[FFmpeg-devel] [PATCH] IPv6 support v.2

Ronald S. Bultje rsbultje
Mon Nov 26 15:22:20 CET 2007


Hi Luca,

On Nov 26, 2007 2:41 AM, Luca Abeni <lucabe72 at email.it> wrote:

> My understanding was that you was going to post some updated patches after
> I committed my ones (you said: "I'll say it this way: there's multiple
> ways to get there, and I think your patches are a good start in one such
> direction to fix the issue, and I'll gladly pick up from there and resolve
> the whole thing").
>

Ok, but I still need to know how:

I could duplicate udp.c's ipv6_resolve_host() everywhere, generalize it and
keep it either in udp.c and export it or move it back into os_support.c
(instead of resolve_host()), or I could remove it completely and integrate
the call into the function caller in udp.c and duplicate that code into
tcp.c.

For udp.c, I would highly prefer to not only remove the ipv4-code path, but
also integrate all the single-use functions that we have now back into their
respective callers (udp_port(), udp_set_url()); this would basically undo
most of the work that you did so far, but I'd consider it a cleanup, since
if you remove the ipv4-code, the result is a bit like spaghetti...

Then there's a tenfold of code in sdp.c, ffserver.c and rtsp.c involving
changing in_addr / sockaddr_in into generalized containers holders and
inet_aton -> inet_pton or getnameinfo; my patches handled most of these
already and would probably still be useful afterwards (although I'd need to
change them to use addrinfo directly instead of this avnet stuff).

In the end, a lot of things and various ways to go, since I did 3 approaches
so far (uses error code from gethostbyname, switch to gethostbyname_r and
switch to getaddrinfo in a wrapper) and none of them got me anywhere, I'd
like a clear path for me before I start, I'm not going to waste another
month on these patches and see them rejected again.

But at this point I suggest you to wait for the libavnet changes before
> refreshing your patches.


OK.

Ronald




More information about the ffmpeg-devel mailing list