[asterisk-dev] Clearing pick-up groups on Zap/ channels
Olle E Johansson
oej at edvina.net
Mon Nov 6 07:16:26 MST 2006
5 nov 2006 kl. 16.43 skrev Gil Kloepfer:
> I've discovered that in configuration files such as the one for
> chan_zap (where all the options "cascade down" rather than being
> specific for each category, there is no way to indicate that
> channels have no pick-up group after the first group has been
> set. For example (simplified), in zapata.conf:
>
> [channels]
> ; These two phones are in call group/pick-up group #1
> callerid="Green Phone"<(256) 428-6121>
> callgroup=1
> pickupgroup=1
> channel => 1
> callerid="Black Phone"<(256) 428-6122>
> channel => 2
>
> ; These three channels should not be in ANY pick-up groups, but
> there
> ; is no way to clear the previous settings
> callerid="CallerID Phone" <(256) 428-6123>
> channel => 3
> callerid="CallerID Phone" <(256) 704-4666>
> channel => 4
> callerid="CallerID Phone" <(630) 372-1564>
> channel => 5
>
> In the case of channels 3, 4, and 5, there is no valid way to
> "clear out"
> the "current" callgroup / pickupgroup without encountering an error
> (you could say pickupgroup=none, but that actually throws an error
> from ast_get_group(), although it happens to work).
>
> I'd like to propose making a change to ast_get_group() to allow
> the option "none" that returns a ast_group_t containing no groups set,
> basically zero. So you could say 'callgroup=none' for example.
>
> Note that callgroup=0 would not do what I am suggesting because 0 is
> a valid group number. The group/callgroup/pickupgroup is actually a
> bit mask (bits numbered 0 to 63).
>
> If this seems reasonable, please indicate so and I will submit a patch
> for both 1.2 and -trunk (they are essentially the same patch).
Sounds very reasonable. Whether we can see this as a bug or a new
feature is up to Russell to decide. If it's a bug fix, which I think,
we need
patches for the 1.2 and 1.4 branches plus trunk. Please open an issue
in the bug tracker, upload patches and we'll discuss there.
/O
More information about the asterisk-dev
mailing list