Summary:	ASTERISK-02284: [patch] MeetMe option to place callers in MOH instead of Hangup when marked user (temporarily) leaves
Reporter:	Tony Mountifield (softins)	Labels:
Date Opened:	2004-08-27 10:29:42	Date Closed:	2004-10-01 23:15:43
Priority:	Major	Regression?	No
Status:	Closed/Complete	Components:	Applications/app_meetme
Versions:		Frequency of Occurrence
Related Issues:
Environment:		Attachments:	( 0) app_meetme-diff.txt ( 1) pls-hold.gsm ( 2) sounds-diff.txt
Description:	For supervised conferences with a marked user, it may be necessary to allow the supervisor to leave the conference temporarily without hanging up the users, but without allowing them to continue the discussion in his/her absence. This patch builds on the recent wxA options by adding a W option that is like w but puts the callers into MOH if all the marked users leave. It plays "please hold" before starting the MOH. Note that the x flag will override W. Three attached files are: patch for app_meetme.c, trivial patch for sounds.txt and sound file pls-hold.gsm. The pls-hold.gsm file was taken from busy-pls-hold.gsm using "sox ... trim 2.0 1.0".
Comments:	By: Tony Mountifield (softins) 2004-08-27 10:33:37 Disclaimer on file (sorry, forgot to say). By: Mark Spencer (markster) 2004-08-27 15:05:26 This patch seems to affect the normal processing of just the MOH flag? By: Tony Mountifield (softins) 2004-08-27 15:51:53 It was necessary to combine the two tests so there was only one place that tested for MOH being turned on or off. My first attempt with separate if-blocks had it being turned on and off every time round the loop! There is a further test to only stream the pls-hold file if it is due to the W flag. This preserves existing M flag functionality. Happy to revise it if there's a better way to do it. By: mochouinard (mochouinard) 2004-08-30 15:39:18 I was gonna submit something similar following bug 1562 in a few days. I did alot of patch into bug 1562 but finally mark and I decided to make alot of small patch. Basicly, we shouldn't have to add a flag for it, since it should be default if the x flag aint set and w is set. I'll try to check it out tonight By: Tony Mountifield (softins) 2004-08-30 17:11:48 Yes, that is actually a very good point. I agree the extra flag is unnecessary. There are other ways I'd like to revise it too, such as using the MOH code within the loop for the initial wait too, rather than a separate loop beforehand. Currently, the MeetmeJoin events for everyone waiting are delayed until the leader joins. I think the initial waiting should be within MeetmeJoin/MeetmeLeave. Haven't yet had chance to work through the logic fully. By: Tony Mountifield (softins) 2004-08-31 03:48:14 On second thoughts, there may be people who would only want MOH while waiting for the marked user to arrive, but want to allow the conference to continue normally even if the marked user(s) subsequently leave. Others may want MOH when all the marked users leave. That was my original thinking behind having both w and W flags. By: mochouinard (mochouinard) 2004-09-05 17:16:05 Please check out bug 2382.  Like I said previously, this was already within 1562 but was decided to split 1562 in multiple patch for easier increment. Let me know if 2382 is allright with you or you ratter do it some other way. By: Tony Mountifield (softins) 2004-09-10 17:11:21 Sorry for the delay - work here has been manic! I checked out your bug ASTERISK-2350 (btw, if you put a # before the bug number, Mantis makes a link to it). It looks more comprehensive than mine, and I particularly like the fact that it has got rid of the MOH loop before entering the conference. I haven't tried your patch out yet, and probably won't get time for a couple of weeks. So if the bug marshalls are happy that it addresses the same issue as this bug, I'm happy for this one to be closed out in deference to yours. By: Mark Spencer (markster) 2004-10-01 23:12:17 Yah looks like ASTERISK-2350 is a bit more complete, but see if you can't help get the prompts made!