[Bug 762715] mpegtsmux: alignment property and timestamps for Live UDP streaming

[GStreamer (GNOME Bugzilla)]

https://bugzilla.gnome.org/show_bug.cgi?id=762715

--- Comment #8 from jgorosdev <jgorosdev at gmail.com> ---
Hi,

Using Comment #2 as unit test shows the following results:

gst-launch-1.0 videotestsrc ! x264enc ! mpegtsmux alignment=7 ! fakesink
silent=false sync=true -v

Likewise, without patch we obtain;

/GstPipeline:pipeline0/GstFakeSink:fakesink0: last-message = chain   *******
(fakesink0:sink) (1316 bytes, dts: none, pts: none, duration: none, offset: -1,
offset_end: -1, flags: 00000000 ) 0x7f2d3cb0c910

And with patch:

/GstPipeline:pipeline0/GstFakeSink:fakesink0: last-message = chain   *******
(fakesink0:sink) (1316 bytes, dts: none, pts: 0:00:04.200000000, duration:
none, offset: -1, offset_end: -1, flags: 00000000 ) 0x7fb030b0e840

is the result.

Another test can be performed, trying the initial description test with the 
sample file number #1 and streaming it with stream pipeline #2.

When the plugin is patched, live stream is performed ( sync = true ).  However,
without the patch an EOS is received instantaneously, which proves that the
property  "sync = true "  does not work correctly, as it worked in version
1.4.4

Finally, just a note about the code comments: The DTS/PTS order is exactly the
same as gstbasesink.c does.

With respect to the second paragraph, I did not realize until now about that
rule in gstreamer.

Well, I think that's all. Thanks for your help. It was of great value for me.

Best Regards

Josu.

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