[Asterisk-Users] h extension seems to wipe variables?

[John Todd]

I would imagine that using some sort of variable temp structure might 
help you.  However, the trick is keying the database tree so that in 
the event of multiple calls at the same time, you're looking at the 
"right" tree during your post-call processing.  Hmm... is ${CHANNEL} 
getting erased in the h context?  That would be the "best" key, I 
suppose.

JT

At 8:47 AM +0200 8/14/03, Florian Overkamp wrote:
>
>At 11:50 13-8-2003 -0500, you wrote:
>>Correct me if i'm wrong but doesn't the cdr modules log the call duration?
>
>True, but in some cases you would want to do more. I have built 
>IVR's that require some after-call handling. For this I tried to use 
>variables as well, but failed. The solution in my case was to pass 
>the variables from the extensions.conf and passing them as arguments 
>to the AGI script rather then reading them inside the AGI script.
>
>>  > I'm trying to do some custom call logging, and I want to call an AGI
>>>  script from a hangup handler to log call durations and things. Although
>>>  the script executes, it isn't retrieving variables from the AGI
>>>  interface. Looking closer, I realised the variables are actually getting
>>>  unset before the h extension is reached.
>>>
>>>  [foo]
>>>  s,1,SetVar,foo=bar
>>>  s,2,Play(audio/a-long-prompt)
>>>  h,1,AGI(log-call-duration.pl)
>>>
>>>  When I do an $agi->get_variable("foo") from the perl, I get the string
>>>  "noresponse" returned.
>>>
>>>  This all works fine if I don't call the AGI from the hangup extension,
>>>  but from a normal one instead.
>
>
>
>
>Met vriendelijke groet,
>Florian Overkamp
>ObSimRef BV (http://www.obsimref.com/)
>_______________________________________________
>Asterisk-Users mailing list
>Asterisk-Users at lists.digium.com
>http://lists.digium.com/mailman/listinfo/asterisk-users