[MPlayer-dev-eng] [PATCHES] Edit Decision Lists

[compn]

On Sun, 22 Jun 2008 20:35:24 +0200, "Olivier Rolland" <billl at users.sf.net> wrote:
>Hi,
>
>Here are two patches dealing with EDL support in mencoder.
>
>The first one is related to this bug
>http://bugzilla.mplayerhq.hu/show_bug.cgi?id=4: it gets rid of
>potential timestamp resets by using demuxer->stream_pts (if defined)
>instead of demuxer->video->sh->pts. Ultimately,
>demuxer->video->sh->pts (and the like) are replaced by calls to
>demuxer_get_current_time (demuxer). It's a little bit different with
>the audio stream because of the calc_a_pts function, that's why I'm
>using static variables to store previous values to detect pts resets.
>All in all, with this patch, the functionnal behaviour is exactely the
>same with movies without timestamp resets.

interesting if it works :)
i was thinking of an option that would give mplayer a timestamp of its own. so e.g. you could do mplayer -mptime -sub 1cd.srt file1.avi file2.avi and have it work with one subtitle file across multiple files...

how does your patch work for e.g. cat *avi | mplayer - 
(not that it is important)

>
>I've made a few tests with DVDs, AVIs, OGMs and MKVs with and without
>-hr-edl-seek and didn't have any regression.
>
>The second patch is less functional. It simply cleans out the EDL code
>in mencoder by removing unneeded parameters and tests, replacing a
>goto with a while and moving the EDL code from the main loop to its
>own function.
>
>I guess something similar can be done for mplayer.
>
>Regards,
>Olivier Rolland

any patches which can factor out common code between mplayer and mencoder  are welcome. likewise patches which split the giant mplayer.c .

-compn