asterisk/asterisk.git
14 years ago* first bits of decoding facility information elements
Nadi Sarrar [Tue, 8 Aug 2006 18:13:40 +0000 (18:13 +0000)]
* first bits of decoding facility information elements
* fail on misdn_cfg_init() if elements in the config enum don't match with the config structs in misdn_config.c
* implemented first bits for encoding ISDN facility information elements via ASN.1 descriptions
* using unnamed semaphore for syncing in misdn_thread
* advanced fax detection: configurable detect timeout and context to jump into

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39378 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoBlocked revisions 39350 via svnmerge
Russell Bryant [Tue, 8 Aug 2006 17:29:09 +0000 (17:29 +0000)]
Blocked revisions 39350 via svnmerge

........
r39350 | russell | 2006-08-08 13:04:10 -0400 (Tue, 08 Aug 2006) | 2 lines

Increase the buffer size for the callid (issue #7675, reported by pssatcs)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39353 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agosome code clean up and catch for a act_hook being called
Matt O'Gorman [Tue, 8 Aug 2006 17:07:41 +0000 (17:07 +0000)]
some code clean up and catch for a act_hook being called
without a packet.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39351 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAdd support for Sigma Designs cards. These basically allow you to offload dialtone...
Joshua Colp [Tue, 8 Aug 2006 17:02:45 +0000 (17:02 +0000)]
Add support for Sigma Designs cards. These basically allow you to offload dialtone generation to the board. If you're using a quicknet board where this might work, give it a try as well. (issue #6092 reported by ywalther - minor mods by moi)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39349 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoClarify volgain option a bit, it needs sox to work.
Joshua Colp [Tue, 8 Aug 2006 15:53:51 +0000 (15:53 +0000)]
Clarify volgain option a bit, it needs sox to work.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39333 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoBug 6237 - add volgain parameter, such that voicemail messages may be amplified after...
Tilghman Lesher [Tue, 8 Aug 2006 15:37:53 +0000 (15:37 +0000)]
Bug 6237 - add volgain parameter, such that voicemail messages may be amplified after recording

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39332 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMarking PRI_CAUSE as deprecated to be replaced by hangup(cause)
Olle Johansson [Tue, 8 Aug 2006 14:47:17 +0000 (14:47 +0000)]
Marking PRI_CAUSE as deprecated to be replaced by hangup(cause)
(Issue #7610)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39316 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years ago* added possibility to change echocancel settings during calls
Christian Richter [Tue, 8 Aug 2006 09:19:06 +0000 (09:19 +0000)]
* added possibility to change echocancel settings during calls
* removed the holded element from the chan_list struct, we know this from the
  state already
* added a few tweaks to make HOLD/RETRIEVE work again (TRANSFER does not work
  yet)
* added possibility to debug mISDN frames via syslog
* added misdn_lib_port_is_blocked function to check if a port is blocked
* removed ec_training=1 from empty_bc, we don't use ec_training anymore
* removed unused misdn_lib_get_l2_status function
* added the nt bit to dummy misdn_bchannel objects
* setting bc->out_fac_type to FACILITY_NONE defaultly
* removed HANDLER_DEBUG stuff for better readability

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39295 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agochange constant name from MAXSTRLEN to INITLEN since it is not the maximum
Russell Bryant [Tue, 8 Aug 2006 06:36:47 +0000 (06:36 +0000)]
change constant name from MAXSTRLEN to INITLEN since it is not the maximum
length, it is the initial length of the string when first allocated

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39274 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerge team/russell/ast_verbose_threadstorage
Russell Bryant [Tue, 8 Aug 2006 06:32:04 +0000 (06:32 +0000)]
Merge team/russell/ast_verbose_threadstorage

- instead of defining a free() wrapper in a bunch of files, define it as
  ast_free() in utils.h and remove the copies from all the files.

- centralize and abstract the code used for doing thread storage. The code
  lives in threadstorage.h, with one function being implemented in utils.c.
  This new API includes generic thread storage as well as special functions
  for handling thread local dynamic length string buffers.

- update ast_inet_ntoa() to use the new threadstorage API
- update ast_state2str() to use the new threadstorage API
- update ast_cli() to use the new threadstorage API

- Modify manager_event() to use thread storage. Instead of using a buffer of
  4096 characters as the workspace for building the manager event, use a thread
  local dynamic string.  Now there is no length limitation on the length of the
  body of a manager event.

