asterisk/asterisk.git
9 years agoconstification, remove include
Russell Bryant [Fri, 26 Feb 2010 08:07:09 +0000 (08:07 +0000)]
constification, remove include

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@249011 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove unnecessary includes, formatting tweak
Russell Bryant [Fri, 26 Feb 2010 08:05:36 +0000 (08:05 +0000)]
Remove unnecessary includes, formatting tweak

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@249010 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoconstification and remove unnecessary include
Russell Bryant [Fri, 26 Feb 2010 08:04:07 +0000 (08:04 +0000)]
constification and remove unnecessary include

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@249009 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 248860 via svnmerge from
Jeff Peeler [Thu, 25 Feb 2010 23:09:54 +0000 (23:09 +0000)]
Merged revisions 248860 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r248860 | jpeeler | 2010-02-25 15:22:06 -0600 (Thu, 25 Feb 2010) | 18 lines

  Ensure that monitor recordings are written to the correct location (again)

  This is an extension to 248757. As such the dialplan test has been extended:

  exten => 5040, 1, monitor(wav,tmp/jeff/monitor_test,b)
  exten => 5040, n, dial(sip/5001)
  exten => 5041, 1, monitor(wav,/tmp/jeff/monitor_test2,b)
  exten => 5041, n, dial(sip/5001)
  exten => 5042, 1, monitor(wav,monitor_test3,b)
  exten => 5042, n, dial(sip/5001)
  exten => 5043, 1, monitor(wav,tmp/jeff/monitor_test3,m)
  exten => 5043, n, changemonitor(monitor_test4)
  exten => 5043, n, dial(sip/5001)
  exten => 5044, 1, monitor(wav,monitor_test4,m)
  exten => 5044, n, changemonitor(tmp/jeff/monitor_test5) ; this looks to fail by design and emits a warning
  exten => 5044, n, dial(sip/5001)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248952 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix incorrect ACL behavior when CIDR notation of "/0" is used.
Mark Michelson [Thu, 25 Feb 2010 22:41:48 +0000 (22:41 +0000)]
Fix incorrect ACL behavior when CIDR notation of "/0" is used.

AST-2010-003

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248946 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 248859 via svnmerge from
Tilghman Lesher [Thu, 25 Feb 2010 21:22:39 +0000 (21:22 +0000)]
Merged revisions 248859 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r248859 | tilghman | 2010-02-25 15:21:05 -0600 (Thu, 25 Feb 2010) | 15 lines

  Some platforms clear /var/run at boot, which makes connecting a remote console... difficult.

  Previously, we only created the default /var/run/asterisk directory at install
  time.  While we could create it in the init script, that would not work for
  those who start asterisk manually from the command line.  So the safest thing
  to do is to create it as part of the Asterisk boot process.  This also changes
  the ownership of the directory, because the pid and ctl files are created after
  we setuid/setgid.

  (closes issue #16802)
   Reported by: Brian
   Patches:
         20100224__issue16802.diff.txt uploaded by tilghman (license 14)
   Tested by: tzafrir
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248861 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 248757 via svnmerge from
Jeff Peeler [Thu, 25 Feb 2010 18:37:56 +0000 (18:37 +0000)]
Merged revisions 248757 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r248757 | jpeeler | 2010-02-25 12:06:54 -0600 (Thu, 25 Feb 2010) | 15 lines

  Ensure that monitor recordings are written to the correct location.

  Recordings should be placed in the monitor directory when a non-absolute path
  is used.

  Exact dialplan used for testing:
  exten => 5040, 1, monitor(wav,tmp/jeff/monitor_test,b)
  exten => 5040, n, dial(sip/5001)
  exten => 5041, 1, monitor(wav,/tmp/jeff/monitor_test2,b)
  exten => 5041, n, dial(sip/5001)
  exten => 5042, 1, monitor(wav,monitor_test3,b)
  exten => 5042, n, dial(sip/5001)

  ABE-2101
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248793 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoBlocked revisions 248668 via svnmerge
Jeff Peeler [Wed, 24 Feb 2010 23:00:40 +0000 (23:00 +0000)]
Blocked revisions 248668 via svnmerge

........
  r248668 | jpeeler | 2010-02-24 16:58:54 -0600 (Wed, 24 Feb 2010) | 8 lines

  Make deletion of temporary greetings work properly with IMAP_STORAGE

  This same patch was merged in 220833, but was skipped in this branch
  erroneously.

  (closes issue #16170)
  Reported by: francesco_r
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248671 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAlso kill the .i files, or else the build process will not recreate them, when we
Tilghman Lesher [Wed, 24 Feb 2010 22:44:55 +0000 (22:44 +0000)]
Also kill the .i files, or else the build process will not recreate them, when we
change flags.  Fixes a weird symbol problem mmichelson was having in a group branch,
but also applies to trunk.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248667 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 248582 via svnmerge from
Tilghman Lesher [Wed, 24 Feb 2010 21:17:26 +0000 (21:17 +0000)]
Merged revisions 248582 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r248582 | tilghman | 2010-02-24 15:02:18 -0600 (Wed, 24 Feb 2010) | 7 lines

  Remove color code sequences from verbose messages that go to logfiles.
  (closes issue #16786)
   Reported by: dodo
   Patches:
         logger2.patch uploaded by dodo (license 989)
   Tested by: tilghman
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248584 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove unnecessary warning message, make a couple of formatting tweaks
Russell Bryant [Wed, 24 Feb 2010 06:39:27 +0000 (06:39 +0000)]
Remove unnecessary warning message, make a couple of formatting tweaks

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248534 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd ASTERISK_FILE_VERSION macro.
Russell Bryant [Wed, 24 Feb 2010 06:38:10 +0000 (06:38 +0000)]
Add ASTERISK_FILE_VERSION macro.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248533 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoUnit test for ast_str API.
Mark Michelson [Tue, 23 Feb 2010 22:29:06 +0000 (22:29 +0000)]
Unit test for ast_str API.

