[asterisk-dev] [Code Review]: INFO Record: on/off handled dynamically and separately with support for dynamic features.

[jrose]

> On Dec. 19, 2011, 4:16 p.m., jrose wrote:
> > Oh yeah, this needs to be indicated in CHANGES.  Noted and put on my to do list.
> 
> Olle E Johansson wrote:
>     I've always wanted a way to turn this off. Just a side note and a feature request that safely can be ignored.
> 
> jrose wrote:
>     I saw your comment in the code actually actually, and figured you'd appreciate this patch. Maybe you need an explanation? The fact is, the feature never did anything unless you mapped the built-in automon feature to some DTMF in features.conf in the first place. With this patch in place, you can just change the general settings to use "" and then they never should never find a feature to map against, even when you've given the incoming peer the ability to use something like automon in the first place.
>     
>     Alternatively, you could set up a feature that just invokes no-op or something. There really isn't any need for a more comprehensive feature to actually shut off INFO record detection if that's what you are asking for.
> 
> jrose wrote:
>     Addendum:  Ok, maybe that needs a little better explanation... so basically if you want this feature to not have any function at all, you'd do something like this...
>     <in sip.conf>
>     ...
>     [general]
>     ...
>     recordonfeature=
>     recordofffeature=
>     ...
>     
>     And then if you want to override that, you can do so for specific peers the same as you would for just about anything.
> 
> Olle E Johansson wrote:
>     Do we have any other feature you have to disable like that? Can we implement a keyword "off" or "false" or something that's similar to other configuration options?

Is there any configuration option in particular you are referencing?  As far as I know, neither 'off' nor 'false' is ever used in the context of chan_sip. Chan_dahdi on the other hand shows setting to blank strings in a few examples like msn and unknownprefix.  It isn't quite the same since I think in these cases the defaults are already blank.

One of the problems of coming up with a keyword is that there really aren't any limits to what you can call a dynamic feature.  Most everything that has stuff like "yes", "no", and other stuff is some kind of state from a list of finite states.


- jrose


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/1634/#review5032
-----------------------------------------------------------


On Dec. 20, 2011, 10:06 a.m., jrose wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/1634/
> -----------------------------------------------------------
> 
> (Updated Dec. 20, 2011, 10:06 a.m.)
> 
> 
> Review request for Asterisk Developers, David Vossel and Matt Jordan.
> 
> 
> Summary
> -------
> 
> The reporter was was wanting to submit a feature change which would have allowed Record requests to search through dynamic features for a feature named automon rather than simply failing if it wasn't found in builtin features.
> 
> David Vossel responded to the issue suggesting he would rather see a setting added to specify the feature.
> 
> This patch goes a step beyond that by allowing the user to specify which feature to use when receiving the message separately depending on whether the incoming message contained Record: on or Record: off.  These applications can be set to the same value when using the built in features like automon/automixmon, or they can be set separately so that on always triggers the recording feature while off always triggers the  appropriate feature to shut the recording application off. In this case, the SNOM phone which this feature was made for will typically show the correct recording status indication after the button is pressed even if it previously fell out of sync somehow. *I would guess by the user being clumsy or mischievous and pressing record over and over again really quickly*
> 
> Thanks to this new setup, it is possible to forgo recording all together in favor of having Record: on louden the volume before blasting the other peer with weasels and then have Record: off raise the volume even more before blasting the other peer with iguanas.  Glorious.
> 
> Sadly, not a single aspect of the reporter's patch has remained, though oddly enough the first chunk in chan_sip was added just a few days ago in an unrelated issue.
> 
> 
> This addresses bug ASTERISK-16507.
>     https://issues.asterisk.org/jira/browse/ASTERISK-16507
> 
> 
> Diffs
> -----
> 
>   /trunk/channels/chan_sip.c 348210 
>   /trunk/channels/sip/include/sip.h 348210 
>   /trunk/configs/sip.conf.sample 348210 
>   /trunk/main/features.c 348210 
> 
> Diff: https://reviewboard.asterisk.org/r/1634/diff
> 
> 
> Testing
> -------
> 
> Use of both built-in and dynamic features applied separately were performed with the settings being applied both globally and to peers individually.  Testing was also done to check that the default settings applied as intended.
> 
> 
> Thanks,
> 
> jrose
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20111220/cd0fe802/attachment.htm>