asterisk/asterisk.git
11 years agoremove superfluous reference counting operations in manage_parkinglot since ao2_inter...
Jeff Peeler [Fri, 3 Oct 2008 22:40:59 +0000 (22:40 +0000)]
remove superfluous reference counting operations in manage_parkinglot since ao2_interator_next increments the ref count automatically

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@146242 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoResolve a subtle bug where we would never successfully be able to get
Sean Bright [Fri, 3 Oct 2008 22:10:18 +0000 (22:10 +0000)]
Resolve a subtle bug where we would never successfully be able to get
the first item in the CLI entry list.  This was preventing '!' from
showing up in either 'help' or in tab completion.

(closes issue #13578)
Reported by: mvanbaak

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@146198 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 146129 via svnmerge
Jeff Peeler [Fri, 3 Oct 2008 20:49:17 +0000 (20:49 +0000)]
Blocked revisions 146129 via svnmerge

........
r146129 | jpeeler | 2008-10-03 15:44:22 -0500 (Fri, 03 Oct 2008) | 7 lines

(closes issue #13425)
Reported by: mdu113
Tested by: mdu113

Similar to r143204, masquerade the channel in the case of Park being called from AGI.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@146130 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agodocument meetme schedule changes (related to issue #11040)
Tilghman Lesher [Fri, 3 Oct 2008 18:30:39 +0000 (18:30 +0000)]
document meetme schedule changes (related to issue #11040)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@146081 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoput a note in CHANGES about the cli_cleanup done during AstriDevCon
Michiel van Baak [Fri, 3 Oct 2008 17:36:30 +0000 (17:36 +0000)]
put a note in CHANGES about the cli_cleanup done during AstriDevCon

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@146053 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoThe dialing API should inherit datastores as well as variables
Terry Wilson [Fri, 3 Oct 2008 17:35:37 +0000 (17:35 +0000)]
The dialing API should inherit datastores as well as variables

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@146052 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoThe 'P' command for ExternalIVR was also added in 1.6.0
Russell Bryant [Thu, 2 Oct 2008 19:30:45 +0000 (19:30 +0000)]
The 'P' command for ExternalIVR was also added in 1.6.0

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145962 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoTCP support for ExternalIVR went in to 1.6.1, not 1.6.0
Russell Bryant [Thu, 2 Oct 2008 19:27:37 +0000 (19:27 +0000)]
TCP support for ExternalIVR went in to 1.6.1, not 1.6.0

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145959 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agofix the 'meetme list', 'meetme list concise', 'meetme list $confno' and 'meetme list...
Michiel van Baak [Thu, 2 Oct 2008 18:02:26 +0000 (18:02 +0000)]
fix the 'meetme list', 'meetme list concise', 'meetme list $confno' and 'meetme list $confno concise' CLI commands

(closes issue #13586)
Reported by: john8675309
Help and feedback from eliel, thanks!

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145915 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoPermit the syntax and synopsis fields to be set (for func_odbc).
Tilghman Lesher [Thu, 2 Oct 2008 17:16:54 +0000 (17:16 +0000)]
Permit the syntax and synopsis fields to be set (for func_odbc).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145846 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agomake this compile under devmode again
Michiel van Baak [Thu, 2 Oct 2008 16:42:30 +0000 (16:42 +0000)]
make this compile under devmode again

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145842 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 145839 via svnmerge
Tilghman Lesher [Thu, 2 Oct 2008 16:40:32 +0000 (16:40 +0000)]
Blocked revisions 145839 via svnmerge

........
  r145839 | tilghman | 2008-10-02 11:39:56 -0500 (Thu, 02 Oct 2008) | 4 lines

  Backport support for some of the keyword modifications used in 1.6 (while warning that
  some options aren't really supported) and add some warning messages.  Some credit to
  oej, who was complaining in #asterisk-dev.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145840 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoThis is much cleaner, methinks.
Sean Bright [Thu, 2 Oct 2008 15:28:48 +0000 (15:28 +0000)]
This is much cleaner, methinks.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145771 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 145751 via svnmerge from
Tilghman Lesher [Thu, 2 Oct 2008 15:17:16 +0000 (15:17 +0000)]
Merged revisions 145751 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r145751 | tilghman | 2008-10-02 10:13:21 -0500 (Thu, 02 Oct 2008) | 3 lines

  Some sanity checks that may have led to prior crashes, found by codefreeze-lap
  (murf) on IRC.  Also some cleanup of incorrectly-used constants.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145752 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoTry a test compile using the GMime library. Some distros install gmime-config
Sean Bright [Wed, 1 Oct 2008 23:48:16 +0000 (23:48 +0000)]
Try a test compile using the GMime library.  Some distros install gmime-config
in the base package instead of the -devel package.  Now we print a notice and
disable GMime support instead of bombing during the main compilation.

(closes issue #13583)
Reported by: arkadia

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145692 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoAdd schedule extensions to app_meetme. In addition, the reporter found a
Tilghman Lesher [Wed, 1 Oct 2008 23:02:25 +0000 (23:02 +0000)]
Add schedule extensions to app_meetme.  In addition, the reporter found a
problem within strptime(3), which we are correcting here with ast_strptime().
(closes issue #11040)
 Reported by: DEA
 Patches:
       20080910__bug11040.diff.txt uploaded by Corydon76 (license 14)
 Tested by: DEA

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145649 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoOkay, this should really do it now. While I did manage
Mark Michelson [Wed, 1 Oct 2008 22:23:50 +0000 (22:23 +0000)]
Okay, this should really do it now. While I did manage
to fix blind transfers with my last commit here, I also
caused an unwanted side-effect. That is, only the first
priority of the 'h' extension would be executed when
a blind transfer occurred instead of all priorities.