Review: https://reviewboard.asterisk.org/r/517

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248489 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 248396 via svnmerge from
David Vossel [Tue, 23 Feb 2010 16:34:39 +0000 (16:34 +0000)]
Merged revisions 248396 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r248396 | dvossel | 2010-02-23 10:26:05 -0600 (Tue, 23 Feb 2010) | 9 lines

  fixes invite with replaces deadlock

  (closes issue #16862)
  Reported by: pwalker
  Patches:
        replaces_deadlock_1.4 uploaded by dvossel (license 671)
  Tested by: pwalker, dvossel
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248397 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMove the REF_DEBUG comment higher in the include list.
Mark Michelson [Mon, 22 Feb 2010 20:19:00 +0000 (20:19 +0000)]
Move the REF_DEBUG comment higher in the include list.

Uncommenting the REF_DEBUG definition where it was in the source
resulted in only a small part of the astobj2 references being logged
to a file. Moving this up higher in the include list causes all references
to be logged as they should be.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248347 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoBlocked revisions 248268 via svnmerge
Olle Johansson [Mon, 22 Feb 2010 13:55:24 +0000 (13:55 +0000)]
Blocked revisions 248268 via svnmerge

........
  r248268 | oej | 2010-02-22 14:52:34 +0100 (Mån, 22 Feb 2010) | 2 lines

  Don't log to debug unless debug is turned on
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248269 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMinor tweaks to comment blocks and includes.
Russell Bryant [Mon, 22 Feb 2010 06:45:52 +0000 (06:45 +0000)]
Minor tweaks to comment blocks and includes.

Fix the copyright lines, tweak doxygen formatting, and remove some unnecessary
includes.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248226 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoTweak copyright and author lines.
Russell Bryant [Mon, 22 Feb 2010 06:43:02 +0000 (06:43 +0000)]
Tweak copyright and author lines.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248225 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoCleanup transmit_* functions, part 1
Michiel van Baak [Sun, 21 Feb 2010 12:09:53 +0000 (12:09 +0000)]
Cleanup transmit_* functions, part 1

Break transmit_tone into transmit_start_tone and transmit_stop_tone as per the skinny protocol.

(closes issue #16874)
Reported by: wedhorn
Patches:
      skinny-clean01.diff uploaded by wedhorn (license 30)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248184 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoImprove support for RTCP reports without report blocks
Olle Johansson [Sat, 20 Feb 2010 22:37:22 +0000 (22:37 +0000)]
Improve support for RTCP reports without report blocks

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248108 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoBlocked revisions 248106 via svnmerge
Olle Johansson [Sat, 20 Feb 2010 22:29:50 +0000 (22:29 +0000)]
Blocked revisions 248106 via svnmerge

........
r248106 | oej | 2010-02-20 23:25:42 +0100 (Lör, 20 Feb 2010) | 2 lines

Make sure we support RTCP compound messages with zero reports

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248107 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agomfcr2 issue 0016844 - Fix portability bit fields and make mfcr2_immediate_accept...
Moises Silva [Fri, 19 Feb 2010 18:38:34 +0000 (18:38 +0000)]
mfcr2 issue 0016844 - Fix portability bit fields and make mfcr2_immediate_accept work again, reported and patched by korihor

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@248003 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agohandle_request_invite revise comment, fix coding guideline issues
David Vossel [Fri, 19 Feb 2010 17:40:26 +0000 (17:40 +0000)]
handle_request_invite revise comment, fix coding guideline issues

I'm working with this code right now trying to analyze a deadlock.
This change is just to clean up a few things before I make a more
complex patch.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247915 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 247910 via svnmerge from
Richard Mudgett [Fri, 19 Feb 2010 17:33:33 +0000 (17:33 +0000)]
Merged revisions 247910 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

................
  r247910 | rmudgett | 2010-02-19 11:18:49 -0600 (Fri, 19 Feb 2010) | 55 lines

  Merged revision 247904 from
  https://origsvn.digium.com/svn/asterisk/be/branches/C.2-...

  ..........
  r247904 | rmudgett | 2010-02-19 10:49:44 -0600 (Fri, 19 Feb 2010) | 49 lines

  Make chan_misdn DTMF processing consistent with other channel technologies.

  The processing of DTMF tones on the receiving side of an ISDN channel is
  inconsistent with the way it is handled in other channels, especially
  DAHDI analog.  This causes DTMF tones sent from an ISDN phone to be
  doubled at the connected party.

  We are using the following 2 options of misdn.conf
  1) astdtmf=yes
  2) senddtmf=yes

  Option one is necessary because the asterisk DSP DTMF detection is better
  than mISDN's internal DSP.  Not as many false positives.

  Option two is necessary to transmit DTMF tones end to end when mISDN
  channels are connected to SIP channels with out of band DTMF for example.

  The symptom is that DTMF tones sent by an ISDN phone are doubled on the
  way through asterisk when two mISDN channels are connected with a Local
  channel in between or if it is bridged to an analog channel.

  The doubling of DTMF tones is because DTMF is passed inband to asterisk by
  the mISDN channel and passed out of band once again after the release of
  the DTMF tone.  Passing it inband is wrong.  Neither an analog channel nor
  SIP channel passes DTMF inband if configured to inband DTMF.  Analog and
  SIP channels filter out the DTMF tones because they use the voice frames
  returned by ast_dsp_process.  But chan_misdn passes the unfiltered input
  voice frames instead.

