[libpri-commits] lmadsen: tag 1.4.12-beta2 r1933 - in /tags/1.4.12-beta2: .version ChangeLog
SVN commits to the libpri project
libpri-commits at lists.digium.com
Mon Aug 30 15:58:29 CDT 2010
Author: lmadsen
Date: Mon Aug 30 15:58:27 2010
New Revision: 1933
URL: http://svnview.digium.com/svn/libpri?view=rev&rev=1933
Log:
Importing files for 1.4.12-beta2 release.
Added:
tags/1.4.12-beta2/.version (with props)
tags/1.4.12-beta2/ChangeLog (with props)
Added: tags/1.4.12-beta2/.version
URL: http://svnview.digium.com/svn/libpri/tags/1.4.12-beta2/.version?view=auto&rev=1933
==============================================================================
--- tags/1.4.12-beta2/.version (added)
+++ tags/1.4.12-beta2/.version Mon Aug 30 15:58:27 2010
@@ -1,0 +1,1 @@
+1.4.12-beta2
Propchange: tags/1.4.12-beta2/.version
------------------------------------------------------------------------------
svn:eol-style = native
Propchange: tags/1.4.12-beta2/.version
------------------------------------------------------------------------------
svn:keywords = none
Propchange: tags/1.4.12-beta2/.version
------------------------------------------------------------------------------
svn:mime-type = text/plain
Added: tags/1.4.12-beta2/ChangeLog
URL: http://svnview.digium.com/svn/libpri/tags/1.4.12-beta2/ChangeLog?view=auto&rev=1933
==============================================================================
--- tags/1.4.12-beta2/ChangeLog (added)
+++ tags/1.4.12-beta2/ChangeLog Mon Aug 30 15:58:27 2010
@@ -1,0 +1,1438 @@
+2010-08-30 Leif Madsen <lmadsen at digium.com>
+
+ * libpri 1.4.12-beta2 released.
+
+2010-08-30 20:49 +0000 [r1918-1929] Richard Mudgett <rmudgett at digium.com>
+
+ * q921.c: Reduced fake MDL-ERROR (Z) message severity to be visible
+ only when debugging enabled.
+
+ * q921.c, q931.c: Q.921/Q.931 message debug output improvements.
+ The Q.931 message decode debug output now will follow the correct
+ Q.921 header decode if Q.921 message dumping is enabled. Also the
+ Q.931 message decode will happen when the message actually goes
+ out on the line instead of when Q.931 passes the message to
+ Q.921. Q.921 may have to request a TEI, bring the connection up,
+ or retransmit previous frames before it can actually send the new
+ message.
+
+ * q921.c: Q.921 improvements from comparing Q.921 SDL diagrams with
+ implementation. * Handle sending and receiving DM response frames
+ as needed. * Added handling of received FRMR frames. * Completed
+ implementation of Q921_AWAITING_RELEASE state. (State is
+ currently unreachable since we have no API to initiate sending
+ the DISC message.) * Better NT PTMP TEI allocation. * Reduced
+ more ERROR level severity messages so users will stop panicking
+ when they see ERROR. This is especially true for the Q.921
+ MDL-ERROR messages. * Added better Q.921 visibility when normal
+ debug message level is enabled.
+
+ * q921.c: ISDN BRI does not recover from line faults Q.921 was
+ getting stuck in state 2 (Q921_ASSIGN_AWAITING_TEI). For some
+ reason the network was removing the TEI. Libpri then immediately
+ tried to get a new TEI assigned. The network did not reply to the
+ N202(3) attempts to get a new TEI. Libpri then just gave up
+ trying but did not leave the state. Some paths in Q.921 Figure
+ B.3 were not implemented. Q.921 now transitions to the
+ Q921_TEI_UNASSIGNED state when the N202 count is exceeded. Q.921
+ will wait there until an incoming or outgoing call is attempted.
+ * Fixed initializing the n202_counter. Not initializing the
+ n202_counter would cause the Q921_TEI_IDENTITY_REQUEST to
+ unexpectedly not go out and due to how state transitions were
+ done, Q.921 would get stuck in the Q921_ASSIGN_AWAITING_TEI
+ state. * Fixed start T202 timer fail causing Q.921 to get stuck
+ in the Q921_ASSIGN_AWAITING_TEI state if the network did not
+ respond to the request. * Fixed handling of
+ Q921_TEI_IDENTITY_REMOVE to do the MDL-REMOVE primitive
+ (q921_mdl_remove()) instead of transitioning directly to the
+ Q921_TEI_UNASSIGNED state. Necessary state clean-up was not
+ getting done. * Minor tweaks to q921_mdl_remove(). The worst
+ problem was erroneously generating an error message. * Fixed
+ potential for sending I-frames with an invalid TEI. The I-frame
+ could have been queued when Q.921 did not have an assigned TEI. *
+ Fixed testing of the q931_receive() return value when a UI-frame
+ is received. (closes issue #17570) Reported by: jcovert Patches:
+ issue17570_v1.4.11.3_v3.patch uploaded by rmudgett (license 664)
+ issue17570_v1.4_v3.patch uploaded by rmudgett (license 664)
+ Tested by: jcovert, rmudgett
+
+2010-08-27 23:37 +0000 [r1894-1912] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, q931.c: Convert most references of Q931_MAX_TEI to use
+ ARRAY_LEN(). * Minor comment correction in q931_destroycall(). *
+ Redundant logic removal q931_destroycall(). "W && X && (Y || W &&
+ Z)" is the same as "W && X && (Y || Z)"
+
+ * rose.c, rosetest.c: Add support to receive ECMA-164 2nd edition
+ OID name ROSE messages. NOTE: To add support to send the old
+ style name messages will require implementing them as new ROSE
+ operation message types. NOTE: To actually use them would likely
+ require implementing another version of the Q.SIG switch type.