Essentially, my last commit corrected the return value
of ast_bridge_call. However, the implementation still
was not 100% correct. Now it is.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145606 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoif (!(x) == 0) is the same as
Mark Michelson [Wed, 1 Oct 2008 21:33:11 +0000 (21:33 +0000)]
if (!(x) == 0) is the same as
if (x).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145579 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoThe logic surrounding the return value of ast_spawn_extension
Mark Michelson [Wed, 1 Oct 2008 21:06:26 +0000 (21:06 +0000)]
The logic surrounding the return value of ast_spawn_extension
within ast_bridge_call was reversed.

This problem was observed when a blind transfer placed from
the callee channel of a test call failed.

While the problem I am solving here is exactly the same
as what was reported in issue #13584, the difference is
that this fix I am applying is trunk-only. Issue #13584
was reported against the 1.4 branch, and my tests
of 1.4's blind transfers appear to work fine.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145553 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 145479 via svnmerge from
Leif Madsen [Wed, 1 Oct 2008 17:26:20 +0000 (17:26 +0000)]
Merged revisions 145479 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r145479 | lmadsen | 2008-10-01 13:18:30 -0400 (Wed, 01 Oct 2008) | 6 lines

  Update the realtime_pgsql.sql script to create the setinterfacevar column.

  (closes issue #13549)
  Reported by: fiddur
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145487 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoInitializing buffer prevents a segfault when arguments are incomplete.
Tilghman Lesher [Wed, 1 Oct 2008 15:44:06 +0000 (15:44 +0000)]
Initializing buffer prevents a segfault when arguments are incomplete.
(closes issue #13471)
 Reported by: alecdavis
 Patches:
       20080916__bug13471.diff.txt uploaded by Corydon76 (license 14)
 Tested by: alecdavis

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145428 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoToo many times have I mistyped the word
Mark Michelson [Wed, 1 Oct 2008 14:44:12 +0000 (14:44 +0000)]
Too many times have I mistyped the word
'install' as 'isntall'

Now this typo shall no longer be a problem since
'make isntall' just builds the 'install' target.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145381 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agotabs to spaces
Russell Bryant [Wed, 1 Oct 2008 12:29:18 +0000 (12:29 +0000)]
tabs to spaces

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145329 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years ago(closes issue #13337)
Jeff Peeler [Tue, 30 Sep 2008 22:21:19 +0000 (22:21 +0000)]
(closes issue #13337)
Reported by: pj
Tested by: pj

Set transport to SIP_TRANSPORT_UDP mode if not specified which fixes calls to get_transport returning UNKNOWN.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145249 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoAdd support for call pickup on Snom phones. Asterisk now includes a magic
Russell Bryant [Tue, 30 Sep 2008 21:32:53 +0000 (21:32 +0000)]
Add support for call pickup on Snom phones.  Asterisk now includes a magic
call-id in the dialog-info event package used with extension state subscriptions
on Snom phones.  Then, when the phone sends an INVITE with Replaces for the
special callid, Asterisk will perform a pickup on the extension that was
subscribed to.

The original code on this issue was submitted by xylome.  However, contributions
have been made by (at least) mgernoth and pkempgen.  The final patch was written
by seanbright, and includes the necessary logic to allow this work in a
technology independent way.

(closes issue #5014)
Reported by: xylome
Patches:
      issue5014-trunk.diff uploaded by seanbright (license 71)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145226 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years ago* Miscellaneous formatting changes to make v1.4 and trunk
Richard Mudgett [Tue, 30 Sep 2008 21:00:54 +0000 (21:00 +0000)]
*  Miscellaneous formatting changes to make v1.4 and trunk
more merge compatible in the mISDN area.

channels/chan_misdn.c
*  Eliminated redundant code in cb_events() EVENT_SETUP

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145200 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerge the cli_cleanup branch.
Michiel van Baak [Sun, 28 Sep 2008 23:32:14 +0000 (23:32 +0000)]
Merge the cli_cleanup branch.

This work is done by lmadsen, junky and mvanbaak
during AstriDevCon.

This is the second audit the CLI got, and
this time lmadsen made sure he had _ALL_ modules
loaded that have CLI commands in them.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145121 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoChange several improper "sizeof" to "strlen", as sizeof in that context would
Tilghman Lesher [Sun, 28 Sep 2008 21:39:07 +0000 (21:39 +0000)]
Change several improper "sizeof" to "strlen", as sizeof in that context would
incorrectly use the size of a pointer, rather than the length of a string.
(Closes issue #13574)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145076 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agorename chandup() and clarify its usage
Kevin P. Fleming [Sun, 28 Sep 2008 17:08:12 +0000 (17:08 +0000)]
rename chandup() and clarify its usage

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@145027 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoremove incorrect comment
Kevin P. Fleming [Sat, 27 Sep 2008 16:17:43 +0000 (16:17 +0000)]
remove incorrect comment

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144951 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agofix bugs caused by r144949 when MALLOC_DEBUG is defined
Kevin P. Fleming [Sat, 27 Sep 2008 16:10:33 +0000 (16:10 +0000)]
fix bugs caused by r144949 when MALLOC_DEBUG is defined

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144950 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 144924-144925 via svnmerge from
Kevin P. Fleming [Sat, 27 Sep 2008 15:52:56 +0000 (15:52 +0000)]
Merged revisions 144924-144925 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r144924 | kpfleming | 2008-09-27 10:00:48 -0500 (Sat, 27 Sep 2008) | 6 lines

  improve header inclusion process in a few small ways:

    - it is no longer necessary to forcibly include asterisk/autoconfig.h; every module already includes asterisk.h as its first header (even before system headers), which serves the same purpose
    - astmm.h is now included by asterisk.h when needed, instead of being forced by the Makefile; this means external modules will build properly against installed headers with MALLOC_DEBUG enabled
    - simplify the usage of some of these headers in the AEL-related stuff in the utils directory
........
  r144925 | kpfleming | 2008-09-27 10:13:30 -0500 (Sat, 27 Sep 2008) | 2 lines

  fix some minor issues with rev 144924
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144949 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agofix a couple of CLI commands that did not have a help description.
Michiel van Baak [Sat, 27 Sep 2008 00:49:24 +0000 (00:49 +0000)]
fix a couple of CLI commands that did not have a help description.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144879 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoUpdate documentation to include default setting. This is for you jtodd!
Joshua Colp [Fri, 26 Sep 2008 23:12:13 +0000 (23:12 +0000)]
Update documentation to include default setting. This is for you jtodd!

