gst/audiorate/gstaudiorate.c: If a buffer doesn't have a timestamp, assume it's contiguous with the previous buffer, ...

Original commit message from CVS:
* gst/audiorate/gstaudiorate.c: (gst_audio_rate_chain):
If a buffer doesn't have a timestamp, assume it's contiguous with
the previous buffer, and synthesise timestamps appropriately.
This commit is contained in:
Michael Smith 2007-05-03 13:16:21 +00:00
parent 14f2bca596
commit 03e4592e41
2 changed files with 11 additions and 0 deletions

View file

@ -1,3 +1,9 @@
2007-05-03 Michael Smith <msmith@fluendo.com>
* gst/audiorate/gstaudiorate.c: (gst_audio_rate_chain):
If a buffer doesn't have a timestamp, assume it's contiguous with
the previous buffer, and synthesise timestamps appropriately.
2007-05-03 Edward Hervey <edward@fluendo.com>
* tests/check/elements/videorate.c: (GST_START_TEST):

View file

@ -510,6 +510,11 @@ gst_audio_rate_chain (GstPad * pad, GstBuffer * buf)
audiorate->in++;
in_time = GST_BUFFER_TIMESTAMP (buf);
if (in_time == GST_CLOCK_TIME_NONE) {
GST_DEBUG_OBJECT (audiorate, "no timestamp, using expected next time");
in_time = audiorate->next_ts;
}
in_size = GST_BUFFER_SIZE (buf);
in_samples = in_size / audiorate->bytes_per_sample;
/* get duration from the size because we can and it's more accurate */