+ Like (NI1 & NI2) and (4ess & 5ess) for example. Patches:
+ libpri37.patch uploaded by rmudgett (license 664) JIRA SWP-2100
+ JIRA LIBPRI-37
+
+ * pri.c: Make pri_dump_event() only print the event name.
+
+2010-08-25 17:17 +0000 [r1883-1884] Richard Mudgett <rmudgett at digium.com>
+
+ * rosetest.c: Added gripe check to rosetest for invalid operation
+ and error codes.
+
+ * q921.c: Don't silently fail to post our fake MDL-ERROR(Z).
+
+2010-08-23 22:13 +0000 [r1878] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: Add silly fake MDL error Z for kicking L2 to reactivate
+ if a DISC or DM is received in proper states on PTP links.
+
+2010-08-06 18:35 +0000 [r1852-1853] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: No audio on inbound J1 calls. Incoming calls specifying
+ the channel using a slot map could not negotiate a B channel
+ correctly. Libpri historically has handled this as an any channel
+ request. However, when chan_dahdi picked a new channel, libpri
+ sent out the recorded slot map and not the new channel selected.
+ Thus the two endpoints would be attached to different B channels
+ and the parties would not hear anything or would hear the wrong
+ parties. This patch restores the historical preference of sending
+ out the channel id using the channel number method if a channel
+ number is available. JIRA LIBPRI-35 Patches:
+ libpri-35_v1.4.11.3.patch uploaded by rmudgett (license 664)
+ libpri-35_v1.4.patch uploaded by rmudgett (license 664) Tested
+ by: rmudgett
+
+ * q921.c: Learn (AGAIN!) why state 7 and state 8 transitions were
+ suppressed.
+
+2010-08-03 23:04 +0000 [r1842-1848] Richard Mudgett <rmudgett at digium.com>
+
+ * q921.c, q931.c: Improved Q.921/Q.931 debug output. * Debug output
+ for a sent Q.931 message in TE PTMP now uses the best available
+ TEI number instead of always using 127. It could still be wrong
+ if layer 2 does not have a TEI assigned. * Q.921 debug output is
+ grouped better so a decoded message is not split by a blank line.
+ * The Q.921 state is also decoded to a name.
+
+ * q921.c: Q921_TIMER_RECOVERY SDL issue in q921_rnr_rx() Added
+ missing code specified by Q.921 (Figure B.8 Page 85) when receive
+ RNR in "Timer Recovery" state. (closes issue #16791) Reported by:
+ alecdavis Patches: rnr_timer_recovery.diff.txt uploaded by
+ alecdavis (license 585)
+
+ * q921.c: Renamed local struct pri *pri variables to struct pri
+ *ctrl in q921.c. The context tagging for my editor is much
+ happier now that the struct and the variable do not have the same
+ name. (At least for this file.)
+
+2010-07-26 Leif Madsen <lmadsen at digium.com>
+
+ * libpri 1.4.12-beta1 released
+
+2010-07-22 17:59 +0000 [r1836] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, libpri.h: Add pri_new_bri_cb() API - Create BRI D-channel
+ with user defined I/O callbacks and data There currently exists a
+ pri_new_cb() API call that allows you to create a PRI with
+ user-defined I/O read and write callbacks, and option userdata.
+ Add the same capability for BRI interfaces by adding a
+ pri_new_bri_cb() API function. (closes issue #16477) Reported by:
+ nic_bellamy Patches: pri_new_bri_cb_api.patch uploaded by nic
+ bellamy (license 299) (with minor cosmetic changes)
+
+2010-07-16 22:55 +0000 [r1828-1833] Richard Mudgett <rmudgett at digium.com>
+
+ * pritest.c, Makefile: pritest hadn't been ported to DAHDI pritest
+ hadn't been ported to DAHDI, made the trivial changes to make it
+ work. (I haven't found the equivalent to zap_playf, so changed
+ the behaviour to an echo box) (closes issue #17274) Reported by:
+ horape Patches: pritest.diff uploaded by horape (license 1035)
+ (with minor cosmetic changes)
+
+ * rosetest.c: Only need to output one version of the version
+ message.
+
+ * rosetest.c: Added a libpri version output line as a sanity check
+ for rosetest.