- Significantly simplify the handling of ast_verbose() ...
  - Instead of using a static char buffer and a lock to make sure only one
    thread can be using ast_verbose() at a time, use a thread local dynamic
    string as the workspace for preparing the verbose message. Instead of
    locking around the entire function, the only locking done now is when the
    message has been built and is being deliviered to the list of registered
    verbose message handlers.
  - This function was doing a strdup() on every message passed to it and
    keeping a queue of the last 200 messages in memory. This has been
    completely removed. The only place this was used was that if there were
    any messages in the verbose queue when a verbose handler was registered,
    all of the messages in the queue would be fed to it.  So, I just made sure
    that the console verbose handler and the network verbose handler (for
    remote asterisk consoles) were registered before any verbose messages.
    pbx_gtkconsole and pbx_kdeconsole will now lose a few verbose messages at
    startup, but I didn't feel the performance hit of this message queue was
    worth saving the initial verbose output for these very rarely used modules.
  - I have removed the last three arguments to the verbose handlers, leaving
    only the string itself because they aren't needed anymore. For example,
    ast_verbose had some logic for telling the verbose handler to add
    a newline if the buffer was completely full. Now that the buffer can grow
    as needed, this doesn't matter anymore.
  - remove unused function, ast_verbose_dmesg() which was to dispatch the
    message queue
  - Convert the list of verbose handlers to use the linked list macros.

- add missing newline characters to a few ast_verbose() calls

- convert the list of log channels to use the linked list macros in logger.c

- fix close_logger() to close all of the files it opened for logging

- update ast_log() to use a thread local dynamic string for its workspace
  for preparing log messages instead of a buffer of size BUFSIZ (8kB on my
  system) allocated on the stack.  The dynamic string in this case is limited
  to only growing to a maximum size of BUFSIZ.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39272 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoThis patch allows for a user to send messages and
Matt O'Gorman [Mon, 7 Aug 2006 22:03:50 +0000 (22:03 +0000)]
This patch allows for a user to send messages and
monitor a jabber connection over manager.
patches from 7673 and 7666 with minor changes.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39248 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMany many code cleanup changes given to me by Oej
Matt O'Gorman [Mon, 7 Aug 2006 21:15:28 +0000 (21:15 +0000)]
Many many code cleanup changes given to me by Oej
Thanks, sorry I didn't put this in forever ago.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39229 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoremove some extraneous 'followme' in prompt names
Kevin P. Fleming [Mon, 7 Aug 2006 18:47:33 +0000 (18:47 +0000)]
remove some extraneous 'followme' in prompt names

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39206 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoThis commit folds in changes to both stringfields (some enhancements to the ...field_...
Steve Murphy [Mon, 7 Aug 2006 18:12:51 +0000 (18:12 +0000)]
This commit folds in changes to both stringfields (some enhancements to the ...field_set() macro, to optimize setting strings to empty, resetting strings to shorter contents, etc.) and to chan_iax2.c, to use stringfields in the user, peer, and pvt structs. Has been running stably on iaxtel, but while iaxtel has a large registration volume, it doesn't seem to have a high call volume. So far, it seems to reduce heap usage by over half. YMMV\! Please report any IAX bugs that might involve stringfields\!

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39203 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAdd placeholder for sla.conf sample in configs/. Please update with
Olle Johansson [Mon, 7 Aug 2006 17:01:40 +0000 (17:01 +0000)]
Add placeholder for sla.conf sample in configs/. Please update with
correct information not to overload the switchboard with calls to sales.
 Thank you! ;-)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39182 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoFix accidental removal of if() in commit 38905
Olle Johansson [Mon, 7 Aug 2006 16:20:23 +0000 (16:20 +0000)]
Fix accidental removal of if() in commit 38905

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39171 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoThe contents of branch teams/murf/bug_7598 are being folded in here.
Steve Murphy [Mon, 7 Aug 2006 12:59:47 +0000 (12:59 +0000)]
The contents of branch teams/murf/bug_7598 are being folded in here.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39151 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoIssue #7651 - always send subscription-state (AuPix)
Olle Johansson [Mon, 7 Aug 2006 09:22:43 +0000 (09:22 +0000)]
Issue #7651 - always send subscription-state (AuPix)
  Small change to the fix in the report.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39132 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agochanged naming of mISDN channels, so that hinting works proper
Christian Richter [Mon, 7 Aug 2006 09:09:17 +0000 (09:09 +0000)]
changed naming of mISDN channels, so that hinting works proper

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39131 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerge my applicationmap_fixup branch to address the issues described in this
Russell Bryant [Mon, 7 Aug 2006 04:15:52 +0000 (04:15 +0000)]
Merge my applicationmap_fixup branch to address the issues described in this
post to the asterisk-dev mailing list:
  http://lists.digium.com/pipermail/asterisk-dev/2006-August/022174.html

This implements full control over both which channel(s) can activate a dynamic
feature, as well as which channel to run the application on.  I also updated
the documentation on the applicationmap in features.conf.sample in hopes that
the configuration is more clear.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39109 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 39081 via svnmerge from
Russell Bryant [Mon, 7 Aug 2006 01:32:51 +0000 (01:32 +0000)]
Merged revisions 39081 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r39081 | russell | 2006-08-06 21:28:29 -0400 (Sun, 06 Aug 2006) | 7 lines