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144829 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 144758 via svnmerge
Mark Michelson [Fri, 26 Sep 2008 22:15:40 +0000 (22:15 +0000)]
Blocked revisions 144758 via svnmerge

........
r144758 | mmichelson | 2008-09-26 17:14:59 -0500 (Fri, 26 Sep 2008) | 7 lines

This patch was applied to 1.4 but it completely
does not apply since the "found" pointer is not
passed in to this function. If this is going to
be backported, it needs to be done differently or
a deeper backport needs to be done.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144759 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years ago(closes issue #13564)
Steve Murphy [Fri, 26 Sep 2008 18:02:06 +0000 (18:02 +0000)]
(closes issue #13564)
Reported by: mnicholson
Patches:
      pbx_lua9.diff uploaded by mnicholson (license 96)

Many thanks to Matt for his upgrade to the lua dialplan
option!

the Description from the bug:

This patch adds a stack trace to errors encountered while executing lua extensions. The patch also handles out of memory errors reported by lua.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144681 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 144677 via svnmerge from
Steve Murphy [Fri, 26 Sep 2008 17:50:35 +0000 (17:50 +0000)]
Merged revisions 144677 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r144677 | murf | 2008-09-26 11:47:13 -0600 (Fri, 26 Sep 2008) | 12 lines

(closes issue #13563)
Reported by: mnicholson
Patches:
      found1.diff uploaded by mnicholson (license 96)

This patch was mainly meant to apply to trunk and 1.6.x,
but I'm applying it to 1.4 also, which should be a perfectly
harmless fix to the vast majority of users who are not using
external switches, but the few who might be affected
will not have to go to the pain of filing a bug report.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144678 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMatt suggests we remove utils/build-extensions-conf.lua, as per bug 12961, it is...
Steve Murphy [Fri, 26 Sep 2008 16:51:30 +0000 (16:51 +0000)]
Matt suggests we remove utils/build-extensions-conf.lua, as per bug 12961, it is no longer necessary.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144635 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years ago(closes issue #13557)
Steve Murphy [Thu, 25 Sep 2008 22:21:28 +0000 (22:21 +0000)]
(closes issue #13557)
Reported by: nickpeirson

The user attached a patch, but the license is not yet
recorded. I took the liberty of finding and replacing
ALL index() calls with strchr() calls, and that
involves more than just main/pbx.c;

chan_oss, app_playback, func_cut also had calls
to index(), and I changed them out. 1.4 had no
references to index() at all.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144569 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years ago(closes issue #13559)
Steve Murphy [Thu, 25 Sep 2008 21:54:11 +0000 (21:54 +0000)]
(closes issue #13559)
Reported by: mnicholson
Patches:
      pbx_lua8.diff uploaded by mnicholson (license 96)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144563 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoI added a little verbage to hashtab for the hashtab_destroy func.
Steve Murphy [Thu, 25 Sep 2008 21:18:12 +0000 (21:18 +0000)]
I added a little verbage to hashtab for the hashtab_destroy func.
It was pretty sparsely documented.

This update fleshes out the pbx_lua module, to
add the switch statements to the extensions in the
extensions.lua file, as well as removing them when
the module is unloaded.

Many thanks to Matt Nicholson for his fine
contribution!

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144523 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years ago(closes issue #13558)
Steve Murphy [Thu, 25 Sep 2008 17:51:11 +0000 (17:51 +0000)]
(closes issue #13558)
Reported by: mnicholson

Considering that the example extensions.lua used nothing but ["12345"] notation,
and that the resulting error message:

[Sep 24 17:01:16] ERROR[12393]: pbx_lua.c:1204 exec: Error executing lua extension: attempt to call a nil value

is not very informative as to the nature of the problem, I think this bug
fix is a big win!

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144482 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 144420 via svnmerge
Steve Murphy [Thu, 25 Sep 2008 16:21:54 +0000 (16:21 +0000)]
Blocked revisions 144420 via svnmerge

I am pushing this up to trunk and beyond, because I am not
certain that this bug will happen in the same way in
trunk and 1.6.x. If someone can document this problem
in 1.6.x or trunk, I can revert this block and push the
change up.

........
r144420 | murf | 2008-09-25 10:12:14 -0600 (Thu, 25 Sep 2008) | 25 lines

(closes issue #12101)
Reported by: MVF
Tested by: neutrino88, urzedo, murf, thiagofernandes

Many thanks to neutrino88 for this patch, which
solves a problem whereby channels get a CANCEL
request, respond to it properly, but end up
in a hung state, infinitely being rescheduled.
This fix is a bit crude, in that catches the
problem at a rather late phase, but it may
prevent infinite rescheduling problems that
might still arise.

It might have been better to find out why,
in the course of protocol handling, the channel
was not destroyed, but we leave that to
future generations.

