seek_time = 0; seek_time = 0;