  To overcome one aspect of the problem, the doubling of DTMF tones when two
  mISDN channels are directly bridged, someone made an 'optimization', where
  in that case the DTMF tone passed out-of-band to the peer channel is not
  translated to an inband tone at the transmit side.  This optimization is
  bad because it does not work in general.  For example, analog channels or
  mISDN channels when bridged through an intermediary local channel will
  generate DTMF tones from out-of-band information.  Also, of course, it
  must not be done when there is no inband DTMF available.

  This patch fixes the issue.  Now chan_misdn will filter the received
  inband DTMF signal the same as other channel types.

  Another change included: No need to build an extra translation path
  because ast_process_dsp does it if required.

  Patches:
   misdn-dtmf.patch

  JIRA ABE-2080
................

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247914 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRevert an errant part of a previous cleanup, to fix a memory corruption issue.
Tilghman Lesher [Thu, 18 Feb 2010 23:13:46 +0000 (23:13 +0000)]
Revert an errant part of a previous cleanup, to fix a memory corruption issue.

(closes issue #16368)
 Reported by: thirionjwf
 Patches:
       res_speech.c.patch uploaded by thirionjwf (license 955)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247841 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoIf the peer record is from realtime, it could be set to 0, due to MySQL not represent...
Tilghman Lesher [Thu, 18 Feb 2010 21:42:53 +0000 (21:42 +0000)]
If the peer record is from realtime, it could be set to 0, due to MySQL not representing NULL well in integer columns.

NULL means the value is not specified for the column, which normally means
the driver uses whatever is the default value.  However, on MySQL, placing
a NULL in either a float or integer column results in a retrieval of the 0
value.  Hence, users get an errant error on load.  This patch suppresses
that error and makes the value as if it was not there.

Note that this cannot be done in the realtime driver, because the lack of
difference between NULL and 0 can only be intepreted correctly by the
driver itself.  If we did it in the realtime driver, then it would be
effectively impossible to set any realtime field to 0, because it would act
as if the field were unspecified and possibly take on a different value.

(closes issue #16683)
 Reported by: wdoekes

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247787 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agofixes confbridge crash when no timing module is loaded.
David Vossel [Thu, 18 Feb 2010 21:23:48 +0000 (21:23 +0000)]
fixes confbridge crash when no timing module is loaded.

(closes issue #16471)
Reported by: kjotte
Patches:
      M16471.diff uploaded by junky (license 177)
Tested by: kjotte, junky

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247770 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agofixes Queue with C option crash
David Vossel [Thu, 18 Feb 2010 20:58:41 +0000 (20:58 +0000)]
fixes Queue with C option crash

(closes issue #16475)
Reported by: okrief
Patches:
      queue_crash.diff uploaded by dvossel (license 671)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247736 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 247651 via svnmerge from
Matthew Nicholson [Thu, 18 Feb 2010 19:39:37 +0000 (19:39 +0000)]
Merged revisions 247651 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r247651 | mnicholson | 2010-02-18 13:38:09 -0600 (Thu, 18 Feb 2010) | 6 lines

  Copy the calling party's account code to the called party if they don't already have one.

  (closes issue #16331)
  Reported by: bluefox
  Tested by: mnicholson
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247652 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix placing ISDN calls on hold preventing native bridging from being reexamined after...
Richard Mudgett [Thu, 18 Feb 2010 18:31:44 +0000 (18:31 +0000)]
Fix placing ISDN calls on hold preventing native bridging from being reexamined after a transfer.

Consider the following scenario:

                 /-- B
A == * == Network
                 \-- C

Party B calls party A (EuroISDN BRI phone)
Party A puts B on hold using the HOLD/RETRIEVE messages.
Party A calls party C.
Party A puts C on hold to talk with party B again.
Party A transfers B to C by hanging up.

The call does not get the opportunity to get re-transferred into the ISDN
network by the native bridge because native bridging is not being
reexamined after the initial transfer.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247609 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 247508 via svnmerge from
Leif Madsen [Thu, 18 Feb 2010 16:54:43 +0000 (16:54 +0000)]
Merged revisions 247508 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r247508 | lmadsen | 2010-02-18 11:53:44 -0500 (Thu, 18 Feb 2010) | 1 line

  Add additional link to best practices document per jsmith.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247509 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 247502 via svnmerge from
Leif Madsen [Thu, 18 Feb 2010 16:41:04 +0000 (16:41 +0000)]
Merged revisions 247502 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r247502 | lmadsen | 2010-02-18 11:38:17 -0500 (Thu, 18 Feb 2010) | 10 lines

  Add best practices documentation.

  (issue #16808)
  Reported by: lmadsen

  (issue #16810)
  Reported by: Nick_Lewis
  Tested by: lmadsen

  Review: https://reviewboard.asterisk.org/r/507/
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247503 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd a new manager event for our buddies status.
Philippe Sultan [Thu, 18 Feb 2010 16:34:08 +0000 (16:34 +0000)]
Add a new manager event for our buddies status.

The new JabberStatus event gives a concise view of the status change to the AMI
clients. Thanks fiddur!

(closes issue #16760)
Reported by: fiddur
Patches:
      244498.2.diff uploaded by fiddur (license 678)
Tested by: fiddur, phsultan

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247500 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 247422 via svnmerge from
Russell Bryant [Thu, 18 Feb 2010 04:20:11 +0000 (04:20 +0000)]
Merged revisions 247422 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r247422 | russell | 2010-02-17 22:19:01 -0600 (Wed, 17 Feb 2010) | 10 lines

  Tweak argument handling for wget in the sounds Makefile.

  1) Fix the check to see if we are using wget to not be full of fail.  The
  configure script populates this variable with the absolute path to wget if
  it is found, so it didn't work.