Fix a crash reported to me by hads on IRC.  This crash would occur with the use
of the "distinctiveringaftercid" option.  Also, on this user's system, the crash
would only occur when built without optimizations.  This is because the bug is
that the code would write past the end of an array that was allocated on the
stack, and the structure of the stack is different with or without optimizations
enabled.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39082 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agominor cleanups
Russell Bryant [Mon, 7 Aug 2006 00:26:11 +0000 (00:26 +0000)]
minor cleanups
- use appropriate types in some assignments
- use ast_strlen_zero()
- don't manually free cid fields since ast_set_callerid() will handle it

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39058 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 39056 via svnmerge from
Joshua Colp [Mon, 7 Aug 2006 00:17:12 +0000 (00:17 +0000)]
Merged revisions 39056 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r39056 | file | 2006-08-06 20:15:51 -0400 (Sun, 06 Aug 2006) | 2 lines

Reset our stream and vstream pointers back to NULL so that any generator that uses them (file based MOH) will not try to close them again. (issue #7668 reported by jmls)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39057 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years ago- add DIALING_OFFHOOK and PRERING to ast_state2str
Russell Bryant [Sat, 5 Aug 2006 20:58:47 +0000 (20:58 +0000)]
- add DIALING_OFFHOOK and PRERING to ast_state2str
- don't include a newline character when building a response for an unknown
  state
(issue #7645, jonty)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39032 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agosuppress a compiler warning about the usage of a potentially uninitialized
Russell Bryant [Sat, 5 Aug 2006 11:25:36 +0000 (11:25 +0000)]
suppress a compiler warning about the usage of a potentially uninitialized
variable

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@39014 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38982 via svnmerge from
Russell Bryant [Sat, 5 Aug 2006 09:03:27 +0000 (09:03 +0000)]
Merged revisions 38982 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38982 | russell | 2006-08-05 05:01:37 -0400 (Sat, 05 Aug 2006) | 6 lines

Always generate a Newstate event in ast_setstate() instead of making it a
Newchannel event if the state was AST_STATE_DOWN.  The Newchannel event will
always be generated in ast_request(), so this just causes a duplicated
Newchannel event in some cases.
(issue #7506, repoted by capouch, fixed by me)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38994 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38972 via svnmerge from
Russell Bryant [Sat, 5 Aug 2006 08:10:20 +0000 (08:10 +0000)]
Merged revisions 38972 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38972 | russell | 2006-08-05 04:08:48 -0400 (Sat, 05 Aug 2006) | 3 lines

remove duplicate queue log entry when the caller exits on a timeout
(issue #7616, ppyy)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38973 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38950 via svnmerge from
Russell Bryant [Sat, 5 Aug 2006 07:22:27 +0000 (07:22 +0000)]
Merged revisions 38950 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38950 | russell | 2006-08-05 03:21:12 -0400 (Sat, 05 Aug 2006) | 3 lines

don't advertise that this function can set a SIP header when it can only
do reads

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38951 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38928 via svnmerge from
Russell Bryant [Sat, 5 Aug 2006 06:39:43 +0000 (06:39 +0000)]
Merged revisions 38928 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38928 | russell | 2006-08-05 02:37:59 -0400 (Sat, 05 Aug 2006) | 3 lines

make sure the priv-callerintros directory exists before trying to create a file
there (issue #7659, patch by hads, with some modifications by me)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38929 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38903-38904 via svnmerge from
Russell Bryant [Sat, 5 Aug 2006 05:26:29 +0000 (05:26 +0000)]
Merged revisions 38903-38904 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38903 | russell | 2006-08-05 01:07:39 -0400 (Sat, 05 Aug 2006) | 2 lines

suppress a compiler warning about the usage of a potentially uninitialized variable

........
r38904 | russell | 2006-08-05 01:08:50 -0400 (Sat, 05 Aug 2006) | 10 lines

Fix an issue that would cause a NewCallerID manager event to be generated
before the channel's NewChannel event.  This was due to a somewhat recent
change that included using ast_set_callerid() where it wasn't before.  This
function should not be used in the channel driver "new" functions.
(issue #7654, fixed by me)

Also, fix a couple minor bugs in usecount handling.  chan_iax2 could have
increased the usecount but then returned an error.  The place where chan_sip
increased the usecount did not call ast_update_usecount()

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38905 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoIt is better to hold a lock then not hold a lock when you are supposed to.
Joshua Colp [Sat, 5 Aug 2006 00:13:44 +0000 (00:13 +0000)]
It is better to hold a lock then not hold a lock when you are supposed to.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38887 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMinor optimization but use a linkedlists.h macro that pops the head off the list...
Joshua Colp [Fri, 4 Aug 2006 23:17:40 +0000 (23:17 +0000)]
Minor optimization but use a linkedlists.h macro that pops the head off the list instead of getting the first entry and then removing it.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38871 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoDon't assume that a referred by URI will always exist (issue #7641 reported and fixed...
Joshua Colp [Fri, 4 Aug 2006 18:40:47 +0000 (18:40 +0000)]
Don't assume that a referred by URI will always exist (issue #7641 reported and fixed by AuPix)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38853 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38825 via svnmerge from
Joshua Colp [Thu, 3 Aug 2006 19:55:05 +0000 (19:55 +0000)]
Merged revisions 38825 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38825 | file | 2006-08-03 15:54:02 -0400 (Thu, 03 Aug 2006) | 2 lines

