[asterisk-dev] [Code Review] sig_pri and chan_misdn do not update connected line after call transfer

rmudgett at digium.com rmudgett at digium.com
Mon Oct 4 14:36:57 CDT 2010


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/958/
-----------------------------------------------------------

(Updated 2010-10-04 14:36:57.461316)


Review request for Asterisk Developers and Russell Bryant.


Changes
-------

Fixed the ringback problem in chan_misdn.  This patch caused the regression.  You cannot UNHOLD the channel after a masquerade and expect all channel technologies to still hear the ringback.  I had to rework the sig_pri and chan_misdn transfer in light of the ringback problem and deadlock concerns.  The masquerade needs to stop the ringback on party A's channel before actually doing the masquerade.


Summary
-------

This patch corrects an oversight and is against the 1.8 branch.  The sig_pri and chan_misdn channel drivers did not update the connected line information after a call transfer.


Diffs (updated)
-----

  /branches/1.8/channels/chan_misdn.c 290176 
  /branches/1.8/channels/sig_pri.c 290176 
  /branches/1.8/main/channel.c 290176 

Diff: https://reviewboard.asterisk.org/r/958/diff


Testing (updated)
-------

Call transfers still work and the connected line information is updated.  Also the expected caller/callee connected line interception macros are run on each channel.

Basic transfer actions:
A calls B
A places B on hold
A calls C
A can optionally swap between B and C by flash hook.
A transfers B to C

Test cases:
A transfers B to C using hangup.
A transfers B to C using hangup while C ringing.
A transfers C to B using hangup.
A transfers B to C using ECTExecute message.
A transfers B to C using ECTExecute message while C ringing.
A transfers C to B using ECTExecute message.

Note that chan_misdn does not support the ECTExecute method.


Thanks,

rmudgett




More information about the asterisk-dev mailing list