+
+ * rosetest.c, rose_qsig_name.c: Calling name not successfully
+ processed on inbound QSIG PRI calls from Mitel PBX The alternate
+ form of the Q.SIG Name sequence that allows manufacturer
+ extensions for CallingName, CalledName, ConnectedName, and
+ BusyName was not consuming the next ASN.1 tag. The code that
+ processed the ASN.1 Name structure was then using a stale tag
+ value. The stale tag value was then rejected with the "Did not
+ expect" message. I have added a test case using the supplied PRI
+ debug output to the rosetest utility to verify that this
+ alternate encoding is tested in the future. (closes issue #17619)
+ Reported by: jims8650 Patches: issue17619_v1.4.11.3.patch
+ uploaded by rmudgett (license 664) Tested by: rmudgett
+
+2010-06-29 16:15 +0000 [r1823] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_internal.h, q931.c: [regression] Calling Number assignment
+ logic change in libpri 1.4.11. Restored the old behaviour if
+ there is more than one calling number in the incoming SETUP
+ message. A network provided number is reported as ANI. (closes
+ issue #17495) Reported by: ibercom Patches:
+ issue_17495_v1.4.11.2.patch uploaded by rmudgett (license 664)
+ issue_17495_v1.4.patch uploaded by rmudgett (license 664) Tested
+ by: ibercom
+
+2010-06-25 20:22 +0000 [r1818] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: PRI: XXX Missing handling for mandatory IE 12 The switch
+ is sending the Connected Address ie (12) and there was no handler
+ for that ie. That is why the reporter was getting the "Missing
+ mandatory ie" message. The simple fix is to just add the missing
+ receive handler for that ie. Since connected line (COLP)
+ functionality has been added and this is just an alternate for
+ the Connected Number ie (0x4C), the handler was already written.
+ I also changed the ie name to what Q.931 calls it: Connected
+ Address. Also some minor code clean up in q931_handle_ie() and
+ ie2str(). JIRA SWP-1678 (closes issue #16915) Reported by: kobaz
+
+2010-06-04 22:45 +0000 [r1810] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_facility.c: Q.SIG calling name in FACILITY message not
+ reported to the upper layer. Q.SIG can send the CallingName,
+ CalledName, and ConnectedName in stand alone FACILITY messages.
+ If the CallingName was not sent in the SETUP message, the caller
+ id name was not reported to the upper layer. (closes issue
+ #17458) Reported by: jsmith Patches:
+ issue17458_post_qsig_name.patch uploaded by rmudgett (license
+ 664) issue17458_post_qsig_name_v1.4.11.1.patch uploaded by
+ rmudgett (license 664) Tested by: rmudgett, jsmith
+
+2010-06-01 22:59 +0000 [r1785-1794] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c: Inband disconnect setting does nothing on BRI spans. The
+ acceptinbanddisconnect flag is not inherited when creating a new
+ TEI and thus rendering the setting (and its respective equivalent
+ in Asterisk) a no-op on BRI setups. (closes issue #15265)
+ Reported by: paravoid Patches: inband-bri.diff uploaded by
+ paravoid (license 200) Tested by: paravoid
+
+ * rose.c, rosetest.c, q931.c: Multi component FACILITY messages
+ only process the first component. The code was only processing
+ the first ROSE component in the facility message. I also updated
+ rosetest.c to have a multiple component ROSE message test.
+ (closes issue #17428) Reported by: patrol-cz Patches:
+ issue17428.patch uploaded by rmudgett (license 664) Tested by:
+ rmudgett, patrol-cz
+
+2010-05-28 22:34 +0000 [r1776] David Vossel <dvossel at digium.com>
+
+ * pri_internal.h, pri_aoc.c, q931.c, pri_facility.c,
+ pri_facility.h, libpri.h: support for sending ETSI advice of
+ charge Review: https://reviewboard.asterisk.org/r/619/
+
+2010-05-28 21:50 +0000 [r1723-1767] Richard Mudgett <rmudgett at digium.com>
+
+ * rose_etsi_mwi.c (added), rose.c, rosetest.c, rose_internal.h,
+ Makefile, rose.h, pri_facility.c, libpri.h, rose_qsig_mwi.c: ETSI
+ Message Waiting Indication (MWI) support. Add the ability to
+ report waiting messages to ISDN endpoints (phones). Relevant
+ specification: EN 300 650 and EN 300 745 Review:
+ https://reviewboard.asterisk.org/r/600/
+
+ * rose.c, rosetest.c, pri_internal.h, rose.h, pri_facility.c,
+ libpri.h: ETSI Malicious Call ID support. Add the ability to
+ report malicious callers. Relevant specification: EN 300 180
+ Review: https://reviewboard.asterisk.org/r/575/
+
+ * pri.c, pri_internal.h, pri_q931.h, q931.c, libpri.h: ETSI Call
+ Waiting support. Add the ability to announce a call to an
+ endpoint when there are no B channels available. A call waiting
+ call is a SETUP message with no B channel selected. Relevant
+ specification: EN 300 056, EN 300 057, EN 300 058 Review:
+ https://reviewboard.asterisk.org/r/569/
+
