[MPlayer-dev-eng] [PATCH] demux_real.c : make the mess even

[Arpi]

Hi,

> > > > > btw2 there is a hack in demux_real, because without that sound took
> > > > > a long time to catch up with the video. This solves it because until
> > > > > the first keyframe arrives, the video is still - and by that time
> > > > > audio will be in sync with video. Or maybe not - could some one tell
> > > > > me what is the real reason than?:)
> > > >
> > > > which hack?
> > >
> > > that with I / P frames
> >
> > hmm?
> 
> if(!(s[0]&0x8) || !(s[0]&0x10)){ // P || I  frame -> swap timestamps

it ha snothing with audio or sth you wrote. it's for simple I/P vs B frame
reordering. (I/P frames are coded before the B frame sbut displayed after,
so we have to swap their timestamps)

> > > there is no debug stuff afaict
> > > or you mean the number of packets reached 0 ?
> > > that is no debug, that is a bug there, commented ;)
> >
> > bug? why?
> according to the fine documentation if num of packets
> is reached, a new data chunk could be coming (with the
> chunk header etc., xine checks for this)

nice. then fix that bug, i don't remember seeing anything like that in your
"patch"

> > so if you have real fixes, not hacks and not 'i uncomment this part of code
> > because i think it's a cool thing' then send an usable patch.
> sure
> i don't think it is cool
> sorry for sending a patch that fixes something that
> works in xine, but not in mplayer
> my bad

not it doesn't fix but probably breaks many things.
mayeb it fixes one or a few streams.
and i don't care if it works in xine or not, xine is a bug buggy hack.
they steal our demuxer and trying to hack it without understanding how it
works.

> before telling me these very usefull thoughts, did you actually look at 
> demux_real? it is one big hack, nobody knows why stuff is there (as it is

wrong. for example i know.

> not documented by realnetworks of course), full of gotos and printfs

ah you're anti-goto coder...

> so I really don't get what you are talking about

i'm talking about code ferezee and a messy ugly shit you sent named 'patch'.
and you expect it to be commited? lol.

> and it is not fun resending every patch 5 times, just because one deleted

easy: then send usable patch at first time...

> well if you have a lot of time, go ahead and fix it The Right Way - if you
no i don't have time

> are into duplicating others work, and telling them how lame they are
> than you should do that imho

i didn't aks you to fix it. you wanted to fix but failed, your 'patch' is
everything but a patch.

btw be patient, as i promised, i'll leave mainatinership right after 0.90
released, maybe the new maintainer will like your "patch" (i doubt).


A'rpi / Astral & ESP-team

--
Developer of MPlayer, the Movie Player for Linux - http://www.MPlayerHQ.hu
    "However, many people beg for its inclusion in Debian. Why?" - Gabucino
  "Because having new software in Debian is good." - Josselin Mouette
"Because having good software in Debian is new." - Gabucino