[asterisk-dev] [Code Review] 3671: media_formats: res_rtp_asterisk and other fixes

[Corey Farrell]

> On June 25, 2014, 7:45 a.m., Joshua Colp wrote:
> > /team/group/media_formats-reviewed-trunk/main/codec.c, lines 348-350
> > <https://reviewboard.asterisk.org/r/3671/diff/1/?file=60618#file60618line348>
> >
> >     Provided we documented that it gets bumped in ref, we could. We'd also have to clean up afterwards.

This is one of Matt's comments.  I suspect his question is why we don't have a function "ast_format_get_codec" that would return a bumped codec.  I'm going to let this be deferred until he does the API renames.


> On June 25, 2014, 7:45 a.m., Joshua Colp wrote:
> > /team/group/media_formats-reviewed-trunk/main/format.c, lines 200-202
> > <https://reviewboard.asterisk.org/r/3671/diff/1/?file=60620#file60620line200>
> >
> >     Just curious - does this happen?

This could happen if someone configures a blank format_cap (disallow=all) - the first format would be NULL (ast_best_codec).

Maybe an assert should go here until we get a better handle on things?  If NULL's to this procedure are possible we probably want to check this second (if both are NULL then they are equal).


> On June 25, 2014, 7:45 a.m., Joshua Colp wrote:
> > /team/group/media_formats-reviewed-trunk/include/asterisk/format.h, lines 326-333
> > <https://reviewboard.asterisk.org/r/3671/diff/1/?file=60615#file60615line326>
> >
> >     This says moved... but I don't see it... is it still around?

This is just Reviewboard trying too hard to find a match, it thinks the "\param" line plus 1 before and after were moved up to line 236 (format.h).  ast_format_is_slinear was actually renamed in a previous commit to ast_format_cache_is_slinear.  The old function name wasn't removed from this header so that's being done now.


On June 25, 2014, 7:45 a.m., Corey Farrell wrote:
> > For cases where is it doing a cleanup/copy it might be useful to add a BUGBUG for when replace exists.

ao2_replace already exists.  Do we want to just directly use that or create a #define ast_format_replace(dst,src) ao2_replace(dst,src) ?

I prefer we just use ao2_replace directly, I think this makes the code clearer.  I think of the #define ast_format_copy(format) as a compatibility macro that should go away during Matt's API rename.


- Corey


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3671/#review12307
-----------------------------------------------------------


On June 25, 2014, 7:52 a.m., Corey Farrell wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/3671/
> -----------------------------------------------------------
> 
> (Updated June 25, 2014, 7:52 a.m.)
> 
> 
> Review request for Asterisk Developers, Joshua Colp and Matt Jordan.
> 
> 
> Repository: Asterisk
> 
> 
> Description
> -------
> 
> This update gives media_formats the ability to receive a call using chan_sip.  Possibly other channel drivers might work, I haven't tried them.
> 
> * ast_format_cap_is_compatible_format needs to be checked against AST_FORMAT_CMP_NOT_EQUAL, not zero/non-zero.  All calls to ast_format_cap_is_compatible_format were fixed.
> * res_rtp_asterisk was updated by Matt Jordan, along with related changes to codec.c, codec.h, format.c, format.c and codec_builtin.c.
> * Switch ast_format_copy from function to macro to ao2_bump.  This allows REF_DEBUG to give better results.
> 
> 
> Diffs
> -----
> 
>   /team/group/media_formats-reviewed-trunk/res/res_speech.c 417190 
>   /team/group/media_formats-reviewed-trunk/res/res_rtp_asterisk.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/translate.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/frame.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/format.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/codec_builtin.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/codec.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/channel.c 417190 
>   /team/group/media_formats-reviewed-trunk/main/bridge.c 417190 
>   /team/group/media_formats-reviewed-trunk/include/asterisk/format.h 417190 
>   /team/group/media_formats-reviewed-trunk/include/asterisk/codec.h 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_unistim.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_skinny.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_sip.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_pjsip.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_oss.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_nbs.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_motif.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_mgcp.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_jingle.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_h323.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_gtalk.c 417190 
>   /team/group/media_formats-reviewed-trunk/channels/chan_alsa.c 417190 
>   /team/group/media_formats-reviewed-trunk/addons/chan_ooh323.c 417190 
>   /team/group/media_formats-reviewed-trunk/addons/chan_mobile.c 417190 
> 
> Diff: https://reviewboard.asterisk.org/r/3671/diff/
> 
> 
> Testing
> -------
> 
> Called from Asterisk 11 to a test server with this code, I was able to hear the 'invalid' message, everything seemed during the call.  I received TONS of ao2 frack's when stopping Asterisk.  The sip.conf peer on both Asterisk servers was setup for disallow=all / allow=ulaw.
> 
> 
> Thanks,
> 
> Corey Farrell
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140625/32870d91/attachment-0001.html>