Many thanks to urzedo and thiagofernandes for
their work in verifying that the patch code
indeed is being executing, and averting the
problem.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144421 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoRecorded merge of revisions 144356 via svnmerge from
Tilghman Lesher [Thu, 25 Sep 2008 01:46:13 +0000 (01:46 +0000)]
Recorded merge of revisions 144356 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r144356 | tilghman | 2008-09-24 20:44:47 -0500 (Wed, 24 Sep 2008) | 6 lines

  Backport Hebrew language to voicemail.
  (closes issue #13155)
   Reported by: greenfieldtech
   Patches:
         voicemail-hebrew-patch-1.4-SVN.c.patch uploaded by greenfieldtech (license 369)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144357 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlanch the 404 error message for those with no sense of humor
Doug Bailey [Wed, 24 Sep 2008 22:05:44 +0000 (22:05 +0000)]
Blanch the 404 error message for those with no sense of humor

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144314 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 144238 via svnmerge from
Christian Richter [Wed, 24 Sep 2008 08:42:55 +0000 (08:42 +0000)]
Merged revisions 144238 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r144238 | crichter | 2008-09-24 10:20:52 +0200 (Mi, 24 Sep 2008) | 1 line

improved helptext of misdn_set_opt.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144257 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoCreate a 'hashcompat' option that permits the results of a CURL() able to be
Tilghman Lesher [Wed, 24 Sep 2008 06:43:05 +0000 (06:43 +0000)]
Create a 'hashcompat' option that permits the results of a CURL() able to be
passed directly into the HASH() function.  Requested via the -users list, and
committed at Astricon in the Code Zone.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144199 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoFix a conflict in flag values
Mark Michelson [Tue, 23 Sep 2008 23:33:33 +0000 (23:33 +0000)]
Fix a conflict in flag values

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144149 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 144066 via svnmerge from
Steve Murphy [Tue, 23 Sep 2008 16:52:32 +0000 (16:52 +0000)]
Merged revisions 144066 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r144066 | murf | 2008-09-23 10:41:49 -0600 (Tue, 23 Sep 2008) | 29 lines

(closes issue #13489)
Reported by: DougUDI
Tested by: murf

(closes issue #13490)
Reported by: seanbright
Tested by: murf

(closes issue #13467)
Reported by: edantie
Tested by: murf, edantie, DougUDI

This crash happens because we are unsafely handling old pointers.
The channel whose cdr is being handled, has been hung up and
destroyed already. I reorganized the code a bit, and tried not
to lose the fork-cdr-chain concepts of the previous code.
I now verify that the 'previous' channel (the channel we
had when the bridge was started), still exists, by looking it up
by name in the channel list. I also do not try to reset the
CDR's of channels involved in bridges.

Testing shows it solves the crash problem, and should not
negatively impact previous fixes involving CDR's generated
during/after blind transfers. (The reason we need to reset
the CDR's on the "beginning" channels in the first place).

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144067 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoWhen a promiscuous redirect contained both a user and
Mark Michelson [Tue, 23 Sep 2008 15:37:00 +0000 (15:37 +0000)]
When a promiscuous redirect contained both a user and
host portion in the Contact URI and specifies a
transport, the parsing done in parse_moved_contact
resulted in a malformed URI.

This commit fixes the parsing so that a proper
Dial string may be formed when the forwarded
call is placed.

(closes issue #13523)
Reported by: mattdarnell
Patches:
      13523v2.patch uploaded by putnopvut (license 60)
Tested by: mattdarnell

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@144025 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 143964 via svnmerge
Steve Murphy [Tue, 23 Sep 2008 14:26:32 +0000 (14:26 +0000)]
Blocked revisions 143964 via svnmerge

........
r143964 | murf | 2008-09-23 08:22:10 -0600 (Tue, 23 Sep 2008) | 21 lines

In at least one machine, we noted that the timestr
was not getting set in the STMT; it was coming out,
usually, as binary garbage to an mssql server.
These changes fixed the problem. The only thing
I can venture forth as a guess, is that the pointer
is being stored in the interface, not a copy of the
string. Because we ripped the build process into a
subroutine, the timestr became a temp. stack variable,
and between the time the STMT got built and the
time it was executed on the server, the string being
pointed to was damaged. At any rate, even if this
theory is false, and some mechanism was at fault,
this fix worked reliably where it didn't before.

Why this bug didn't bite last week, I have no idea.
This change basically defines the timestr buffer
in the calling function, extending the life of the
buffer to cover both the STMT's building and
processing to the server.

I AM BLOCKING THIS FIX FROM TRUNK, because we
tested out a trunk release, and it is working
fine.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143974 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 143903 via svnmerge from
Sean Bright [Mon, 22 Sep 2008 22:50:07 +0000 (22:50 +0000)]
Merged revisions 143903 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r143903 | seanbright | 2008-09-22 18:49:00 -0400 (Mon, 22 Sep 2008) | 8 lines

Use the advertised header size in .au files instead of just assuming they
are 24 bytes (the minimum).

(closes issue #13450)
Reported by: jamessan
Patches:
      pcm-header.diff uploaded by jamessan (license 246)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143904 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agofix privacymanager example so it shows how to use the
Michiel van Baak [Sun, 21 Sep 2008 09:53:01 +0000 (09:53 +0000)]
fix privacymanager example so it shows how to use the
PRIVACYMRGSTATUS variable

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143843 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agodocument the new context argument for privacymanager
Michiel van Baak [Sun, 21 Sep 2008 09:31:54 +0000 (09:31 +0000)]
document the new context argument for privacymanager
so people can do pattern matching on the input

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143840 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agofix privacy documentation. We no longer do priority jumping +101
Michiel van Baak [Sun, 21 Sep 2008 09:27:08 +0000 (09:27 +0000)]
fix privacy documentation. We no longer do priority jumping +101

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143837 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agomake 'module unload chan_skinny.so' actually work.
Michiel van Baak [Sat, 20 Sep 2008 10:39:20 +0000 (10:39 +0000)]
make 'module unload chan_skinny.so' actually work.

(closes issue #13524)
Reported by: wedhorn
Patches:
      unload.diff uploaded by wedhorn (license 30)
  With small tweak by me to prevent a crash

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143799 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 143736 via svnmerge from
Sean Bright [Sat, 20 Sep 2008 00:52:20 +0000 (00:52 +0000)]
Merged revisions 143736 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r143736 | seanbright | 2008-09-19 20:50:10 -0400 (Fri, 19 Sep 2008) | 9 lines

Make vmail.cgi work with mailboxes defined in users.conf, too.

