diff options
| author | jstebbins <[email protected]> | 2012-09-11 09:45:21 +0000 |
|---|---|---|
| committer | jstebbins <[email protected]> | 2012-09-11 09:45:21 +0000 |
| commit | 16d4d5215502aa3b621013223beb8235469b677e (patch) | |
| tree | 7b317ab405bb661f09190ee9e63fc703c79a6bfa | |
| parent | 720a8bb4ad4eb4eee2ce7d145448b4c3f120b482 (diff) | |
libhb: handle overlapping timestamps better
This shouldn't happen except in the case of bad sources or an hb bug.
But if encountered, it provoked a feedback loop of frame dropping and
duplication where there should have been a single frame drop.
git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@4952 b64f7644-9d1e-0410-96f1-a4d463321fa5
| -rw-r--r-- | libhb/vfr.c | 8 |
1 files changed, 8 insertions, 0 deletions
diff --git a/libhb/vfr.c b/libhb/vfr.c index 8bd731b74..8e50d5b89 100644 --- a/libhb/vfr.c +++ b/libhb/vfr.c @@ -539,6 +539,14 @@ static int hb_vfr_work( hb_filter_object_t * filter, pv->total_lost_time += temp_duration; } + else if ( in->s.stop <= pv->last_stop[0] ) + { + // This is generally an error somewhere (bad source or hb bug). + // But lets do our best to straighten out the mess. + ++pv->drops; + hb_buffer_close(&in); + return HB_FILTER_OK; + } /* Cache frame start and stop times, so we can renumber time stamps if dropping frames for VFR. */ |