[asterisk-dev] [Code Review] Manage translation table between SIP and ISDN hangup causes

[Birger Harzenetter]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/2227/#review7492
-----------------------------------------------------------


I see a number of issues here.

In my experience the numeric SIP response is not always unique.
You may have to check the text as well to find out what exactly happened.

I think it would be a lot easier to use numeric cause codes as well.
Otherwise you need to document all the names as well.
(Are there names for all possible causes?)

(this is a core thing only related to this work)
There's some ambiguity on the cause side as well as we don't have the location data available.
I think that should be added to give this effort more meaning.
Some causes can mean quite different things depending on where they come from.
Likewise the sensible way to handle the situation can also vary depending on the location.

There are some translations that don't make sense to me.
RFC 3398 isn't the best reference I think (rejected > forbidden or out of order > bad gateway, what gateway?)
But I'll leave that for later.


- Birger


On Dec. 5, 2012, 5:34 a.m., Olle E Johansson wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/2227/
> -----------------------------------------------------------
> 
> (Updated Dec. 5, 2012, 5:34 a.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Summary
> -------
> 
> The SIP2CAUSE hangup code conversion tables has up to now been hard-coded in Asterisk. In some cases, like when building in-house ISDN/Q.SIG to SIP gateways, there's a need to manipulate this conversion. 
> 
> With this code, advanced users can add a "private" conversion. This is added in front of the built-in conversions.
> 
> Asterisk conversion tables does not change in this patch. Everything should work as before. To shrink the chan_sip.c file a small bit I decided to move this functionality into a new source code file.
> 
> Adding:
> - new source code file sip2cause.c and include file sip2cause.h
> - new configuration file sip2cause.conf
> 
> Reviewboard doesn't seem accept the new files, so they have to be found in the branch itself.
> 
> http://svn.digium.com/svn/asterisk/team/oej/earl-grey-sip2cause-configurable-trunk
> 
> The new files are:
> * http://svnview.digium.com/svn/asterisk/team/oej/earl-grey-sip2cause-configurable-trunk/configs/sip2cause.conf.sample
> * http://svnview.digium.com/svn/asterisk/team/oej/earl-grey-sip2cause-configurable-trunk/channels/sip/sip2cause.c
> * http://svnview.digium.com/svn/asterisk/team/oej/earl-grey-sip2cause-configurable-trunk/channels/sip/include/sip2cause.h
> 
> 
> This addresses bug ASTERISK-20759.
>     https://issues.asterisk.org/jira/browse/ASTERISK-20759
> 
> 
> Diffs
> -----
> 
>   /trunk/channels/chan_sip.c 377205 
>   /trunk/channels/sip/include/sip_utils.h 377205 
> 
> Diff: https://reviewboard.asterisk.org/r/2227/diff
> 
> 
> Testing
> -------
> 
> Tested all kinds of weird translations. This file should cause some errors (AST_CAUSE_SKREP doesn't exist, 903 is not a valid SIP reason code etc etc. 
> 
> [sip2cause]
> 604 => AST_CAUSE_SKREP
> 404 => UNALLOCATED
> 599 Bad => USER_BUSY
> 486 => NORMAL_CLEARING
> 603 => UNALLOCATED
>         
> [cause2sip]
> SKREP => 503 Service Failure
> UNALLOCATED => 903 Go to hell
> UNALLOCATED => 499 I don't want to do that.
> USER_BUSY => 503 I am not feeling well
> 
> 
> Thanks,
> 
> Olle E
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20121206/a50be2de/attachment.htm>