(closes issue #13187)
Reported by: netvoice
Patches:
      20080911__bug13187.diff.txt uploaded by Corydon76 (license 14)
      (Slightly modified to take alchamist's comments on mantis into account)
Tested by: msales, alchamist, seanbright

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143737 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoThis blocks 143674 from trunk; it appears to already done in trunk, since ast_odbc_di...
Steve Murphy [Fri, 19 Sep 2008 21:41:39 +0000 (21:41 +0000)]
This blocks 143674 from trunk; it appears to already done in trunk, since ast_odbc_direct_execute creates a new stmt for each attempt.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143697 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoWe should only unsubscribe to the device state event
Mark Michelson [Fri, 19 Sep 2008 15:43:28 +0000 (15:43 +0000)]
We should only unsubscribe to the device state event
subscription if we have previously subscribed. Otherwise
a segfault will occur.

(closes issue #13476)
Reported by: jonnt
Patches:
      13476.patch uploaded by putnopvut (license 60)
Tested by: jonnt

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143609 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 143534 via svnmerge from
Steve Murphy [Thu, 18 Sep 2008 23:41:33 +0000 (23:41 +0000)]
Merged revisions 143534 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r143534 | murf | 2008-09-18 16:11:51 -0600 (Thu, 18 Sep 2008) | 1 line

A micro-fix, in sip_park_thread, where d is freed before the func is done using it.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143559 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 143475 via svnmerge
Jeff Peeler [Thu, 18 Sep 2008 21:02:02 +0000 (21:02 +0000)]
Blocked revisions 143475 via svnmerge

........
r143475 | jpeeler | 2008-09-18 16:01:23 -0500 (Thu, 18 Sep 2008) | 1 line

compare peer and chan instead of looking at the parked user (pu), which could have possibly already have been freed by the parking thread
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143476 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 143404 via svnmerge from
Tilghman Lesher [Wed, 17 Sep 2008 20:57:58 +0000 (20:57 +0000)]
Merged revisions 143404 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r143404 | tilghman | 2008-09-17 15:55:47 -0500 (Wed, 17 Sep 2008) | 6 lines

  When callerid is blank, we want to use "unknown caller" in those cases, too.
  (closes issue #13486)
   Reported by: tomo1657
   Patches:
         20080917__bug13486.diff.txt uploaded by Corydon76 (license 14)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143405 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoIf attempting to free a NULL pointer when MALLOC_DEBUG
Mark Michelson [Wed, 17 Sep 2008 20:25:40 +0000 (20:25 +0000)]
If attempting to free a NULL pointer when MALLOC_DEBUG
is set, don't bother searching for a region to free, just
immediately exit.

This has the dual benefit of suppressing a warning message
about freeing memory at (nil) and of optimizing the free()
replacement by not having to do any futile searching for
the proper region to free.

(closes issue #13498)
Reported by: pj
Patches:
      13498.patch uploaded by putnopvut (license 60)
Tested by: pj

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143400 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 143337 via svnmerge from
Mark Michelson [Wed, 17 Sep 2008 18:26:35 +0000 (18:26 +0000)]
Merged revisions 143337 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r143337 | mmichelson | 2008-09-17 13:24:15 -0500 (Wed, 17 Sep 2008) | 6 lines

Allow for "G.729" if offered in an SDP even though
it is not RFC 3551 compliant. Some Cisco switches
will send this in an SDP, and it doesn't hurt to
be able to accept this.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143340 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 143270 via svnmerge
Jeff Peeler [Tue, 16 Sep 2008 21:44:52 +0000 (21:44 +0000)]
Blocked revisions 143270 via svnmerge

........
r143270 | jpeeler | 2008-09-16 16:44:21 -0500 (Tue, 16 Sep 2008) | 9 lines

(closes issue #13493)
Reported by: mdu113
Patches:
      bug13493.diff uploaded by jpeeler (license 325)
Tested by: mdu113

Changed park_call_full to hold the parkinglot lock a little longer, which protects the parkeduser struct from being freed out from underneath. Made sure that the parking extension is added to the parking context while holding the lock thereby ensuring that there are no spurious warnings from removal attempts when a hangup occurs while the parking lot is being announced.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143271 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBlocked revisions 143204 via svnmerge
Jeff Peeler [Tue, 16 Sep 2008 16:35:32 +0000 (16:35 +0000)]
Blocked revisions 143204 via svnmerge

........
r143204 | jpeeler | 2008-09-16 11:14:25 -0500 (Tue, 16 Sep 2008) | 10 lines

(closes issue #13465)
(closes issue #13425)
Reported by: mdu113
Patches:
      bug13425.diff uploaded by jpeeler (license 325)
Tested by: mdu113

The main change here was to masquerade the channel if the channel that was to be parked was running a PBX on it. The PBX thread can then maintain full control of the channel (the zombie) as it expects to while allowing the parking thread full control of the real (parked) channel.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143226 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 143140 via svnmerge from
Tilghman Lesher [Mon, 15 Sep 2008 21:31:36 +0000 (21:31 +0000)]
Merged revisions 143140 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r143140 | tilghman | 2008-09-15 16:29:32 -0500 (Mon, 15 Sep 2008) | 6 lines

  Set the raw formats at the same time as the other formats.
  (closes issue #13240)
   Reported by: jvandal
   Patches:
         20080813__bug13240.diff.txt uploaded by Corydon76 (license 14)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143141 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoplug a couple of memleaks in chan_skinny.
Michiel van Baak [Sun, 14 Sep 2008 22:16:34 +0000 (22:16 +0000)]
plug a couple of memleaks in chan_skinny.