+ * pri_internal.h, pri_aoc.c (added), Makefile, pri_facility.c,
+ rose_etsi_aoc.c, pri_facility.h, libpri.h: ETSI Advice Of Charge
+ (AOC) event reporting. This feature passes ETSI AOC-S, AOC-D, and
+ AOC-E message information to the upper layer (i.e. Asterisk) for
+ processing. Relevant specification: EN 300 182 Consideration was
+ made for the possible future addition of Q.SIG AOC support
+ (ECMA-212) with the events passed to the upper layer. Review:
+ https://reviewboard.asterisk.org/r/538/
+
+ * pri.c, pri_internal.h, q931.c, pri_facility.c, pri_facility.h,
+ libpri.h: Added ETSI Explicit Call Transfer (ECT) support. Added
+ ability to send and receive ETSI ECT messages to eliminate
+ tromboned calls. Added ETSI support to an existing API call to
+ send ECT messages when the upper level indicates to transfer
+ specified calls. The libpri API was extended to indicate to the
+ upper layer that the far end requests the transfer of the
+ indicated calls. Review: https://reviewboard.asterisk.org/r/521/
+
+2010-05-26 16:01 +0000 [r1714] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, doc/cc_ptmp_monitor_flattened.fsm, pri_internal.h,
+ Makefile, q931.c, doc/cc_ptp_agent.fsm, doc/cc_qsig_agent.fsm,
+ pri_cc.c (added), rose_qsig_cc.c (added), prisched.c,
+ doc/cc_ptmp_agent_flattened.fsm, rose_etsi_cc.c (added),
+ doc/cc_ptmp_monitor.fsm, rosetest.c, rose.c, rose_internal.h,
+ doc/cc_ptmp_agent.fsm, pri_q931.h,
+ doc/cc_qsig_monitor_flattened.fsm,
+ doc/cc_ptp_monitor_flattened.fsm, rose.h, pri_facility.c,
+ pri_facility.h, doc/cc_qsig_agent_flattened.fsm,
+ doc/cc_ptp_agent_flattened.fsm, doc (added), libpri.h,
+ doc/cc_qsig_monitor.fsm, doc/cc_ptp_monitor.fsm: Add Call
+ Completion Suppplementary Service Call Completion Supplementary
+ Service (CCSS) added for the following switch types: ETSI PTMP,
+ ETSI PTP, Q.SIG. Specifications: ETS 300 359 CCBS for PTMP and
+ PTP ETS 301 065 CCNR for PTMP and PTP ECMA-186 Call Completion
+ for Q.SIG Several support services were added to support CC:
+ Dummy Call Reference. Q.931 REGISTER message. Dynamic expansion
+ of the number of available timers (up to 8192). Enhanced facility
+ message handling. Current implementation limitations preclude the
+ following: CC service retention is not supported. Q.SIG path
+ reservation is not supported. (closes issue #14292) Reported by:
+ tomaso Tested by: rmudgett JIRA SWP-1493 Review:
+ https://reviewboard.asterisk.org/r/522/
+
+2010-06-29 Leif Madsen <lmadsen at digium.com>
+
+ * libpri 1.4.11.3 released.
+
+2010-06-29 11:17 +0000 [r1824] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_internal.h, q931.c: [regression] Calling Number assignment
+ logic change in libpri 1.4.11. Restored the old behaviour if
+ there is more than one calling number in the incoming SETUP
+ message. A network provided number is reported as ANI. (closes
+ issue #17495) Reported by: ibercom Patches:
+ issue_17495_v1.4.11.2.patch uploaded by rmudgett (license 664)
+ issue_17495_v1.4.patch uploaded by rmudgett (license 664) Tested
+ by: ibercom
+
+2010-06-02 Leif Madsen <lmadsen at digium.com>
+
+ * libpri 1.4.11.2 released.
+
+2010-06-04 17:45 +0000 [r1810] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_facility.c: Q.SIG calling name in FACILITY message not
+ reported to the upper layer. Q.SIG can send the CallingName,
+ CalledName, and ConnectedName in stand alone FACILITY messages.
+ If the CallingName was not sent in the SETUP message, the caller
+ id name was not reported to the upper layer. (closes issue #17458)
+ Reported by: jsmith Patches: issue17458_post_qsig_name.patch uploaded
+ by rmudgett (license 664) issue17458_post_qsig_name_v1.4.11.1.patch
+ uploaded by rmudgett (license 664) Tested by: rmudgett, jsmith
+
+2010-06-02 Leif Madsen <lmadsen at digium.com>
+
+ * libpri 1.4.11.1 released.
+
+2010-06-02 11:26 +0000 [r1805-1806] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c: Inband disconnect setting does nothing on BRI spans.
+ The acceptinbanddisconnect flag is not inherited when creating
+ a new TEI and thus rendering the setting (and its respective
+ equivalent in Asterisk) a no-op on BRI setups. (closes issue
+ #15265) Reported by: paravoid Patches: inband-bri.diff uploaded
+ by paravoid (license 200) Tested by: paravoid
+
+ * q931.c, rose.c, rosetest.c: Multi component FACILITY messages
+ only process the first component. The code was only processing
+ the first ROSE component in the facility message. I also updated
+ rosetest.c to have a multiple component ROSE message test.
+ (closes issue #17428) Reported by: patrol-cz Patches:
+ issue17428.patch uploaded by rmudgett (license 664) Tested by:
+ rmudgett, patrol-cz
+
+2010-05-20 Russell Bryant <russell at digium.com>
+
+ * libpri 1.4.11 released.