Treat the file as invalid if we have no valid formats for it (issue #7643 reported by KNK)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38826 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoUpdate documentation on realtime; add a workaround for lack of realtime hints by...
Tilghman Lesher [Thu, 3 Aug 2006 19:05:06 +0000 (19:05 +0000)]
Update documentation on realtime; add a workaround for lack of realtime hints by using func_odbc

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38823 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years ago* removed pp_l2_check (fixed L2 bug in mISDNuser)
Christian Richter [Thu, 3 Aug 2006 16:38:00 +0000 (16:38 +0000)]
* removed pp_l2_check (fixed L2 bug in mISDNuser)
* added blocking flag to stack object. A port can be blocked/unblocked from the
  cli
* added EVENT_PORT_ALARM to send alarm infos to the chan_misdn.c layer (later
  we can add a manager event for that)
* added block_on_alarm option, to block the port whenever a ALARM occurs
* added need_busy flag to indicate if we've sended a CONTROL_BUSY already
* changed a bunch of cb_log(-1,..) to cb_log(0,..) due to funny behaviour in
  recent asterisk ast_log messages..
* fixed a few ETSI state violations, especially when finishing calls in
  different seldom states
* changed debug levels a lot to make the log more readable in low debuglevels
* some first fixes for the HOLD/RETRIEVE stuff (doesn't work totally still)
* removed the PRECONNECTED state stuff
* added cause 27 when we get a CLEANUP directly after a outgoing SETUP, this
  creates a CHANISUNAVAIL instead of a NOANSWER
* removed the addr pointer from "misdn show stacks" that's not needed anymore
  and makes the output more unreadable
* added cause saving on RELEASE/RELEASE_COMPLETE
* set cause to 16 on prepare_bc
* removed stack getting from ph_control functions, we don't really need it
  there
* added beroec api

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38801 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoUpdate statement on FreeTDS with realtime, due to revisions in res_odbc
Tilghman Lesher [Thu, 3 Aug 2006 13:16:00 +0000 (13:16 +0000)]
Update statement on FreeTDS with realtime, due to revisions in res_odbc

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38785 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38761 via svnmerge from
Tilghman Lesher [Thu, 3 Aug 2006 05:24:43 +0000 (05:24 +0000)]
Merged revisions 38761 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38761 | tilghman | 2006-08-03 00:22:19 -0500 (Thu, 03 Aug 2006) | 2 lines

Bug 7648 - Checking wrong count for plurality on new messages for Dutch language

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38762 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38731 via svnmerge from
Kevin P. Fleming [Wed, 2 Aug 2006 19:32:38 +0000 (19:32 +0000)]
Merged revisions 38731 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38731 | kpfleming | 2006-08-02 14:29:56 -0500 (Wed, 02 Aug 2006) | 3 lines

fix brain-damage I introduced when trying to fix the CANCEL/BYE sending mechanism for pending INVITES
accept unknown 1xx responses as 183 responses (as RFC3261 mandates we should do)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38732 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agodtmf support. not everything else, trying to clear out those other bugs
Matt O'Gorman [Wed, 2 Aug 2006 01:00:24 +0000 (01:00 +0000)]
dtmf support. not everything else, trying to clear out those other bugs
but more to come i guess.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38714 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38686 via svnmerge from
Kevin P. Fleming [Tue, 1 Aug 2006 23:09:28 +0000 (23:09 +0000)]
Merged revisions 38686 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38686 | kpfleming | 2006-08-01 18:07:06 -0500 (Tue, 01 Aug 2006) | 2 lines

ensure that the 'feature digit timeout' value is taken into account when deciding how long the bridge should run (this fixes a problem report where a digit press that did not invoke a feature is never passed across the bridge)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38687 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38654 via svnmerge from
Joshua Colp [Tue, 1 Aug 2006 19:21:30 +0000 (19:21 +0000)]
Merged revisions 38654 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38654 | file | 2006-08-01 15:20:05 -0400 (Tue, 01 Aug 2006) | 2 lines

Close the stream when file based MOH stop. This won't get rid of their position in the file but it will cause the translation path to be setup again. (issue #7634 reported by asimpson)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38655 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoThis branch will resolve the bug 7635.
Steve Murphy [Tue, 1 Aug 2006 19:03:32 +0000 (19:03 +0000)]
This branch will resolve the bug 7635.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38652 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoSet permissions completely open and modify by umask
Tilghman Lesher [Tue, 1 Aug 2006 13:15:15 +0000 (13:15 +0000)]
Set permissions completely open and modify by umask

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38633 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38611 via svnmerge from
Kevin P. Fleming [Mon, 31 Jul 2006 21:31:44 +0000 (21:31 +0000)]
Merged revisions 38611 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38611 | kpfleming | 2006-07-31 16:14:11 -0500 (Mon, 31 Jul 2006) | 4 lines

don't reissue hangup requests for SIP channels that have expired their RTP timeouts (one time is enough)
don't rescan the SIP private structure list too fast, it can cause channels to not be able to hang up (issue #7495, and probably others)
use ast_softhangup_nolock() since we already hold the channel's lock