(closes issue #13452)
Reported by: pj
Patches:
      memleak5.diff uploaded by wedhorn (license 30)
Tested by: wedhorn, pj, mvanbaak

(closes issue #13294)
Reported by: pj

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143082 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoEverytime a compile fails, a puppy dies.
Sean Bright [Sat, 13 Sep 2008 14:15:08 +0000 (14:15 +0000)]
Everytime a compile fails, a puppy dies.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143034 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoRepair IAXVAR implementation so that it works again (regression?)
Tilghman Lesher [Sat, 13 Sep 2008 13:54:15 +0000 (13:54 +0000)]
Repair IAXVAR implementation so that it works again (regression?)
(closes issue #13354)
 Reported by: adomjan
 Patches:
       20080828__bug13354.diff.txt uploaded by Corydon76 (license 14)
       20080829__bug13354__1.6.0.diff.txt uploaded by Corydon76 (license 14)
 Tested by: Corydon76, adomjan

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@143031 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoCreate a new config file status, CONFIG_STATUS_FILEINVALID for differentiating
Tilghman Lesher [Fri, 12 Sep 2008 23:30:03 +0000 (23:30 +0000)]
Create a new config file status, CONFIG_STATUS_FILEINVALID for differentiating
when a file is invalid from when a file is missing.  This is most important when
we have two configuration files.  Consider the following example:

Old system:
sip.conf     users.conf     Old result               New result
========     ==========     ==========               ==========
Missing      Missing        SIP doesn't load         SIP doesn't load
Missing      OK             SIP doesn't load         SIP doesn't load
Missing      Invalid        SIP doesn't load         SIP doesn't load
OK           Missing        SIP loads                SIP loads
OK           OK             SIP loads                SIP loads
OK           Invalid        SIP loads incompletely   SIP doesn't load
Invalid      Missing        SIP doesn't load         SIP doesn't load
Invalid      OK             SIP doesn't load         SIP doesn't load
Invalid      Invalid        SIP doesn't load         SIP doesn't load

So in the case when users.conf doesn't load because there's a typo that
disrupts the syntax, we may only partially load users, instead of failing with
an error, which may cause some calls not to get processed.  Worse yet, the old
system would do this with no indication that anything was even wrong.

(closes issue #10690)
 Reported by: dtyoo
 Patches:
       20080716__bug10690.diff.txt uploaded by Corydon76 (license 14)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142992 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142927 via svnmerge from
Jeff Peeler [Fri, 12 Sep 2008 22:24:13 +0000 (22:24 +0000)]
Merged revisions 142927 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142927 | jpeeler | 2008-09-12 17:22:28 -0500 (Fri, 12 Sep 2008) | 6 lines

(closes issue #12965)
Reported by: rlsutton2

Prevents local channels from playing MOH at each other which was causing ast_generic_bridge to loop much faster.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142929 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142865 via svnmerge from
Tilghman Lesher [Fri, 12 Sep 2008 20:49:46 +0000 (20:49 +0000)]
Merged revisions 142865 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r142865 | tilghman | 2008-09-12 15:37:18 -0500 (Fri, 12 Sep 2008) | 11 lines

  Create rules for disallowing contacts at certain addresses, which may
  improve the security of various installations.  As this does not change
  any default behavior, it is not classified as a direct security fix for
  anything within Asterisk, but may help PBX admins better secure their
  SIP servers.
  (closes issue #11776)
   Reported by: ibc
   Patches:
         20080829__bug11776.diff.txt uploaded by Corydon76 (license 14)
   Tested by: Corydon76, blitzrage
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142866 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoRecorded merge of revisions 142807 via svnmerge from
Michiel van Baak [Fri, 12 Sep 2008 18:22:17 +0000 (18:22 +0000)]
Recorded merge of revisions 142807 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r142807 | mvanbaak | 2008-09-12 19:59:25 +0200 (Fri, 12 Sep 2008) | 2 lines

  fix copyright year range
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142808 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoWhen checking for an encoded character, make sure the string isn't blank, first.
Tilghman Lesher [Fri, 12 Sep 2008 16:54:44 +0000 (16:54 +0000)]
When checking for an encoded character, make sure the string isn't blank, first.
(Closes issue #13470)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142748 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142744 via svnmerge from
Tilghman Lesher [Fri, 12 Sep 2008 16:38:55 +0000 (16:38 +0000)]
Merged revisions 142744 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142744 | tilghman | 2008-09-12 11:38:02 -0500 (Fri, 12 Sep 2008) | 4 lines

Missing merge from 1.2 fixes errant exit on DTMF, only when language is Italian
(cf commit 34242)
(Closes issue #7353)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142745 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142740 via svnmerge from
Tilghman Lesher [Fri, 12 Sep 2008 16:29:01 +0000 (16:29 +0000)]
Merged revisions 142740 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142740 | tilghman | 2008-09-12 11:27:32 -0500 (Fri, 12 Sep 2008) | 4 lines

Don't return a free'd pointer, when a file cannot be opened.
(closes issue #13462)
 Reported by: wackysalut

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142741 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142675 via svnmerge from
Steve Murphy [Fri, 12 Sep 2008 04:50:48 +0000 (04:50 +0000)]
Merged revisions 142675 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142675 | murf | 2008-09-11 22:29:34 -0600 (Thu, 11 Sep 2008) | 29 lines

Tested by: sergee, murf, chris-mac, andrew, KNK

This is a "second attempt" to restore the previous "endbeforeh" behavior
in 1.4 and up. In order to capture information concerning all the
legs of transfers in all their infinite combinations, I was forced
to this particular solution by a chain of logical necessities, the
first being that I was not allowed to rewrite the CDR mechanism from
the ground up!

This change basically leaves the original machinery alone, which allows
IVR and local channel type situations to generate CDR's as normal, but
a channel flag can be set to suppress the normal running of the h exten.
That flag would be set by the code that runs the h exten from the
ast_bridge_call routine, to prevent the h exten from being run twice.
Also, a flag in the ast_bridge_config struct passed into ast_bridge_call
can be used to suppress the running of the h exten in that routine. This
would happen, for instance, if you use the 'g' option in the Dial app.