  2) Allow some extra arguments to be passed in for wget.  This is just a simple
  change to allow our Bamboo build script to tell wget to be quiet and not fill
  up our logs with download status output.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247423 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix a couple of bugs in test tab completion.
Mark Michelson [Wed, 17 Feb 2010 22:44:53 +0000 (22:44 +0000)]
Fix a couple of bugs in test tab completion.

1. Add missing unlock of lists.
2. Swap order of arguments to test_cat_cmp in complete_test_name.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247381 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoTab completion for test categories and names for "test show registered" and "test...
Mark Michelson [Wed, 17 Feb 2010 22:40:43 +0000 (22:40 +0000)]
Tab completion for test categories and names for "test show registered" and "test execute" CLI commands.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247380 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix two problems in ast_str functions found while writing a unit test.
Mark Michelson [Wed, 17 Feb 2010 21:22:40 +0000 (21:22 +0000)]
Fix two problems in ast_str functions found while writing a unit test.

1. The documentation for ast_str_set and ast_str_append state that
the max_len parameter may be -1 in order to limit the size of the
ast_str to its current allocated size. The problem was that the max_len
parameter in all cases was a size_t, which is unsigned. Thus a -1 was
interpreted as UINT_MAX instead of -1. Changing the max_len parameter
to be ssize_t fixed this issue.

2. Once issue 1 was fixed, there was an off-by-one error in the case
where we attempted to write a string larger than the current allotted
size to a string when -1 was passed as the max_len parameter. When trying
to write more than the allotted size, the ast_str's __AST_STR_USED was
set to 1 higher than it should have been. Thanks to Tilghman for quickly
spotting the offending line of code.

Oh, and the unit test that I referenced in the top line of this commit
will be added to reviewboard shortly. Sit tight...

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247335 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd support for GROUP_MATCH_COUNT regex matching on category
Jeff Peeler [Wed, 17 Feb 2010 19:51:53 +0000 (19:51 +0000)]
Add support for GROUP_MATCH_COUNT regex matching on category

Current support for regex matching was previously only available on the group.
Also, error reporting for regex failures has been added. In addition to this
feature enhancement a unit test has been written to check the regular expression
logic to ensure the count operation is working as expected.

(closes issue #16642)
Reported by: kobaz
Patches:
      groupmatch2.patch uploaded by kobaz (license 834)

Review: https://reviewboard.asterisk.org/r/503/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247295 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agomodified device2extension_test's category
David Vossel [Wed, 17 Feb 2010 19:23:35 +0000 (19:23 +0000)]
modified device2extension_test's category

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247282 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agounit test for combined device state mapping and device to exten state mapping
David Vossel [Wed, 17 Feb 2010 19:18:44 +0000 (19:18 +0000)]
unit test for combined device state mapping and device to exten state mapping

Review: https://reviewboard.asterisk.org/r/516/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247262 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoaddition of dynamic parkinglots feature
David Vossel [Wed, 17 Feb 2010 18:29:48 +0000 (18:29 +0000)]
addition of dynamic parkinglots feature

This feature allows for parkinglots to be created dynamically within
the dialplan.  Thanks to all who were involved with getting this patch
written and tested!

(closes issue #15135)
Reported by: IgorG
Patches:
      features.dynamic_park.v3.diff uploaded by IgorG (license 20)
      2009090400_dynamicpark.diff.txt uploaded by mvanbaak (license 7)
      dynamic_parkinglot.diff uploaded by dvossel (license 671)
Tested by: eliel, IgorG, acunningham, mvanbaak, zktech

Review: https://reviewboard.asterisk.org/r/352/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247248 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 247168 via svnmerge from
Mark Michelson [Wed, 17 Feb 2010 16:24:54 +0000 (16:24 +0000)]
Merged revisions 247168 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r247168 | mmichelson | 2010-02-17 10:24:17 -0600 (Wed, 17 Feb 2010) | 3 lines

  Make sure that when autofill is disabled that callers not in the front of the queue cannot place calls.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247169 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRTP documentation states that you can pass NULL as the module, so make sure that...
Tilghman Lesher [Wed, 17 Feb 2010 07:01:13 +0000 (07:01 +0000)]
RTP documentation states that you can pass NULL as the module, so make sure that's really the case.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247125 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMake all of the various rtpqos parameters in this branch available from the CHANNEL...
Tilghman Lesher [Wed, 17 Feb 2010 06:25:15 +0000 (06:25 +0000)]
Make all of the various rtpqos parameters in this branch available from the CHANNEL function.

Also includes a test for retrieving rtpqos parameters, including a NULL RTP
driver.  Additionally, some further separation of the SIP internal API into
headers was necessary.

(closes issue #16652)
 Reported by: kkm
 Patches:
       20100204__issue16652.diff.txt uploaded by tilghman (license 14)

Review: https://reviewboard.asterisk.org/r/501/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247124 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd va_end calls to __ast_str_helper.
Mark Michelson [Tue, 16 Feb 2010 23:44:33 +0000 (23:44 +0000)]
Add va_end calls to __ast_str_helper.

According to the man page for stdarg(3),
"Each invocation of va_copy() must be matched by a
corresponding invocation of va_end() in the same
function."

There were several cases in __ast_str_helper where
va_copy was not matched with a corresponding call
to va_end.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247076 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agogenerate connected line info update from info in h.323 packets
Alexandr Anikin [Tue, 16 Feb 2010 22:58:22 +0000 (22:58 +0000)]
generate connected line info update from info in h.323 packets
Tested by: benngard

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@247035 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd some clarifying documentation to the ast_str_set and ast_str_append functions.
Mark Michelson [Tue, 16 Feb 2010 21:15:38 +0000 (21:15 +0000)]
Add some clarifying documentation to the ast_str_set and ast_str_append functions.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246985 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoswap openssl with OpenSSL in warning message.
David Vossel [Tue, 16 Feb 2010 21:03:34 +0000 (21:03 +0000)]
swap openssl with OpenSSL in warning message.

