[Bug 736655] basesink: preroll issue for some clips which audio is shorter than video

[GStreamer (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=736655

--- Comment #105 from kevin <kevinbing.song at gmail.com> ---
(In reply to Sebastian Dröge (slomo) from comment #100)
> Review of attachment 295411 [details] [review]:
> 
> ::: tests/check/pipelines/streamsynchronizer.c
> @@ +37,3 @@
> +
> +      g_mutex_lock (&test_mutex);
> +      if (gst_event_parse_group_id (event, &group_id)) {
> 
> This should always succeed here, just add an assertion instead of the else
> case
> 
Accepted.
> @@ +126,3 @@
> +      GST_STATE_CHANGE_FAILURE);
> +
> +  g_usleep (1 * G_USEC_PER_SEC);
> 
> Why the sleeps?
> 
Don't need it. Removed.
> @@ +131,3 @@
> +
> +  g_object_unref (bus);
> +  g_object_unref (G_OBJECT (pipeline));
> 
> Use gst_object_unref() and gst_object_ref(), and the G_OBJECT() casts are
> not needed then
> 
Accepted.
> @@ +144,3 @@
> +  run_streamsynchronizer_handle_eos ("videotestsrc name=videosrc ! "
> +      "streamsynchronizer name=streamsync audiotestsrc name=audiosrc
> num-buffers=0 ! "
> +      "streamsync. streamsync. ! fakesink sync=true streamsync. ! fakesink
> sync=true");
> 
> Make a separate testcase for each of these pipelines. That will make the
> test runner output more informative :)
> 
Accepted.
> @@ +156,3 @@
> +
> +  suite_add_tcase (s, tc_chain);
> +  tcase_add_test (tc_chain, test_streamsynchronizer_handle_eos);
> 
> Maybe we should also add some tests for the normal behaviour of
> streamsynchronizer while we're at it. Testing that it passes through data
> correctly, and especially aligns stream starts when they're switched to a
> new group.
Maybe can add the test when fix segment start which from different source is
different:)

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.