Running this routine 'early' allows not only the CDR() func to be used
in the h extension for reading CDR variables, but also allows them to
be modified before the CDR is posted to the backends.

While I dearly hope that this patch overcomes all problems, and
introduces no new problems, reality suggests that surely someone
will have problems. In this case, please re-open 13251 (or 13289),
and we'll see if we can't fix any remaining issues.

** trunk note: some code to suppress the h exten being run
from app_queue was added; for the 'continue' option available
only in trunk/1.6.x.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142676 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoBuild under dev-mode
Sean Bright [Fri, 12 Sep 2008 00:49:31 +0000 (00:49 +0000)]
Build under dev-mode

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142635 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142575 via svnmerge from
Steve Murphy [Thu, 11 Sep 2008 23:12:53 +0000 (23:12 +0000)]
Merged revisions 142575 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142575 | murf | 2008-09-11 16:55:49 -0600 (Thu, 11 Sep 2008) | 20 lines

(closes issue #13364)
Reported by: mdu113

Well, fundamentally, the problems revealed in 13364 are
because of the ForkCDR call that is done before the dial.
When the bridge is in place, it's dealing with the first
(and wrong) cdr in the list.

So, I wrote a little func to zip down to the first non-locked
cdr in the chain, and thru-out the ast_bridge_call, these
results are used instead of raw chan->cdr and peer->cdr pointers.
This shouldn't affect anyone who isn't forking cdrs before a
dial, and should correct the cdr's of those that do.

So, this change ends up correcting the dstchannel
and userfield; the disposition was fixed by a previous
patch, it was OK coming into this problem.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142576 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoAdd usegmtime, as per the recent -users list discussion, and also add my
Tilghman Lesher [Thu, 11 Sep 2008 21:45:07 +0000 (21:45 +0000)]
Add usegmtime, as per the recent -users list discussion, and also add my
explanation to the file, since that additional text helps people understand
the concept.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142536 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142474 via svnmerge from
Steve Murphy [Wed, 10 Sep 2008 22:11:27 +0000 (22:11 +0000)]
Merged revisions 142474 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142474 | murf | 2008-09-10 15:58:17 -0600 (Wed, 10 Sep 2008) | 30 lines

(closes issue #12318)
Reported by: krtorio

I made a small change to the code that handles local channel situations.
In that code, I copy the answer time from the peer cdr, to the bridge_cdr,
but I wasn't also copying the disposition from the peer cdr.

So, Now I copy the disposition, and I've tested against
these cases:

1. phone 1 never answers the phone; no cdr is generated at all.
   this should show up as a manager command failure or something.

2. phone 2 never answers. CDR is generated, says NO ANSWER

3. phone 2 is busy. CDR is generated, says BUSY

4. phone 2 answers: CDR is generated, times are correct; disposition
   is ANSWERED, which is correct. The start time is the time that
   the manager dialed the first phone. The answer time is the time
   the second phone picks up.

I purposely left the cid and src fields blank; since this call really
originates from the manager, there is no 'easy' data to put in these
fields. If you feel strongly that these fields should be filled in,
re-open this bug and I'll dig further.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142475 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142416 via svnmerge from
Sean Bright [Wed, 10 Sep 2008 19:09:03 +0000 (19:09 +0000)]
Merged revisions 142416 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142416 | seanbright | 2008-09-10 15:05:46 -0400 (Wed, 10 Sep 2008) | 9 lines

Fix detection of PWLIB and OpenH323 version when spacing in the headers
isn't consistent.

(closes issue #13426)
Reported by: bamby
Patches:
      detect_openh323.diff uploaded by bamby (license 430)
      (Modified by me to use sed instead of tr)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142417 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142358 via svnmerge from
Tilghman Lesher [Wed, 10 Sep 2008 16:55:31 +0000 (16:55 +0000)]
Merged revisions 142358 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142358 | tilghman | 2008-09-10 11:54:29 -0500 (Wed, 10 Sep 2008) | 2 lines

Publish new extra sounds version.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142359 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142354 via svnmerge from
Russell Bryant [Wed, 10 Sep 2008 16:41:55 +0000 (16:41 +0000)]
Merged revisions 142354 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142354 | russell | 2008-09-10 11:39:53 -0500 (Wed, 10 Sep 2008) | 7 lines

It is a normal situation that a task gets put in the scheduler that should run
as soon as possible.  Accept "0" as an acceptable time to run, and also treat
negative as "run now", and don't print a debug message about it.

(inspired by a message asking about the "request to schedule in the past"
 debug message on the -dev list)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142355 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMove last change to CHANGES up to the 1.6.2 section
Russell Bryant [Wed, 10 Sep 2008 15:57:50 +0000 (15:57 +0000)]
Move last change to CHANGES up to the 1.6.2 section

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142318 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoDisable autoprune by default.
Philippe Sultan [Tue, 9 Sep 2008 22:08:56 +0000 (22:08 +0000)]
Disable autoprune by default.
(closes issue #13411)
Reported by: caio1982
Patches:
      res_jabber_autoprune1.diff uploaded by caio1982 (license 22)
Tested by: caio1982

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142280 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142218 via svnmerge from
Mark Michelson [Tue, 9 Sep 2008 19:16:30 +0000 (19:16 +0000)]
Merged revisions 142218 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142218 | mmichelson | 2008-09-09 14:15:28 -0500 (Tue, 09 Sep 2008) | 14 lines

Make sure that the branch sent in CANCEL requests
matches the branch of the INVITE it is cancelling.