(issue #16673)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246981 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agowarning message if openssl support is missing while attempting tls connection
David Vossel [Tue, 16 Feb 2010 20:54:48 +0000 (20:54 +0000)]
warning message if openssl support is missing while attempting tls connection

(closes issue #16673)
Reported by: michaesc
Patches:
      tls_error_msg.diff uploaded by dvossel (license 671)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246980 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd unit test for dialplan pattern matching.
Mark Michelson [Tue, 16 Feb 2010 18:29:42 +0000 (18:29 +0000)]
Add unit test for dialplan pattern matching.

This test works by reading input from arrays to build a sample
dialplan. From there, patterns are attempted to be matched against
said dialplan, with the expected match given. We then search in our
example dialplan to see if we find a match and if what we find matches
what we expected it to match.

(closes issue #16809)
Reported by: lmadsen
Tested by: mmichelson

Review: https://reviewboard.asterisk.org/r/504/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246942 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agofixes sample rate conversion issue with Monitor application
David Vossel [Tue, 16 Feb 2010 17:07:41 +0000 (17:07 +0000)]
fixes sample rate conversion issue with Monitor application

When using ast_seekstream with the read/write streams of a monitor,
the number of samples we are seeking must be of the same rate as the
stream or the jump calculation will be incorrect.  This patch adds logic
to correctly convert the number of samples to jump to the sample rate
the read/write stream is using.

For example, if the call is G722 (16khz) and the read/write stream is
recording a 8khz wav, seeking 320 samples of 16khz audio is not the
same as seeking 320 samples of 8khz audio when performing the ast_seekstream
on the stream.

ABE-2044

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246899 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRevert changes for now, pending discussion
Tilghman Lesher [Tue, 16 Feb 2010 15:36:53 +0000 (15:36 +0000)]
Revert changes for now, pending discussion

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246863 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd a few more targets for DEBUG_THREADLOCALS
Tilghman Lesher [Tue, 16 Feb 2010 05:25:11 +0000 (05:25 +0000)]
Add a few more targets for DEBUG_THREADLOCALS

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246827 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoChange the blanket rules to delete .lastclean on all CFLAGS menuselect targets to...
Tilghman Lesher [Tue, 16 Feb 2010 00:52:45 +0000 (00:52 +0000)]
Change the blanket rules to delete .lastclean on all CFLAGS menuselect targets to be more particular.

This change builds upon the recent change to menuselect to add 'touch_on_change'
as an attribute of both categories and members.  This should allow only the most
invasive defines to cause a complete rebuild, while defines which only affect a
subset of modules will only cause a rebuild of that smaller set.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246789 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAllow Timer B to be set on the peer, and ensure SIP rules are followed (or warn)...
Tilghman Lesher [Tue, 16 Feb 2010 00:19:38 +0000 (00:19 +0000)]
Allow Timer B to be set on the peer, and ensure SIP rules are followed (or warn) in comparison to Timer T1.

(closes issue #16643)
 Reported by: nahuelgreco
 Patches:
       20100204__issue16643.diff.txt uploaded by tilghman (license 14)
 Tested by: oej

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246724 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 246709 via svnmerge from
Tilghman Lesher [Mon, 15 Feb 2010 23:43:28 +0000 (23:43 +0000)]
Merged revisions 246709 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r246709 | tilghman | 2010-02-15 17:42:33 -0600 (Mon, 15 Feb 2010) | 5 lines

  Make the menuselect instructions correct by allowing 'make menuselect' to actually solve dependency problems.

  (Previously, it would fail out again with the same message about running
  'make menuselect', which was NOT at all helpful.)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246710 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRestore triedtopribridge flag code removed in -r211197.
Richard Mudgett [Mon, 15 Feb 2010 22:08:23 +0000 (22:08 +0000)]
Restore triedtopribridge flag code removed in -r211197.

Ooops.  Failed to note that we were inside a for loop and
pri_channel_bridge() needs to be executed only once.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246669 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoInstead of just automatically filtering out in the Makefile, give an indication of...
Tilghman Lesher [Mon, 15 Feb 2010 21:37:58 +0000 (21:37 +0000)]
Instead of just automatically filtering out in the Makefile, give an indication of dependencies in menuselect.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246667 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agochan_sip parse code refactoring plus two new unit tests
David Vossel [Mon, 15 Feb 2010 15:45:02 +0000 (15:45 +0000)]
chan_sip parse code refactoring plus two new unit tests

Code Refactoring Changes
- read_to_parts() moved to reqresp_parser.c and has been renamed as
  get_name_and_number()
- get_in_brackets() moved to reqresp_parser.c
- find_closing_quotes() added to sip_utils.h
Logic Changes
- get_name_and_number() now uses parse_uri() and get_calleridname()
  for parsing. Before this change only names within quotes were
  found, when names not within quotes are possible.
New Unit Tests
-sip_get_name_and_number_test
-sip_get_in_brackets_test

(closes issue #16707)
Reported by: Nick_Lewis
Patches:
      issue16706.diff uploaded by dvossel (license 671)

Review: https://reviewboard.asterisk.org/r/499/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246627 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 246545 via svnmerge from
David Vossel [Fri, 12 Feb 2010 23:32:33 +0000 (23:32 +0000)]
Merged revisions 246545 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r246545 | dvossel | 2010-02-12 17:30:17 -0600 (Fri, 12 Feb 2010) | 16 lines

  lock channel during datastore removal

  On channel destruction the channel's datastores are removed and
  destroyed.  Since there are public API calls to find and remove
  datastores on a channel, a lock should be held whenever datastores are
  removed and destroyed.  This resolves a crash caused by a race
  condition in app_chanspy.c.