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38612 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agomake the counting of ingress, outgress, and total frames thread-safe
Russell Bryant [Mon, 31 Jul 2006 17:27:42 +0000 (17:27 +0000)]
make the counting of ingress, outgress, and total frames thread-safe

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38587 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoBlocked revisions 38585 via svnmerge
Joshua Colp [Mon, 31 Jul 2006 17:11:51 +0000 (17:11 +0000)]
Blocked revisions 38585 via svnmerge

........
r38585 | file | 2006-07-31 13:09:10 -0400 (Mon, 31 Jul 2006) | 2 lines

Add missing code to bring transferee channel out of MOH/autoservice under certain circumstance (issue #7611 reported by guillecabeza with minor mods by myself)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38586 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoPoking at a structure when it may not even be allocated is not healthy. Essentially...
Joshua Colp [Mon, 31 Jul 2006 15:24:05 +0000 (15:24 +0000)]
Poking at a structure when it may not even be allocated is not healthy. Essentially make sure an RTCP structure exists before trying to delete it's scheduled item. (issue #7514 reported by jmls fixed by AuPix)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38573 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38546-38547 via svnmerge from
Russell Bryant [Mon, 31 Jul 2006 04:10:43 +0000 (04:10 +0000)]
Merged revisions 38546-38547 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38546 | russell | 2006-07-31 00:04:02 -0400 (Mon, 31 Jul 2006) | 2 lines

Make the frame counting done with TRACE_FRAMES defined thread-safe

........
r38547 | russell | 2006-07-31 00:06:16 -0400 (Mon, 31 Jul 2006) | 2 lines

one more small tweak for thread-safety of TRACE_FRAMES

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38548 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agosuppress some compiler warnings ...
Russell Bryant [Sun, 30 Jul 2006 23:16:23 +0000 (23:16 +0000)]
suppress some compiler warnings ...

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38533 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoallow the gtk console to build and load
Russell Bryant [Sun, 30 Jul 2006 23:11:07 +0000 (23:11 +0000)]
allow the gtk console to build and load

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38532 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAdd details about flash and wait to doxygen about DTMF.
Tilghman Lesher [Sun, 30 Jul 2006 15:02:34 +0000 (15:02 +0000)]
Add details about flash and wait to doxygen about DTMF.
Reformat the lines to break at column 80.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38521 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoBlocked revisions 38501 via svnmerge
Joshua Colp [Sat, 29 Jul 2006 23:19:36 +0000 (23:19 +0000)]
Blocked revisions 38501 via svnmerge

........
r38501 | file | 2006-07-29 19:18:00 -0400 (Sat, 29 Jul 2006) | 2 lines

How many attempts does it take to make a SIP URI parser that works well? I'm up to 5 personally. On to the good stuff - parse the domain first, user second, and get rid of port & options/params last. (issue #7616 reported by andrew)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38502 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAllow updates to match specific lines, allow specification of object or
Mark Spencer [Sat, 29 Jul 2006 22:02:37 +0000 (22:02 +0000)]
Allow updates to match specific lines, allow specification of object or
no when appending.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38489 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoupdate cleancount because the channel structure changed today
Kevin P. Fleming [Sat, 29 Jul 2006 00:19:18 +0000 (00:19 +0000)]
update cleancount because the channel structure changed today

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38476 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoplay a beep tone into the spied-on channel if we are about to whisper to them
Kevin P. Fleming [Fri, 28 Jul 2006 23:47:05 +0000 (23:47 +0000)]
play a beep tone into the spied-on channel if we are about to whisper to them

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38466 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoadd ExtenSpy variant of ChanSpy
Kevin P. Fleming [Fri, 28 Jul 2006 23:36:06 +0000 (23:36 +0000)]
add ExtenSpy variant of ChanSpy
implement whisper mode for ExtenSpy/ChanSpy

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38465 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agothe pure attribute was not implemented before gcc 2.96. Since we have this
Russell Bryant [Fri, 28 Jul 2006 23:30:18 +0000 (23:30 +0000)]
the pure attribute was not implemented before gcc 2.96.  Since we have this
version check here anwyay, define attribute_pure to be empty if it's an earlier
version.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38464 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoadd macros for the pure and const attributes to compiler.h, in case they ever
Russell Bryant [Fri, 28 Jul 2006 22:50:54 +0000 (22:50 +0000)]
add macros for the pure and const attributes to compiler.h, in case they ever
need to be handled differently for a specific compiler

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38454 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAdd the function attribute "pure" or "const" to various functions that perform
Russell Bryant [Fri, 28 Jul 2006 22:14:49 +0000 (22:14 +0000)]
Add the function attribute "pure" or "const" to various functions that perform
int to string or string to int operations.

