[FFmpeg-devel] [PATCH] RTSP-MS 10/15: ASF header parsing

[Michael Niedermayer]

On Tue, Feb 03, 2009 at 12:15:47PM -0500, Ronald S. Bultje wrote:
> Hi,
> 
> On Tue, Feb 3, 2009 at 8:31 AM, Michael Niedermayer <michaelni at gmx.at> wrote:
> > to be honest i really think the rtp/rtsp code needs MUCH more cleanup
> > and documentation done before any new additions not less and later /
> > a promise to do 1 of 100 cleanups afterwards
> 
> You want documentation of struct members, or cleanup of code?

i want them documented then when i understand the code i can say if it
needs a cleanup.


> 
> The cleanup request is unfair, 

its unfair to ask me to review additions to the mess called our rt*p code.
i cant make much sense of large parts of our rt*p code currently ...


> Luca is working on a newly written RTSP
> demuxer (from scratch) and you can't stall other people's work for
> that.

luca, how far are you with that?


> Likewise, you can't ask me to compete with him. I can do the
> documentation of struct members, with the risk of wasting time at
> something Luca will replace sooner or later anyway. Luca, what's the
> current state? 

> Do you have a patch that works?

do you have some code that can saftely be commited? i mean something
that doesnt interfere with the current code, that way others could work
on it too even if it does not work yet

this also reminds me about mans ts demuxer ...


> 
> Also, I hope you're not expecting me to have a smaller cumulative LOC
> after addition of RTP/ASF or RDT support. These are
> non-standard-compliant, and thus require (much) extra code. The if
> (server == RM) do_special_options; else do_normal_options; spaghetti
> is kid of ugly in a way, but the alternative is to come up with a
> complete framework for doing all these tiny little things that each
> server wants (if (server_struct->special_options)
> server_struct->special_options(); else do_normal_options;). I
> personally think it'd just get more complex to understand.

i cant say anything here, its impossible to suggest some design / talk
about it without
knowing the various protocols well, i do know RTP/RTSP a little i dont
know the MS/REAL variants at all


> 
> If you insist, I'll do this. But I need a list of what has to be done.
> I understand the RTSP/RTP code quite well so you have to tell me
> what's not clear to you.

The whole highlevel structure is undocumented
i mean like
a diagram of what parts are part of initialization, and demuxing
like which function calls what and how packets are passed around

a little like demuxer->parser->decoder
but with much more details

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

If a bugfix only changes things apparently unrelated to the bug with no
further explanation, that is a good sign that the bugfix is wrong.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090203/e15592fd/attachment.pgp>