  (closes issue #16678)
  Reported by: tim_ringenbach
  Patches:
        datastore_destroy_race.diff uploaded by tim ringenbach (license 540)
  Tested by: dvossel
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246546 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoBlocked revisions 246460 via svnmerge
Jason Parker [Fri, 12 Feb 2010 18:54:19 +0000 (18:54 +0000)]
Blocked revisions 246460 via svnmerge

........
  r246460 | qwell | 2010-02-12 12:52:28 -0600 (Fri, 12 Feb 2010) | 1 line

  Fix some silly formatting, and remove unnecessary option_debug checks
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246461 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agofixes areas where port should be removed from domain during parsing
David Vossel [Fri, 12 Feb 2010 17:59:39 +0000 (17:59 +0000)]
fixes areas where port should be removed from domain during parsing

A patch was committed recently that converted duplicate uri parsing
code to use the parse_uri function.  There were two instances where
this conversion did not mimic previous behavior exactly because the
port was not being parsed off the end of the domain. In order to do
this, a dummy pointer argument needs to be passed into parse_uri so
it will know it must parse out the port from the domain.  If a port
output paramenter is not present,   the domain is returned with the
port still attached.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246420 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoUpdated doc for OSP lookup application.
TransNexus OSP Development [Fri, 12 Feb 2010 08:30:05 +0000 (08:30 +0000)]
Updated doc for OSP lookup application.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246382 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agofixes some test description formatting inconsistencies so log file looks nice
David Vossel [Thu, 11 Feb 2010 21:57:37 +0000 (21:57 +0000)]
fixes some test description formatting inconsistencies so log file looks nice

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246338 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoastobj2 unit test and bug fix
David Vossel [Thu, 11 Feb 2010 18:42:25 +0000 (18:42 +0000)]
astobj2 unit test and bug fix

A bug was discovered during the creation of the astobj2 unit test.
When OBJ_MULTIPLE | OBJ_UNLINK is used, the objects being returned
had a ref count issue.  This patch resolves that.

Review: https://reviewboard.asterisk.org/r/496/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246299 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd a test module for the event API, test_event.c.
Russell Bryant [Wed, 10 Feb 2010 23:19:16 +0000 (23:19 +0000)]
Add a test module for the event API, test_event.c.

This module includes a single test so far that creates events using two
different methods and does some verification on the result to make sure
the correct data can be retrieved from the event that was created.

One bug was found in the event API while developing this test, which makes
me happy.  :-)

Review: https://reviewboard.asterisk.org/r/495/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246260 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoadditional parse_uri test and documentation
David Vossel [Wed, 10 Feb 2010 23:13:49 +0000 (23:13 +0000)]
additional parse_uri test and documentation

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246249 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agores_pktccops needs to be able to export a symbol for chan_mgcp
Tilghman Lesher [Wed, 10 Feb 2010 21:55:42 +0000 (21:55 +0000)]
res_pktccops needs to be able to export a symbol for chan_mgcp

(closes issue #16782)
 Reported by: nahuelgreco
 Patches:
       res_pktccops.exports uploaded by nahuelgreco (license 162)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246208 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFussy compiler on another machine...
Tilghman Lesher [Wed, 10 Feb 2010 21:24:10 +0000 (21:24 +0000)]
Fussy compiler on another machine...

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246204 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix weird issue with unit tests on optimized build - turned out to be a signing issue.
Tilghman Lesher [Wed, 10 Feb 2010 21:19:35 +0000 (21:19 +0000)]
Fix weird issue with unit tests on optimized build - turned out to be a signing issue.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246200 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 246115 via svnmerge from
David Vossel [Wed, 10 Feb 2010 17:49:34 +0000 (17:49 +0000)]
Merged revisions 246115 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r246115 | dvossel | 2010-02-10 11:44:20 -0600 (Wed, 10 Feb 2010) | 8 lines

  fixes random deadlock in app_queue with use_weight during reload

  (closes issue #16677)
  Reported by: tim_ringenbach
  Patches:
        app_queue_use_weight_deadlock.diff uploaded by tim ringenbach (license 540)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246116 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoChange channel state on local channels for busy,answer,ring.
Jeff Peeler [Wed, 10 Feb 2010 16:47:37 +0000 (16:47 +0000)]
Change channel state on local channels for busy,answer,ring.

Previously local channels channel state never changed. This became problematic
when the state of the other side of the local channel was lost, for example
during a masquerade. Changing the state of the local channel allows for the
scenario to be detected when the channel state is set to ringing, but the peer
isn't ringing. The specific problem scenario is described in 164201. Although
this was noted on one of the issues, here is the tested dialplan verified to
work:

exten => 9700,1,Dial(Local/*9700@default&Local/0009700@default)

exten => *9700,1,Set(GLOBAL(TESTCHAN)=${CHANNEL:0:${MATH(${LEN(${CHANNEL})}-1):0:2}}1)
exten => *9700,n,wait(3) ;3 works, 1 did not
exten => *9700,n,Dial(SIP/5001)

exten => 0009700,1,Wait(1) ;1 works, 3 did not
exten => 0009700,n,ChannelRedirect(${TESTCHAN},parkedcalls,701,1)

(closes issue #14992)
Reported by: davidw

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246070 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoSolaris doesn't like outputting a NULL to a %s in format strings.
Tilghman Lesher [Wed, 10 Feb 2010 16:01:28 +0000 (16:01 +0000)]
Solaris doesn't like outputting a NULL to a %s in format strings.