"pure" essentially says that this function has no side effects aside from its
result, and the result depends on nothing else other than its arguments and
global variables.  "const" is a more strict form of "pure", where the function
also doesn't access any global variables.

From the gcc manual: "Such a function can be subject to common subexpression
elimination and loop optimization just as an arithmetic operator would be."
This also tells the compiler that it is safe to call the function fewer times
than the code says to, given the same arguments, since the result will always
be the same.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38452 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agojust for fun, let's lock the whisper buffer for writing AND reading :-)
Kevin P. Fleming [Fri, 28 Jul 2006 19:58:53 +0000 (19:58 +0000)]
just for fun, let's lock the whisper buffer for writing AND reading :-)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38441 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAs discussed and decided on the asterisk-dev mailing list ...
Russell Bryant [Fri, 28 Jul 2006 19:17:56 +0000 (19:17 +0000)]
As discussed and decided on the asterisk-dev mailing list ...
- Fix some breakage I introduced a while ago that made the timestamps option
  not functional for CLI verbose output.
- Remove the use of the timestamps option for log output, since it was not
  functional.
- clarify text referring to the timestamps option so that it is clear that it
  only applies to CLI verbose output

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38439 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agomove slinfactory structure definition back to header... it's just easier to use this way
Kevin P. Fleming [Fri, 28 Jul 2006 18:59:59 +0000 (18:59 +0000)]
move slinfactory structure definition back to header... it's just easier to use this way
add infrastructure for whispering onto a channel

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38422 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38420 via svnmerge from
Joshua Colp [Fri, 28 Jul 2006 18:53:43 +0000 (18:53 +0000)]
Merged revisions 38420 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38420 | file | 2006-07-28 14:49:00 -0400 (Fri, 28 Jul 2006) | 2 lines

Make a copy of the request URI in check_user_full instead of modifying the one on the structure, and also strip params properly from the user portion of the SIP URI so as to preserve the domain (issue #7552 reported by dan42)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38421 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agomore slinfactory structure definition to inside implementation module
Kevin P. Fleming [Fri, 28 Jul 2006 18:03:57 +0000 (18:03 +0000)]
more slinfactory structure definition to inside implementation module
make read/write/hold work on samples, not bytes
add an API call to find out how many samples are available in a slinfactory

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38418 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agomore simplification, and correct a bug i introduced in the last commit
Kevin P. Fleming [Thu, 27 Jul 2006 23:16:08 +0000 (23:16 +0000)]
more simplification, and correct a bug i introduced in the last commit
fix prototype for a channel walking function to use a const input pointer
use existing channel walk by name prefix instead of reproducing that code in this app

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38389 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoremove local channel finding wrappers
Kevin P. Fleming [Thu, 27 Jul 2006 23:00:27 +0000 (23:00 +0000)]
remove local channel finding wrappers
move guts of dialplan application into separate function, so it can be shared bythe new application i'm about to add :-)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38374 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoblock this patch from 1.2, it's already fixed here as part of a larger commit
Kevin P. Fleming [Thu, 27 Jul 2006 22:24:19 +0000 (22:24 +0000)]
block this patch from 1.2, it's already fixed here as part of a larger commit

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38371 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agorestore infinite timeout for ast_waitfor() call
Kevin P. Fleming [Thu, 27 Jul 2006 22:21:46 +0000 (22:21 +0000)]
restore infinite timeout for ast_waitfor() call

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38369 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agovarious cleanups:
Kevin P. Fleming [Thu, 27 Jul 2006 22:20:52 +0000 (22:20 +0000)]
various cleanups:
  use API call for finding channel by name prefix
  code formatting to match guidelines (lost about half the of the indenting)
  remove useless automatic variable initializations
  don't set the spying channel's read format to SLINEAR when we don't do anything with the voice frames we read from it anyway
  use proper option argument checking for volume argument

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38368 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38347 via svnmerge from
Kevin P. Fleming [Thu, 27 Jul 2006 15:42:49 +0000 (15:42 +0000)]
Merged revisions 38347 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38347 | kpfleming | 2006-07-27 10:40:03 -0500 (Thu, 27 Jul 2006) | 2 lines

do a better job avoiding translation path teardown/setup when not needed

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38348 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agofix seg fault when the parked call that timed out was the last one in the list
Russell Bryant [Thu, 27 Jul 2006 04:44:49 +0000 (04:44 +0000)]
fix seg fault when the parked call that timed out was the last one in the list
of parked calls (fixes issue #7565)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38330 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38328 via svnmerge from
Russell Bryant [Thu, 27 Jul 2006 04:28:09 +0000 (04:28 +0000)]
Merged revisions 38328 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38328 | russell | 2006-07-27 00:25:41 -0400 (Thu, 27 Jul 2006) | 4 lines

