[Bug 740243] New: mpegpsdemux: Issue with incoming TIME segment and base_time

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=740243
  GStreamer | gst-plugins-bad | git

           Summary: mpegpsdemux: Issue with incoming TIME segment and
                    base_time
    Classification: Platform
           Product: GStreamer
           Version: git
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: Normal
         Component: gst-plugins-bad
        AssignedTo: gstreamer-bugs at lists.freedesktop.org
        ReportedBy: bilboed at bilboed.com
         QAContact: gstreamer-bugs at lists.freedesktop.org
     GNOME version: ---


When mpegpsdemux works in push mode with incoming TIME segments, something is
completely wrong with segment calculation.

Incoming segment: start:0, stop:-1
First SCR detected becomes the base_time : base_time = first_scr (say 5m for
this example)
 ... But then the demuxer adjusts that *again* by the difference between
segment.start and first_scr, so essentially base_time = first_scr + first_scr
(we're at 10m now)

When the demuxer comes round to creating the output segments (in
_demux_send_segment) it takes the incoming one ... and then "adjusts" it by
base_time.

  We therefore end up pushing a segment with start == base_time == 2 *
first_scr

  The first PTS/DTS of that stream are around first_scr => everything gets
dropped until it reaches 2 * first_scr.

This was introduced by:
commit a9c0609ae84960edc47469c423b03eebac00ded7
Author: Mark Nauwelaerts <mark.nauwelaerts at collabora.co.uk>
Date:   Fri Oct 12 15:44:03 2012 +0200

    mpegdemux: handle upstream TIME segments

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.