Detect all platforms that don't like that, either, and ensure that when documentation is
missing, we pass a non-NULL pointer when outputting the corresponding documentation.

(closes issue #16689)
 Reported by: bklang
 Patches:
       20100209__issue16689__with_tests.diff.txt uploaded by tilghman (license 14)

Review: https://reviewboard.asterisk.org/r/497/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246030 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoEnable warnings on atypical conditions for the FILTER function (suggested by mmichels...
Tilghman Lesher [Wed, 10 Feb 2010 15:36:57 +0000 (15:36 +0000)]
Enable warnings on atypical conditions for the FILTER function (suggested by mmichelson on the -dev list).

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@246022 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 245944 via svnmerge from
Tilghman Lesher [Wed, 10 Feb 2010 14:06:12 +0000 (14:06 +0000)]
Merged revisions 245944 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r245944 | tilghman | 2010-02-10 07:37:13 -0600 (Wed, 10 Feb 2010) | 2 lines

  Include examples of FILTER usage in extension patterns where a "." may be a risk.
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245945 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoVarious updates to the unit test API.
Russell Bryant [Tue, 9 Feb 2010 23:32:14 +0000 (23:32 +0000)]
Various updates to the unit test API.

1) It occurred to me that the difference in usage between the error ast_str and
the ast_test_update_status() usage has turned out to be a bit ambiguous in
practice.  In a lot of cases, the same message was being sent to both.
In other cases, it was only sent to one or the other.  My opinion now is that
in every case, I think it makes sense to do both; we should output it to the
CLI as well as save it off for logging purposes.

This change results in most of the changes in this diff, since it required
changes to all existing unit tests.  It also allowed for some simplifications
of unit test API implementation code.

2) Update ast_test_status_update() to include the file, function, and line
number for the code providing the update.

3) There are some formatting tweaks here and there.  Hopefully they aren't too
distracting for code review purposes.  Reviewboard's diff viewer seems to do a
pretty good job of pointing out when something is a whitespace change.

4) I moved the md5_test and sha1_test into the test_utils module.  It seemed
like a better approach since these tests are so tiny.

5) I changed the number of nodes used in heap_test_2 from 1 million to
100 thousand.  The only reason for this was to reduce the time it took
for this test to run.

6) Remove an unused function prototype that was at the bottom of utils.h.

7) Simplify test_insert() using the LIST_INSERT_SORTALPHA() macro.  The one
minor difference in behavior is that it no longer checks for a test registered
with the same name.

8) Expand the code in test_alloc() to provide specific error messages for each
failure case, to clearly inform developers if they forget to set the name,
summary, description, etc.

9) Tweak the output of the "test show registered" CLI command.  I swapped the
name and category to have the category first.  It seemed more natural since
that is the sort key.

10) Don't output the status ast_str in the "test show results" CLI command.
This is going to tend to be pretty verbose, so just leave that for the
detailed test logs (test generate results).

Review: https://reviewboard.asterisk.org/r/493/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245864 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agofixes a merging error for the iaxs and iaxsl off by one fix
David Vossel [Tue, 9 Feb 2010 23:18:58 +0000 (23:18 +0000)]
fixes a merging error for the iaxs and iaxsl off by one fix

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245804 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 245792 via svnmerge from
David Vossel [Tue, 9 Feb 2010 23:07:17 +0000 (23:07 +0000)]
Merged revisions 245792 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r245792 | dvossel | 2010-02-09 16:55:38 -0600 (Tue, 09 Feb 2010) | 12 lines

  Fixes iaxs and iaxsl size off by one issue.

  2^15 = 32768 which is the maximum allowed iax2 callnumber.
  Creating the iaxs and iaxsl array of size 32768 means the maximum
  callnumber is actually out of bounds.  This causes a nasty crash.

  (closes issue #15997)
  Reported by: exarv
  Patches:
        iax_fix.diff uploaded by dvossel (license 671)
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245793 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoEnsure frames are only freed once.
Tilghman Lesher [Tue, 9 Feb 2010 18:06:30 +0000 (18:06 +0000)]
Ensure frames are only freed once.

(closes issue #16361)
 Reported by: vlad
 Patches:
       20100208__issue16361.diff.txt uploaded by tilghman (license 14)
 Tested by: kenny, bloodoff, misaksen

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245729 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoThis commit removes an extra newline in T.38 generated SDP packets. This bug was...
Matthew Nicholson [Tue, 9 Feb 2010 17:40:04 +0000 (17:40 +0000)]
This commit removes an extra newline in T.38 generated SDP packets. This bug was caused by the fix introduced in r243860.

(closes issue #16766)
Reported by: raivisr
Patches:
      t38-sdp-newline-fix1.diff uploaded by mnicholson (license 96)
Tested by: raivisr

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245727 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoDon't offer MMR or JBIG transcoding during T.38 negotiation.
Kevin P. Fleming [Tue, 9 Feb 2010 16:24:52 +0000 (16:24 +0000)]
Don't offer MMR or JBIG transcoding during T.38 negotiation.

After further discussion with Steve Underwood, we should not (yet) be offering
to receive MMR or JBIG transcoded streams from T.38 endpoints. A future spandsp
release will support those features, and then they can be enabled during
negotiation

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245680 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix return value of get_ie_str() and get_ie_str_hash() for non-existent IE.
Russell Bryant [Mon, 8 Feb 2010 23:43:00 +0000 (23:43 +0000)]
Fix return value of get_ie_str() and get_ie_str_hash() for non-existent IE.

I found this bug while developing a unit test for event allocation.  Testing
is awesome.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245624 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoUNREGISTER instead of REGISTER in unload_module().
Russell Bryant [Mon, 8 Feb 2010 23:34:01 +0000 (23:34 +0000)]
UNREGISTER instead of REGISTER in unload_module().