+
+2010-05-19 21:50 +0000 [r1703] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: T309 should not do anything with the global call
+ reference call record.
+
+2010-05-19 21:30 +0000 [r1702] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: It's amazing what a tiny bug in the Q.921 SDL diagram can
+ do to cause trouble.... Fix issue where V_R was not reset and N_R
+ was consequentially transmitted incorrectly. Particularly in
+ layer 2 initiated re-establishments.
+
+2010-05-11 22:14 +0000 [r1688] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h, q931.c, libpri.h: Dialing an invalid
+ extension causes incomplete hangup sequence. Revision -r1489
+ corrected a deviation from Q.931 Section 5.3.2. However, this
+ resulted in an unexpected behaviour change to the upper layer
+ (Asterisk). This change restores the legacy hangup behaviour if
+ the new API call is not used. Use pri_hangup_fix_enable() to
+ follow Q.931 Section 5.3.2 call hangup better. (closes issue
+ #17104) Reported by: shawkris Tested by: rmudgett
+
+2010-04-26 19:54 +0000 [r1664-1675] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Simplified some protocol discriminator handling code.
+
+ * q931.c: Garbage on the end of Q.931 messages causing calls to
+ fail to connect. The DAHDI driver had a bug where an extra byte
+ appeared on the end of Q.931 messages. This garbage byte caused
+ the message to be discarded with the diagnostic "XXX Message
+ longer than it should be?? XXX". The Q.931 message will no longer
+ be discarded if there were earlier ie's in the message. This
+ patch also addresses the potential problem of reading beyond the
+ buffer when trying to parse the garbage data. Thanks to roeften
+ for the base patch. (closes issue #14378) Reported by: timking
+
+ * q921.c: Avoid using a cast.
+
+ * q931.c: Cleanup some pri debug output line presentation.
+
+2010-04-19 22:40 +0000 [r1625-1630] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_internal.h, pri_q931.h, q931.c, pridump.c: PTMP NT mode call
+ reference value ambiguity. Since the TE side can pick CR values
+ independently, the TE CR needs to be qualified by TEI to
+ distinguish CR values from other devices. Without doing this,
+ multiple phones on the BRI line will have intermittent call
+ failures. JIRA LIBPRI-30 Also eliminated some wierdness in
+ q931_status() and several places where it is called.
+
+ * q921.c: Fix potential crash when pridump.c calls q921_dump() with
+ NULL pri ptr.
+
+2010-04-15 18:43 +0000 [r1596] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h: Make some internal routines available to
+ other libpri components.
+
+2010-04-09 21:43 +0000 [r1577] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Only one PROCEEDING message per call please.
+
+2010-03-18 15:50 +0000 [r1534-1547] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Move a comment closer to where it has relevance.
+
+ * pri.c, pri_internal.h, q921.c, q931.c, pri_facility.c:
+ Miscellaneous simple reorganization. 1) Make PRI_MASTER() no
+ longer check for a NULL parameter. It is the caller's
+ responsibility. Not many callers could have passed a NULL without
+ crashing before or after anyway. 2) Replace calls to
+ q931_is_ptmp() with PTMP_MODE(). They were equivalent. 3) Made
+ the following boolean config options bit fields: sendfacility,
+ overlapdial, chan_mapping_logical, and service_message_support.
+
+2010-03-02 23:47 +0000 [r1511] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h, q921.c: Restore ability to change the
+ Q.921 K value. The Q.921 rewrite only used value of PRI_TIMER_K
+ right after it was set to the default. The Q.921 window size was
+ thus no longer alterable by the user. (closes issue #16909)
+ Reported by: alecdavis Patches: pritimer.libpri.diff.txt uploaded
+ by alecdavis (license 585) Tested by: alecdavis
+
+2010-02-11 21:47 +0000 [r1488-1489] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Must send DISCONNECT if we have sent a response to a
+ SETUP message. Q.931 Section 5.3.2 a) says we send a
+ RELEASE_COMPLETE to reject a call SETUP if we have not already
+ sent a message in response to the SETUP message.
+
+ * pri.c, libpri.h: Minor comment changes.
+
+2010-02-11 17:35 +0000 [r1482] Wendell Thompson <wthompson at digium.com>
+
+ * Makefile: Added CPUARCH option for selecting a 32-bit build from
+ the command line.
+
+2010-02-08 23:29 +0000 [r1470-1476] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: Revert useless check of pri->t200_timer value, since
+ scheduler routines check the value anyways.
+
+ * q921.c: Make sure we set the l3initiated flag when PTP links are
+ attempted to be re-established
+
+2010-02-05 23:34 +0000 [r1464] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c: pri_restart() is no longer needed since the Q.921 rewrite.
+ Don't output error message for a deprecated function.
+
+2010-01-29 21:55 +0000 [r1457] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: Sense of statement was inverted from what it should have
+ been. Might have caused false T200 operation on reception of
+ I-frames.
+
+2010-01-29 19:32 +0000 [r1451] Richard Mudgett <rmudgett at digium.com>
+
+ * q921.c: Only set eres if there actually is an event to pass up.
