[asterisk-dev] 13 November, 2012 New SIP stack update

[Daniel Pocock]

On 14/11/12 10:24, Saúl Ibarra Corretgé wrote:
> Mark Michelson wrote:
>> Hi folks,
>>
>> This is just a periodic update on the current planned work for the new
>> SIP channel driver for the next couple of weeks. I had meant to send
>> this out yesterday, but I didn't because I suck and am a terrible
>> person :)
>>
>> The hot topic on the mailing list at the moment is the proposed method
>> for Asterisk's use of a SIP stack. On that front, we've opened up
>> dialogs with the PJSIP and ReSIProcate developers to find out what our
>> role can be in their development and find out the reality of being able
>> to provide or use packaged versions of their SIP stacks.
>>
>> Things you can expect to see on the wiki soon:
>>
>> * Brent Eagles is doing an investigation into exactly what would be
>> required in order to write a C++ SIP channel driver.
>> * I will soon be starting on a page to diagram a proposed stack-agnostic
>> SIP architecture for the SIP channel driver.
>> * We're working to start creating issues in JIRA dealing with the tasks
>> that need to be done. These will make their way onto the main SIP
>> project page[1] once they are created.
>>
>> I'll send an e-mail out in a couple of weeks with what was accomplished
>> and what will be worked on next.
>>
> 
> Hi Mark,
> 
> Is Sofia out of the picture then? Seems so from the outside.
> 
> 

Hi Saúl,

I'd be interested to hear your feelings about pjsip/python-sipsimple

For example,

- why did you end up keeping patches locally, was it awkward to work
with upstream, or license issues, or something else?

- would you consider making a fork based on your patches and sharing it
with Asterisk as a standalone library?

- Or if Asterisk potentially went with reSIProcate, would you also
potentially have the resources to change python-sipsimple to use
reSIProcate?  I understand such change wouldn't happen quickly, of course

- same for Sofia SIP - if Asterisk uses Sofia and modernizes it, would
you be tempted to use it in place of your current solution?

Regards,

Daniel