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245623 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoUse memmove() instead of memcpy() for a case where the buffers overlap.
Russell Bryant [Mon, 8 Feb 2010 23:23:45 +0000 (23:23 +0000)]
Use memmove() instead of memcpy() for a case where the buffers overlap.

Once again, valgrind is freaking awesome.  That is all.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245610 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove object files from the channels/sip/ directory on make clean.
Russell Bryant [Mon, 8 Feb 2010 23:20:43 +0000 (23:20 +0000)]
Remove object files from the channels/sip/ directory on make clean.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245597 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoActually use _ASTLDFLAGS in the main/ and channels/ Makefiles.
Tilghman Lesher [Mon, 8 Feb 2010 22:31:40 +0000 (22:31 +0000)]
Actually use _ASTLDFLAGS in the main/ and channels/ Makefiles.

They were previously passed correctly, but they simply weren't used.  This
caused issues with various platforms whose builds needed to pass special
linker flags via the configure script.

(closes issue #16596)
 Reported by: pprindeville
 Patches:
       asterisk-1.6-astldflags.patch uploaded by pprindeville (license 347)
 Tested by: tilghman

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245578 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMerged revisions 245496 via svnmerge from
Jason Parker [Mon, 8 Feb 2010 20:41:05 +0000 (20:41 +0000)]
Merged revisions 245496 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.4

........
  r245496 | qwell | 2010-02-08 14:39:50 -0600 (Mon, 08 Feb 2010) | 4 lines

  Remove reference of documentation in source directory.

  People don't always build Asterisk from source (distro packages, anybody?).
........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245497 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoBlocked revisions 245422 via svnmerge
Olle Johansson [Mon, 8 Feb 2010 11:59:10 +0000 (11:59 +0000)]
Blocked revisions 245422 via svnmerge

........
r245422 | oej | 2010-02-08 12:57:52 +0100 (Mån, 08 Feb 2010) | 2 lines

Res_features depends on res_adsi in 1.4

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245423 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd the libvpb-dev package as a dependency.
Russell Bryant [Mon, 8 Feb 2010 04:51:00 +0000 (04:51 +0000)]
Add the libvpb-dev package as a dependency.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245385 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd a todo for pbx_gtkconsole for updating to gtk2.
Russell Bryant [Mon, 8 Feb 2010 04:43:55 +0000 (04:43 +0000)]
Add a todo for pbx_gtkconsole for updating to gtk2.

This module needs to be converted to gtk2, or we will eventually have to just
remove it from the tree.  gtk1 isn't even packaged anymore in the distro I'm
using.  I suspect nobody uses this and that nobody would notice if we removed
it.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245384 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd more packages required for building Asterisk modules.
Russell Bryant [Mon, 8 Feb 2010 04:40:16 +0000 (04:40 +0000)]
Add more packages required for building Asterisk modules.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245383 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoMake chan_usbradio compile.
Russell Bryant [Mon, 8 Feb 2010 04:31:03 +0000 (04:31 +0000)]
Make chan_usbradio compile.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245382 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd a SHA1 test module.
Russell Bryant [Mon, 8 Feb 2010 00:42:21 +0000 (00:42 +0000)]
Add a SHA1 test module.

Review: https://reviewboard.asterisk.org/r/492/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245345 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove unnecessary include, ast_md5_hash() comes from utils.h.
Russell Bryant [Mon, 8 Feb 2010 00:24:54 +0000 (00:24 +0000)]
Remove unnecessary include, ast_md5_hash() comes from utils.h.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245344 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoAdd an MD5 test module.
Russell Bryant [Sun, 7 Feb 2010 21:56:04 +0000 (21:56 +0000)]
Add an MD5 test module.

Review: https://reviewboard.asterisk.org/r/491/

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245307 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoFix a couple of spelling errors, and add format module dependencies.
Russell Bryant [Sun, 7 Feb 2010 21:43:11 +0000 (21:43 +0000)]
Fix a couple of spelling errors, and add format module dependencies.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245306 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoTweak formatting and add minor updates to some comments.
Russell Bryant [Sun, 7 Feb 2010 20:57:42 +0000 (20:57 +0000)]
Tweak formatting and add minor updates to some comments.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245269 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove an extra space.
Russell Bryant [Sun, 7 Feb 2010 20:56:57 +0000 (20:56 +0000)]
Remove an extra space.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245268 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove parsing of constantssrc from reload_config.
Mark Michelson [Sun, 7 Feb 2010 19:51:54 +0000 (19:51 +0000)]
Remove parsing of constantssrc from reload_config.

This config option is already handled by the function handle_common_options
and it is unnecessary to parse the value again.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245230 65c4cc65-6c06-0410-ace0-fbb531ad65f3

9 years agoRemove useless sip options related to hash table size.
Mark Michelson [Sat, 6 Feb 2010 14:43:03 +0000 (14:43 +0000)]
Remove useless sip options related to hash table size.

First off, these options weren't actually doing anything.
By the time the options were parsed, the peer and dialog
containers had already been allocated with their default
values.

Second, hash table size is something that doesn't really
make sense to change in a config file. If a user is that
interested in changing the hashtable size, he can modify
the source itself.

I have removed the parsing of the hash_peer, hash_user,
and hash_dialog options. I have removed the hash_user_size
variable altogether since it is not used at all. I also
changed hash_peer_size and hash_dialog_size to be constant,
and have changed the symbols to be in all caps as constants
typically are. I have also removed the entire section in
sip.conf.sample regarding configurable hashtable sizes.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@245192 65c4cc65-6c06-0410-ace0-fbb531ad65f3