+ (issue 16713)
+
+2010-01-29 17:27 +0000 [r1445] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: Fix bug in which an event was lost if an I-frame was
+ received during a timer recovery state (related to #16713)
+
+2010-01-26 21:04 +0000 [r1439] Richard Mudgett <rmudgett at digium.com>
+
+ * q921.c: Don't be so noisy when D channel is down.
+
+2010-01-19 21:53 +0000 [r1426] Richard Mudgett <rmudgett at digium.com>
+
+ * q921.c, q931.c: Fix TE PTMP side sending FACILITY messages on the
+ dummy call reference. Only the NT PTMP side can send Q.931
+ broadcast messages. Also removed an inaccurate comment in Q.921
+ and made q921_mdl_handle_error_callback() call the correct struct
+ pri free function.
+
+2010-01-15 18:28 +0000 [r1414] Richard Mudgett <rmudgett at digium.com>
+
+ * q921.c, q931.c: Make some debugging messages conditional and some
+ minor reformating changes.
+
+2010-01-13 19:37 +0000 [r1406] Matthew Fredrickson <creslin at digium.com>
+
+ * pri.c, pri_internal.h, pri_q921.h, Makefile, pri_q931.h, q921.c,
+ q931.c: Merge of Q.921 rewrite branch for wider testing.
+
+2009-12-09 20:59 +0000 [r1374] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Extracted q931_fill_ring_event() from
+ post_handle_q931_message(). Done so it is easier to see what was
+ done in ccbs branch.
+
+2009-11-21 02:40 +0000 [r1345-1351] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h: Fix debug output so built up output lines
+ are readable again. A recent change to Asterisk put the span
+ number at the begining of each line. This is a good thing if you
+ need to debug multiple spans or forget which span you are
+ debugging. Unfortunately, any pri_message() output that is not a
+ complete line is messed up. The pri_message() function now will
+ accumulate line output until a '\n' is seen on the end.
+
+ * pri_internal.h, q931.c, pri_facility.c: Delay processing of
+ facility ie's after all other ie's are processed. * Some ROSE
+ message processing depends on the presence of other ies. The
+ DivertingLegInformation1, and 3 messages will be used as the
+ default connected line number if the connected number ie is not
+ present. The redirecting number ie is used as a default to the
+ redirecting number in the DivertingLegInformation2 message if the
+ ROSE message does not contain it and the redirecting number ie is
+ present. * Some ROSE message processing depends upon other ie
+ values. The StatusRequest, CCBS-T-Call, and CcRingout messages
+ collectively need the BC, HLC, LLC, called number, called
+ subaddress, calling number, and calling subaddress ie information
+ to be available.
+
+2009-11-18 00:36 +0000 [r1331] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Merged revision 1328 from
+ https://origsvn.digium.com/svn/libpri/team/mattf/libpri-1.4-q921-rewrite
+ .......... r1328 | mattf | 2009-11-17 15:16:11 -0600 (Tue, 17 Nov
+ 2009) | 1 line outboundbroadcast isn't set at this time, since it
+ is set after the message is transmited, so we must use other
+ criteria to determine the need for broadcast on a setup
+ ..........
+
+2009-11-14 00:20 +0000 [r1310-1322] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h, q931.c, pri_facility.c, pri_facility.h,
+ libpri.h: Reimplement callback mechanism to handle APDU response
+ messages that we care about. 1) No sent messages will remain in
+ the APDU queue unless they have an active timer to remove them.
+ The dummy call reference call and global call reference call
+ structures will not act like a memory leak to sent messages. 2)
+ The new T-RESPONSE timer will be the generic response guard if
+ the standards do not otherwise specify a timer for a message
+ response. 3) The callback will be called. If it is called because
+ of a response message, then the callback has an opportunity to
+ indicate if more responses are expected.
+
+ * libpri.h: We now have 32 timers. No need to reserve minimum space
+ anymore.
+
+ * pri_internal.h, pri_facility.c: There must be only one source for
+ the invoke id values per D channel group. If there are
+ potentially multiple sources for the invoke id sequence then we
+ could get confused if there are multiple outstanding messages
+ with the same invoke id that get responses.
+
+2009-11-11 00:22 +0000 [r1291] Matthew Fredrickson <creslin at digium.com>
+
+ * pri_internal.h, q921.c: Make sure we also revive links for PRIs,
+ not just PTMP TE BRIs when we get a disconnect message
+
+2009-11-10 21:51 +0000 [r1283] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_internal.h, q931.c, pri_facility.c, pri_facility.h: The
+ facility ie queue needs to remove facilities that have been sent.
+ The facility ie queue needs to remove facilities that have been
+ sent. Otherwise, the queue just grows until the call is
+ terminated. AOC messages can clog the queue during a long call
+ and the dummy call reference may never be deleted. Also removed
+ unneeded elements of struct apdu_event. The callback function was
+ not a good idea since many facility messages do not have
+ responses and the callback would prevents removal of events from
+ the list.
+
+2009-11-10 20:25 +0000 [r1276] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: Re-add back in support for TE initiated layer 2
+ activation
+
+2009-11-10 19:27 +0000 [r1268-1275] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h, q931.c, libpri.h: Add dummy call reference
+ support. Fixes problem where PTMP NT mode responds erroneously to
+ a FACILITY message from a phone on the dummy call reference.