(closes issue #13381)
Reported by: atca_pres
Patches:
      13381v2.patch uploaded by putnopvut (license 60)
Tested by: atca_pres

(closes issue #13198)
Reported by: rickead2000
Tested by: rickead2000

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142219 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoCleaned up comment
Richard Mudgett [Tue, 9 Sep 2008 17:30:52 +0000 (17:30 +0000)]
Cleaned up comment

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142181 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoThis is the trunk version of the patch to close
Mark Michelson [Tue, 9 Sep 2008 17:15:29 +0000 (17:15 +0000)]
This is the trunk version of the patch to close
issue 12979. The difference between this and the
1.6.0 and 1.6.1 versions is that this is a much more
invasive change. With this, we completely get rid
of the interfaces list, along with all its helper
functions.

Let me take a moment to say that this change personally
excites me since it may mean huge steps forward regarding
proper lock order in app_queue without having to strew
seemingly unnecessary locks all over the place. It also
results in a huge reduction in lines of code and complexity.
Way to go Brett!

(closes issue #12979)
Reported by: sigxcpu
Patches:
      20080710_issue12979_queue_custom_state_interface_trunk_2.diff uploaded by bbryant (license 36)
Tested by: sigxcpu, putnopvut

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142146 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142079 via svnmerge from
Mark Michelson [Tue, 9 Sep 2008 16:20:41 +0000 (16:20 +0000)]
Merged revisions 142079 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142079 | mmichelson | 2008-09-09 11:19:17 -0500 (Tue, 09 Sep 2008) | 21 lines

When determining if codecs used by SIP peers allow
the media to be natively bridged, use the jointcapability
instead of the peercapability.

It seems that the intent of using the peercapability was to
expand the choice of codecs for the call to increase the
chances of being able to native bridge the channels. The
problem is that if a codec were settled on for the native
bridge and that wasn't a codec that was configured to be used
by Asterisk for that peer, then Asterisk would send a
REINVITE with no codecs in the SDP which is a bug no matter
how you slice it.

(closes issue #13076)
Reported by: ramonpeek
Patches:
      13076.patch uploaded by putnopvut (license 60)
Tested by: tbelder

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142080 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 142063 via svnmerge from
Russell Bryant [Tue, 9 Sep 2008 15:44:10 +0000 (15:44 +0000)]
Merged revisions 142063 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r142063 | russell | 2008-09-09 10:40:24 -0500 (Tue, 09 Sep 2008) | 5 lines

Ensure that the stored CDR reference is still valid after the bridge before
poking at it.  Also, keep the channel locked while messing with this CDR.

(fixes crashes reported in issue #13409)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142064 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMinor fix to doco
Bradley Latus [Tue, 9 Sep 2008 12:34:32 +0000 (12:34 +0000)]
Minor fix to doco

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@142000 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoUse ast_debug for debug messages. I was wondering why debug
Mark Michelson [Tue, 9 Sep 2008 12:32:38 +0000 (12:32 +0000)]
Use ast_debug for debug messages. I was wondering why debug
messages weren't showing up when I had set the debug level
high for just app_queue.c. It's because we were only checking
the global option_debug variable instead of using the awesome
macro which checks both the global and file-specific value

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141998 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoFix a memory leak in chan_oss
Mark Michelson [Tue, 9 Sep 2008 10:20:58 +0000 (10:20 +0000)]
Fix a memory leak in chan_oss

(closes issue #13311)
Reported by: eliel
Patches:
      chan_oss.c.patch uploaded by eliel (license 64)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141995 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoModify ast_answer() to not hold the channel lock while calling ast_safe_sleep()
Russell Bryant [Tue, 9 Sep 2008 01:47:56 +0000 (01:47 +0000)]
Modify ast_answer() to not hold the channel lock while calling ast_safe_sleep()
or when calling ast_waitfor().  These are inappropriate times to hold the channel
lock.  This is what has caused "could not get the channel lock" messages from
chan_sip and has likely caused a negative impact on performance results of SIP
in Asterisk 1.6.  Thanks to file for pointing out this section of code.

(closes issue #13287)
(closes issue #13115)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141949 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoOptimization: The only reason we should check member
Mark Michelson [Mon, 8 Sep 2008 23:00:06 +0000 (23:00 +0000)]
Optimization: The only reason we should check member
status is if the queue has a joinempty or a leavewhenempty
setting which could cause the caller to not join the queue
or exit the queue. Prior to this patch, we could potentially
traverse the entire queue's member list for no reason since
even if the members are currently not available in some way
we're going to let the caller join the queue anyway.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141906 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoUm, apparently I didn't actually finish merging before committing.
Mark Michelson [Mon, 8 Sep 2008 22:14:40 +0000 (22:14 +0000)]
Um, apparently I didn't actually finish merging before committing.
Bad bad bad

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141868 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 141809 via svnmerge from
Mark Michelson [Mon, 8 Sep 2008 21:18:49 +0000 (21:18 +0000)]
Merged revisions 141809 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r141809 | mmichelson | 2008-09-08 16:10:10 -0500 (Mon, 08 Sep 2008) | 14 lines

Fix pedantic mode of chan_sip to only check the
remote tag of an endpoint once a dialog has
been confirmed. Up until that point, it is possible
and legal for the far-end to send provisional
responses with a different To: tag each time. With
this patch applied, these provisional messages
will not cause a matching problem.

(closes issue #11536)
Reported by: ibc
Patches:
      11536v2.patch uploaded by putnopvut (license 60)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141810 65c4cc65-6c06-0410-ace0-fbb531ad65f3

11 years agoMerged revisions 141806 via svnmerge from
Russell Bryant [Mon, 8 Sep 2008 21:05:01 +0000 (21:05 +0000)]
Merged revisions 141806 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
r141806 | russell | 2008-09-08 16:02:36 -0500 (Mon, 08 Sep 2008) | 7 lines

When doing an async goto, detect if the channel is already in the middle of a
masquerade.  This can happen when chan_local is trying to optimize itself out.
If this happens, fail the async goto instead of bursting into flames.

(closes issue #13435)
Reported by: geoff2010

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@141807 65c4cc65-6c06-0410-ace0-fbb531ad65f3