[Home]

Summary:ASTERISK-11610: Comment on AST_MASQ_NOSTREAM in park_call_full misleading; announcement *can* go to parked channel.
Reporter:David Woolley (davidw)Labels:
Date Opened:2008-03-11 08:43:41Date Closed:2008-03-12 17:49:19
Priority:TrivialRegression?No
Status:Closed/CompleteComponents:Resources/res_features
Versions:Frequency of
Occurrence
Related
Issues:
Environment:Attachments:
Description:The comment in this code fragment from main/features.c in the trunk (and res/res_features.c in 1.4) is not true, as simple invocation of the Park application will result in the announcement going to the channel being parked.  Furthermore, I cannot work out in what scenario it would be useful to abort the announcement if the masquerade gets committed, so I'm not sure what the comment is really trying to say.  (Masquerading isn't committed for the parking sequence until after the announcement is made, and the only interesting case I can think of is where the same channel is specified for both chan and peer.)


 495   if (peer && ((pu->parkingnum != -1 && ast_strlen_zero(orig_chan_name)) || !strcasecmp(peer->name, orig_chan_name))) { /* Only say number if it's a number and the channel hasn't been masqueraded away */
 496     /* Make sure we don't start saying digits to the channel being parked */
 497     ast_set_flag(peer, AST_FLAG_MASQ_NOSTREAM);
 498     ast_say_digits(peer, pu->parkingnum, "", peer->language);
 499     ast_clear_flag(peer, AST_FLAG_MASQ_NOSTREAM);
 500   }


****** ADDITIONAL INFORMATION ******

I've actually observed the announcement going to the parked channel on 1.4 (SVN as of about five weeks ago - that would be 100626 wrt to the file in question), but the code looks not to have any signicant change between that and the trunk version (or to the latest 1.4 SVN version), so I believe the issue is still valid for the trunk.

Note, I don't consider there to be any problem with the announcement going to the parked channel, in that case.
Comments:By: Joshua C. Colp (jcolp) 2008-03-11 09:59:23

This was put in to solve the instance where attended transfers are being used as blind transfers. So for instance:

A is talking to B
A wants to park, so they start an attended transfer to 700.
A begins hearing 7, but as they do they complete the attended transfer.
In the old code B would have continued hearing the digits (01 in my example) but with the above code they will not hear it.
By: Digium Subversion (svnbot) 2008-03-11 10:01:23

Repository: asterisk
Revision: 107465

U   trunk/main/features.c

------------------------------------------------------------------------
r107465 | file | 2008-03-11 10:01:20 -0500 (Tue, 11 Mar 2008) | 4 lines

Clarify comment about masquerading and playback of the parking slot.
(closes issue ASTERISK-11610)
Reported by: davidw

------------------------------------------------------------------------

http://svn.digium.com/view/asterisk?view=rev&revision=107465
By: Digium Subversion (svnbot) 2008-03-12 17:49:19

Repository: asterisk
Revision: 108355

_U  branches/1.6.0/
U   branches/1.6.0/main/features.c

------------------------------------------------------------------------
r108355 | russell | 2008-03-12 17:49:18 -0500 (Wed, 12 Mar 2008) | 12 lines

Merged revisions 107465 via svnmerge from
https://origsvn.digium.com/svn/asterisk/trunk

........
r107465 | file | 2008-03-11 10:05:17 -0500 (Tue, 11 Mar 2008) | 4 lines

Clarify comment about masquerading and playback of the parking slot.
(closes issue ASTERISK-11610)
Reported by: davidw

........

------------------------------------------------------------------------

http://svn.digium.com/view/asterisk?view=rev&revision=108355