+ LibPRI behaved as if the dummy call reference were an invalid
+ call reference and proceeded to respond on the global call
+ reference.
+
+ * pri_internal.h: Remove unused callingsubaddr[].
+
+2009-11-03 17:19 +0000 [r1255-1261] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_q931.h, q931.c: Unknown IE 50 (cs5, Unknown Information
+ Element) Add code to recognize the code set 5 ie 50 (calling
+ party category) to suppress the unknown IE message. (closes issue
+ #13828) Reported by: fdecher Patches:
+ libpri_ie50_cs5-trunk.diff3.txt uploaded by alecdavis (license
+ 585) Tested by: alecdavis
+
+ * q931.c: NT PTMP did not report busy when calling a busy phone.
+ The caller would not get a busy indication when calling a busy
+ phone. Timer T303 is not supposed to be stopped when
+ RELEASE_COMPLETE received. When T303 expires we will now report
+ the last clearing cause code to the caller if we received one.
+
+2009-10-23 23:47 +0000 [r1242-1249] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_internal.h, pri_facility.c: Add subaddress handling to
+ existing messages and functions. Connected line updates when
+ transfering calls does not completely support subaddresses yet.
+
+ * pri.c, pri_internal.h, pri_q921.h, pri_q931.h, q921.c, q931.c,
+ pri_facility.c, pri_facility.h, libpri.h: Add BRI PTMP NT mode,
+ HOLD/RETRIEVE, Call rerouting/deflection, and keypad facility
+ support. * Added support for BRI PTMP NT mode. (Overlap dialing
+ NT -> TE not supported.) * Added handling of received
+ HOLD/RETRIEVE messages and the optional ability to transfer a
+ held call on disconnect similar to an analog phone. * Added
+ CallRerouting/CallDeflection support for Q.SIG, ETSI PTP, ETSI
+ PTMP. Will reroute/deflect an outgoing call when receive the
+ message. Can use the DAHDISendCallreroutingFacility to send the
+ message for the supported switches. * Added ability to
+ send/receive keypad digits in the SETUP message. Send keypad
+ digits in SETUP message:
+ Dial(DAHDI/g1[/K<keypad_digits>][/extension]) Access any received
+ keypad digits in SETUP message by: ${CHANNEL(keypad_digits)}
+ (closes issue #15048) Tested by: rmudgett, mattf
+
+2009-10-22 16:16 +0000 [r1230] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h, pri_q931.h, q931.c, libpri.h: Add support
+ for calling and called subaddress. Partial support for COLP
+ subaddress. The Telecom Specs in NZ suggests that SUB ADDRESS is
+ always on, so doing "desk to desk" between offices each with an
+ asterisk box over the ISDN should then be possible, without a
+ whole load of DDI numbers required. (closes issue #15604)
+ Reported by: alecdavis Patches: libpri_subaddr_trunk.diff11.txt
+ uploaded by alecdavis (license 585) Some minor modificatons were
+ made. Tested by: alecdavis, rmudgett Review:
+ https://reviewboard.asterisk.org/r/406/
+
+2009-10-19 22:49 +0000 [r1219-1220] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Fix call state ie transmission. Sending a STATUS message
+ failed to include the call state ie for some reason. We will now
+ always send a call state ie when a message ie list includes one.
+
+ * q931.c: Fix comparision of invalid party name and number structs
+ in comparison functions.
+
+2009-10-15 22:34 +0000 [r1212] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_facility.c: Take diverted-to-number from DivLegInfo1 and use
+ it as connected number. If no connected number is signaled in the
+ CONNECT message we will use the last diverted to number. (issue
+ #14292) Reported by: tomaso Patches:
+ divleginfo1_to_connectednum.patch uploaded by tomaso (license
+ 564) (Used as a guide since it no longer will apply.) (This patch
+ is unrelated to the issue.)
+
+2009-10-14 19:03 +0000 [r1191-1205] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: In q931_getcall(): Simplify test and add related switch
+ types.
+
+ * q931.c: Reduce future conflicts when adding ie's to the SETUP
+ message.
+
+2009-10-12 17:17 +0000 [r1177] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c, pri_internal.h, pri_q931.h, q931.c, pri_facility.c:
+ Miscellaneous changes: * Removed unnecessary
+ Q931_IE_CONNECTED_NUM ie from setup_ack_ies[]. * Added internal
+ state Q931_CALL_STATE_NOT_SET to Q.931 state enum. * Made
+ q931_is_ptmp() take a const pointer. * pri_facility.c: Some
+ preparations for subaddressing. * pri.c: Eliminate use of a magic
+ number.
+
+2009-10-09 23:20 +0000 [r1169] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Conditional out noisy and redundant ASN.1 parse dump of
+ facility ie contents. 1) Outgoing messages have the facility ie
+ ASN.1 decoded and dumped when the ie is added to the message. The
+ whole message is then dumped. 2) Incoming messages have the
+ facility ie ASN.1 decoded and dumped when the ie is processed.
+ The whole message has already been dumped.