Fix crash when using the "regexten" option with MALLOC_DEBUG enabled.  This was
not reported in the bug tracker but the same bug has been demonstrated in other
places in the code.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38329 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38310 via svnmerge from
Kevin P. Fleming [Thu, 27 Jul 2006 02:46:22 +0000 (02:46 +0000)]
Merged revisions 38310 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38310 | kpfleming | 2006-07-26 21:43:49 -0500 (Wed, 26 Jul 2006) | 2 lines

don't do useless translation destroy/build when the channel is already in the correct format

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38311 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38288 via svnmerge from
Russell Bryant [Thu, 27 Jul 2006 02:00:20 +0000 (02:00 +0000)]
Merged revisions 38288 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38288 | russell | 2006-07-26 21:58:41 -0400 (Wed, 26 Jul 2006) | 3 lines

fix a crash when MALLOC_DEBUG is enabled and the regexten is enabled. The crash
would occur when the extension got removed. (fixes issue #7484)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38291 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agofix crashes with MALLOC_DEBUG enabled that were a result of my recent thread
Russell Bryant [Thu, 27 Jul 2006 01:49:57 +0000 (01:49 +0000)]
fix crashes with MALLOC_DEBUG enabled that were a result of my recent thread
storage changes (fixes issue #7595)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38286 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agofix some breakage of ast_cli() that resulted in seg faults on Josh's machine.
Russell Bryant [Thu, 27 Jul 2006 01:29:57 +0000 (01:29 +0000)]
fix some breakage of ast_cli() that resulted in seg faults on Josh's machine.
I'm not sure why this never caused problems for me ...

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38283 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agofix setting the CFLAGS for building codec libs so that they are built with
Russell Bryant [Wed, 26 Jul 2006 23:12:47 +0000 (23:12 +0000)]
fix setting the CFLAGS for building codec libs so that they are built with
astmm support and astmm doesn't get really upset and complain that it is being
asked to free memory that was never allocated

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38271 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoTypo found on the -dev list
Tilghman Lesher [Wed, 26 Jul 2006 19:59:25 +0000 (19:59 +0000)]
Typo found on the -dev list

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38258 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoOnly unlock these if they were locked on entry
Tilghman Lesher [Wed, 26 Jul 2006 17:43:17 +0000 (17:43 +0000)]
Only unlock these if they were locked on entry

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38250 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerged revisions 38234 via svnmerge from
Joshua Colp [Wed, 26 Jul 2006 15:27:45 +0000 (15:27 +0000)]
Merged revisions 38234 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38234 | file | 2006-07-26 11:26:06 -0400 (Wed, 26 Jul 2006) | 2 lines

Put default callerid into contact when the one specified is either NULL or has a zero string length. (issue #7590 reported by key2)

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38235 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agomerge the changes from my ast_cli_tls branch ...
Russell Bryant [Wed, 26 Jul 2006 07:48:48 +0000 (07:48 +0000)]
merge the changes from my ast_cli_tls branch ...

Instead of having a static buffer size of 16k, start out with 256 bytes and
increase the size of the buffer as needed.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38223 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoThis exact deadlock situation that I observed can't happen in trunk due to the
Russell Bryant [Tue, 25 Jul 2006 19:51:31 +0000 (19:51 +0000)]
This exact deadlock situation that I observed can't happen in trunk due to the
recent hold changes so that MOH is not started on the bridged channel directly.
However, the change is still not a bad idea.

Merged revisions 38200 via svnmerge from
https://origsvn.digium.com/svn/asterisk/branches/1.2

........
r38200 | russell | 2006-07-25 15:43:38 -0400 (Tue, 25 Jul 2006) | 6 lines

This resolves a deadlock that a tech support customer was getting frequently
when his users would answer call waiting. If another thread is currently
holding the zt_pvt lock for the first channel, unlock both channels and let
asterisk retry the native bridge, just like what is done for the second channel
directly below these changes.

........

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38201 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoAdded a small fix to exclude the AMD optimized assembly file from being included...
Steve Murphy [Mon, 24 Jul 2006 20:58:27 +0000 (20:58 +0000)]
Added a small fix to exclude the AMD optimized assembly file from being included in the compile.
I don't particular like the approach, which basically tries to exclude the file based on
NOT being in a list of arches. This, IMHO, is backwards, it should be a list of AMD arches
instead. "Better to light a candle, than to curse the darkness", is apropo here. But since I don't
have an exhaustive list of what those arches could be, I leave as-is for now.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38187 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoblocked rev 38167 on the 1.2 branch from trunk, because it's
Steve Murphy [Mon, 24 Jul 2006 17:24:47 +0000 (17:24 +0000)]
blocked rev 38167 on the 1.2 branch from trunk, because it's
coded differently on trunk, would conflict and have to be hand-merged
anyway.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38168 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoStart out with cleared RTP payload structures instead of defaults. This should preven...
Joshua Colp [Mon, 24 Jul 2006 15:47:59 +0000 (15:47 +0000)]
Start out with cleared RTP payload structures instead of defaults. This should prevent issues where if a stream (audio/stream) is not present and it's RTP payload structure is combined with the overall capability then the capability would be every codec that Asterisk supports.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38158 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agomove middle-of-the-block declaration at the top.
Luigi Rizzo [Mon, 24 Jul 2006 11:46:27 +0000 (11:46 +0000)]
move middle-of-the-block declaration at the top.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38149 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agofix uninitialized variable
Luigi Rizzo [Mon, 24 Jul 2006 11:45:06 +0000 (11:45 +0000)]
fix uninitialized variable

