[FFmpeg-devel] how to test MMS over TCP protocol in ffmpeg

[zhentan feng]

Hi

On Sat, Mar 6, 2010 at 7:45 AM, Ronald S. Bultje <rsbultje at gmail.com> wrote:

> Hi,
>
> On Fri, Mar 5, 2010 at 1:00 PM, Alan Steremberg <alans at wunderground.com>
> wrote:
> > On Thu, Mar 4, 2010 at 6:59 AM, Ronald S. Bultje <rsbultje at gmail.com>
> wrote:
> >> On Thu, Mar 4, 2010 at 4:23 AM, zhentan feng <spyfeng at gmail.com> wrote:
> >> > Is this the method for testing the code ok?
> >>
> >> Yes. But I'm surprised the code didn't work to begin with. Did you
> >> change the original patch, or is this the original one? AFAIK, many
> >> iPhone apps successfully use this patch (e.g. [1]). I was hoping the
> >> patch would work already.
> >
> > I never got the patch to work for Wunder Radio, i just take the data from
> > libmms , and feed it in on my own.
>
> Ugh and shit. :-).
>
> Zhentan, if this is true, I'd recommend working from libmms as a
> starting point. It's LGPL so license is not an issue here. It's also
> relatively small and the individual subprotocols are well-contained in
> their own files. You shouldn't need the wrapper, just rip the relevant
> code from the mmst part in libmms and work from there.
>
>
I'll check libmms.


> Sorry for this, I had actually assumed Ryan's patches worked... :-(.
> I'll help you test next week to make sure we're moving in the right
> direction here.
>
> [...]
>
As mentioned in my last mail, Ryan's original code can be found by [1].
there is some differences with my primary work in soc svn.
I am trying to fix the bug for the current version. Certainly, libmms will
be a good reference.

zhentan

[1]
http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20070102/3b08b76a/attachment.obj
-- 
Best wishes~