+
+2009-10-07 18:34 +0000 [r1151-1152] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_facility.c: Sent Q.SIG call rerouting message fixes. a)
+ Account for pSS1InfoElement where the bearer capability for the
+ new rerouted call is described. The call could be a fax or data
+ call, let's use the appropriate signaling to avoid call rejects
+ on the other end. b) Handle calling number appropriately, number
+ can be prohibited or non-existent. c) Add calling name if
+ available. d) Use the diversion counter from DivLegInfo2 (if was
+ present in the incoming Setup) and increment. (issue #14292)
+ Reported by: tomaso Patches:
+ libpri-1.4-2009-01-29-rerouting-0.1.9.patch uploaded by tomaso
+ (license 564) (Used as a guide since it no longer will apply.)
+ (This patch is unrelated to the issue.)
+
+ * pri.c, pri_q921.h, q921.c: Merged revisions 1143, 1144 from
+ https://origsvn.digium.com/svn/libpri/tags/1.4.10.2 ..........
+ r1144 | mattf | 2009-09-29 10:32:23 -0500 (Tue, 29 Sep 2009) | 1
+ line This fix is more like the fix that was used to resolve the
+ issue for the PRI case .......... r1143 | mattf | 2009-09-28
+ 14:07:01 -0500 (Mon, 28 Sep 2009) | 1 line Changes for 1.4.10.2
+ ..........
+
+2009-09-22 17:12 +0000 [r1120] Jeff Peeler <jpeeler at digium.com>
+
+ * q931.c: Fix call reference to be associated with the D channel
+ message was received The problem is that once a call reference
+ was associated with a particular D channel, it always was. This
+ created an issue with NFAS when the secondary D channel became
+ active as the messages were still being sent on the non-active D
+ channel. (closes issue #14959) Reported by: remiq Patches:
+ bug14959.patch uploaded by jpeeler (license 325) Tested by: remiq
+
+2009-09-22 02:23 +0000 [r1107] Richard Mudgett <rmudgett at digium.com>
+
+ * pri.c: Update pri_event2str() to current defined events.
+
+2009-09-18 00:31 +0000 [r1097] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Another place where timeout events with subcmds should
+ clear any old subcmds.
+
+2009-09-15 22:24 +0000 [r1084] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Timeout events with subcmds should clear any old subcmds.
+
+2009-09-14 22:32 +0000 [r1077] Matthew Fredrickson <creslin at digium.com>
+
+ * q921.c: Output multiline output with multiple calls to
+ pri_message, so that logs look normal in Asterisk
+
+2009-09-13 22:54 +0000 [r1072] Dwayne M. Hubbard <dwayne.hubbard at gmail.com>
+
+ * pri_q931.h, q931.c: Add SERVICE message support for the
+ 'national' switchtype This set of changes integrates SERVICE
+ message support for the 'national' switchtype. The 'national'
+ switchtype uses the 0x43 protocol discriminator. The 'national'
+ SERVICE/SERVICE ACKNOWLEDGE and AT&T SERVICE/SERVICE ACKNOWLEDGE
+ message values are opposite of each other. This is handled by
+ first determining which protocol discriminator is in use, then
+ responding with the appropriate SERVICE ACKNOWLEDGE value. AT&T
+ SERVICE messages use the 0x3 protocol discriminator. (closes
+ issue #15803) Reported by: dhubbard Review:
+ https://reviewboard.asterisk.org/r/347/
+
+2009-09-02 20:19 +0000 [r1059-1061] Richard Mudgett <rmudgett at digium.com>
+
+ * pri_internal.h, pri_q931.h, q931.c: Converted Q931_CALL_STATE_xxx
+ defines to an enum.
+
+ * q921.c, q931.c: Optimized calls to pri_schedule_del(). There is
+ no need to test if a scheduled event is running before calling
+ pri_schedule_del().
+
+ * prisched.c: Cleaned up scheduled events handling code. * Fixed
+ pri_schedule_event() to return 0 on error instead of -1. Zero is
+ a safer value to return. Users would not think that a timer was
+ scheduled. * Fixed potential for pri_schedule_del() to write out
+ of bounds of pri_sched[]. The out of bounds access could occur
+ when pri_schedule_event() returned -1. * Made use all pri_sched[]
+ entries. pri_sched[0] was previously unused. * Removed some
+ unneeded code and recursion since scheduling only runs on master
+ D channel structures. * Added doxygen comments. * Renamed struct
+ pri *pri variables to struct pri *ctrl in this file.
+
+2009-08-31 22:57 +0000 [r1042-1051] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c, libpri.h: Make event channel parameter encoding
+ consistent. Also make sure that service maintenance messages have
+ the channel id parameters reinitialized for each message since
+ they are sent over the global call reference.
+
+ * pri.c, pri_internal.h, q931.c, libpri.h: Split justsignalling
+ into cis_call and cis_auto_disconnect functionality.
+
+2009-08-26 15:24 +0000 [r1006-1028] Richard Mudgett <rmudgett at digium.com>
+
+ * q931.c: Fix BRI PTP broken by -r790.
+
+ * q931.c: Put connected name in display ie for CONNECT message.
+
+ * q931.c: Fix ie ordering in some ie lists for send_message().
+
+ * q931.c: Make dump_channel_id() handle variable length fields.
+ Also did some other minor miscellaneous changes.
+
[... 652 lines stripped ...]
More information about the libpri-commits
mailing list