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38148 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoOnly deal with getting the supported payloads on audio if an audio RTP stream exists
Joshua Colp [Mon, 24 Jul 2006 03:42:27 +0000 (03:42 +0000)]
Only deal with getting the supported payloads on audio if an audio RTP stream exists

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38139 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoast_malloc is sufficient here, since snprintf is called on the buffer
Russell Bryant [Sun, 23 Jul 2006 15:21:44 +0000 (15:21 +0000)]
ast_malloc is sufficient here, since snprintf is called on the buffer
immediately after it gets allocated, which will automatically NULL terminate
the string

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38129 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoMerge team/russell/ast_cli_tls into the trunk.
Russell Bryant [Sun, 23 Jul 2006 15:19:16 +0000 (15:19 +0000)]
Merge team/russell/ast_cli_tls into the trunk.

This improves the performance of ast_cli() by not doing a heap memory
allocation and deallocation every single time the function is called. Instead,
a thread-specific buffer is allocatted the first time the function is called
and automatically free'd when the thread exits. Also note that this buffer will
only be allocatted in threads that actually call this function, which is
probably only the threads spawned to service connected asterisk consoles.

This does introduce a new limitation on the maximum length of the resulting
string from the arguments passed to ast_cli. Previously there was no limit
since it was just allocating a buffer big enough every time the function was
called. The current buffer size is 16kB. If there is ever a case where we want
to print more than 16k characters in a single call to ast_cli(), this will have
to be increased.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38127 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agofix a seg fault in this application if no context paramater is given
Russell Bryant [Sun, 23 Jul 2006 05:20:59 +0000 (05:20 +0000)]
fix a seg fault in this application if no context paramater is given
(issue #7571)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38118 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoresolve another XXX comment by implementing proper handling of control frames
Russell Bryant [Sun, 23 Jul 2006 05:06:47 +0000 (05:06 +0000)]
resolve another XXX comment by implementing proper handling of control frames
in ast_write(), which is to call the channel's indicate function if it exists

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38117 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agotake the advice of an XXX comment and use an atomic operation to decrement a
Russell Bryant [Sun, 23 Jul 2006 04:58:56 +0000 (04:58 +0000)]
take the advice of an XXX comment and use an atomic operation to decrement a
variable.  Also, change a use of strdup to ast_strdup and remove a duplicated
error message.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38108 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoremove a stray newline
Russell Bryant [Sun, 23 Jul 2006 03:29:19 +0000 (03:29 +0000)]
remove a stray newline

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38099 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoupdate to mention svn instead of cvs
Russell Bryant [Sun, 23 Jul 2006 02:50:25 +0000 (02:50 +0000)]
update to mention svn instead of cvs

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38089 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agovarious cleanups regarding coding guidelines issues
Russell Bryant [Sun, 23 Jul 2006 02:41:02 +0000 (02:41 +0000)]
various cleanups regarding coding guidelines issues
 - malloc to ast_malloc
 - malloc + memset to ast_calloc
 - sizeof(struct foo) to sizeof(*bar)
 - remove indentation of the entire body of a function by returning immediately
   on an allocation failure
(issue #7581, tempest1)

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38088 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoast_pbx_outgoing_cdr_failed() is not used outside of pbx.c so make it static
Russell Bryant [Sat, 22 Jul 2006 00:38:30 +0000 (00:38 +0000)]
ast_pbx_outgoing_cdr_failed() is not used outside of pbx.c so make it static

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38078 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agouse ast_atomic_fetchadd_int in chan_zap, sip, and iax2 for usecount handling
Russell Bryant [Sat, 22 Jul 2006 00:33:56 +0000 (00:33 +0000)]
use ast_atomic_fetchadd_int in chan_zap, sip, and iax2 for usecount handling
and fix a couple little things in passing
 - usecnt was not initialized in chan_iax2
 - ast_update_use_count() was not called after incrementing the count in chan_sip

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38077 65c4cc65-6c06-0410-ace0-fbb531ad65f3

14 years agoremove an XXX comment and document that ast_autoservice_start() will return -1
Russell Bryant [Sat, 22 Jul 2006 00:08:21 +0000 (00:08 +0000)]
remove an XXX comment and document that ast_autoservice_start() will return -1
if the channel is already in the autoservice list.

Why is this a valid case to return -1, you ask?  Well, there should never be
any code where it is not clear if the channel is in autoservice or not because
trying to read frames from a channel that is in the autoservice list will lead
to bad results because more than one thread will be waiting on frames to arrive
on the channel and then trying to read them.

git-svn-id: https://origsvn.digium.com/svn/asterisk/trunk@38076 65c4cc65-6c06-0410-ace0-fbb531ad65f3