diff options
author | russell <russell@f38db490-d61c-443f-a65b-d21fe96a405b> | 2005-06-29 21:02:35 +0000 |
---|---|---|
committer | russell <russell@f38db490-d61c-443f-a65b-d21fe96a405b> | 2005-06-29 21:02:35 +0000 |
commit | cb75e9881004441889f2ba69065b637d308cdfa0 (patch) | |
tree | d95501498835b668ee3187f53aba3f4fb71f09d9 | |
parent | 2b24a0e7168c3a1b204ed2d47efcbbc04e545e12 (diff) |
fix callerid matching in extensions.conf
formatting fixes for the ChangeLog
git-svn-id: http://svn.digium.com/svn/asterisk/branches/v1-0@6014 f38db490-d61c-443f-a65b-d21fe96a405b
-rwxr-xr-x | CHANGES | 176 | ||||
-rwxr-xr-x | pbx/pbx_config.c | 13 |
2 files changed, 100 insertions, 89 deletions
@@ -1,48 +1,50 @@ - NOTE: Corrections or additions to the ChangeLog may be submitted - to http://bugs.digium.com. Documentation and formatting - fixes are not listed here. A complete listing of changes - is available through the Asterisk-CVS mailing list hosted - at http://lists.digium.com. + NOTE: Corrections or additions to the ChangeLog may be submitted to + http://bugs.digium.com. Documentation and formatting fixes are not + not listed here. A complete listing of changes is available through + the Asterisk-CVS mailing list hosted at http://lists.digium.com. - -- chan_mgcp - -- *70 is used to disable call waiting. Call waiting will now be re-enabled - on hangup. +Asterisk 1.0.9 + + -- fix bug in callerid matching in the dialplan that was introduced in 1.0.8 Asterisk 1.0.8 -- chan_zap - -- Asterisk will now also look in the regular context for the fax extension while - executing a macro. Previously, for this to work, the fax extension would have - to be included in the macro definition. + -- Asterisk will now also look in the regular context for the fax extension + while executing a macro. Previously, for this to work, the fax extension + would have to be included in the macro definition. -- On some systems, ALERTING will be sent after PROCEEDING, so code has been added to account for this case. - -- If no extension is specified on an overlap call, the 's' extension will be used. - -- Add support for feautres of 2nd gen hardware + -- If no extension is specified on an overlap call, the 's' extension will + be used. -- chan_sip - -- We no longer send a "to" tag on "100 Trying" messages, as it is inappropriate - to do so. - -- We now respond correctly to an invite for T.38 with a "488 Not acceptable here" - -- We now discard saved tags on 401/407 responses in case the provider we're talking - to tries to pull a dirty trick on us and change it. - -- rtptimeout options will now be correctly set on a peer basis rather than only global + -- We no longer send a "to" tag on "100 Trying" messages, as it is + inappropriate to do so. + -- We now respond correctly to an invite for T.38 with a "488 Not acceptable + here" + -- We now discard saved tags on 401/407 responses in case the provider we're + talking to tries to pull a dirty trick on us and change it. + -- rtptimeout options will now be correctly set on a peer basis rather than + only global -- chan_mgcp -- Fixed setting of accountcode -- Fixed where *67 to block callerid only worked for first call -- chan_agent - -- We now will not pass audio until the agent has acked the call if the configuration + -- We now will not pass audio until the agent has acked the call if the + configuration is set up for the agent to do so. -- chan_alsa -- Fixed problems with the unloading of this module -- res_agi - -- A fix has been added to prevent calls from being hung up when more than one - call is executing an AGI script calling the GET DATA command. - -- AGI scripts will now continue to run even if a file was not found with the - GET DATA command. - -- When calling SAY NUMBER with a number like 09, we will now say "nine" instead - of "zero" + -- A fix has been added to prevent calls from being hung up when more than + one call is executing an AGI script calling the GET DATA command. + -- AGI scripts will now continue to run even if a file was not found with + the GET DATA command. + -- When calling SAY NUMBER with a number like 09, we will now say "nine" + instead of "zero" -- app_dial - -- There was a problem where text frames would not be forwarded before the channel - has been answered. + -- There was a problem where text frames would not be forwarded before the + channel has been answered. -- app_disa -- Fixed the timeout used when no password is set -- app_queue @@ -52,15 +54,17 @@ Asterisk 1.0.8 -- say.c -- A problem has been fixed with saying the date in Spanish. -- Makefile - -- A line was missing for the autosupport script that caused "make rpm" to fail + -- A line was missing for the autosupport script that caused "make rpm" to + fail -- format_wav_gsm - -- Fixed a problem with wav formatting that prevented files from being played - in some media players + -- Fixed a problem with wav formatting that prevented files from being + played in some media players -- pbx_spool - -- Fixed if the last line of text in a file for the call spool did not contain - a new line, it would not be processed + -- Fixed if the last line of text in a file for the call spool did not + contain a new line, it would not be processed -- logger - -- Fixed the logger so that color escape sequences wouldn't be sent to the logs + -- Fixed the logger so that color escape sequences wouldn't be sent to the + logs -- format_sln -- A lot of changes were made to correctly handle signed linear format on big endian machines @@ -70,79 +74,91 @@ Asterisk 1.0.8 Asterisk 1.0.7 -- chan_sip - -- The fix for some codec availibility issues in 1.0.6 caused music on hold problems, - but has now been fixed. + -- The fix for some codec availibility issues in 1.0.6 caused music on hold + problems, but has now been fixed. -- chan_skinny -- A check has been added to avoid a crash. -- chan_iax2 - -- A feature has been added to CVS head to have the option of sending timestamps with - trunk frames. It is not supported in 1.0, but a change has been made so that it - will at least not choke if sent trunk timestamps. + -- A feature has been added to CVS head to have the option of sending + timestamps with trunk frames. It is not supported in 1.0, but a change + has been made so that it will at least not choke if sent trunk + timestamps. -- app_voicemail -- Some checks have been added to avoid a crash. -- speex - -- The path /usr/include/speex has been added for a place to look for the speex header. + -- The path /usr/include/speex has been added for a place to look for the + speex header. Asterisk 1.0.6 -- chan_iax2: -- Fixed a bug dealing with a division by zero that could cause a crash -- chan_sip: - -- Behavior was changed so that when a registration fails due to DNS resolution issues, - a retry will be attempted in 20 seconds. - -- Peer settings were not reset to null values when reloading the configuration file. - Behavior has been changed so that these values are now cleared. + -- Behavior was changed so that when a registration fails due to DNS + resolution issues, a retry will be attempted in 20 seconds. + -- Peer settings were not reset to null values when reloading the + configuration file. Behavior has been changed so that these values are + now cleared. -- 'restrictcid' now properly works on MySQL peers. -- Only use the default callerid if it has been specified. - -- Asterisk was not sending the same From: line in SIP messages during certain times. - Fixed to make sure it stays the same. This makes some providers happier, to a working state. - -- Certain circumstances involving a blank callerid caused asterisk to segmentation fault. - -- There was a problem incorrectly matching codec availablity when global preferences were - different from that of the user. To fix this, processing of SDP data has been moved - to after determining who the call is coming from. - -- Asterisk would run out of RTP ports while waiting for SUBSCRIBE's to expire even though - an RTP port isn't needed in this case. This has been fixed by releasing the ports early. + -- Asterisk was not sending the same From: line in SIP messages during + certain times. Fixed to make sure it stays the same. This makes some + providers happier, to a working state. + -- Certain circumstances involving a blank callerid caused asterisk to + segmentation fault. + -- There was a problem incorrectly matching codec availablity when global + preferences were different from that of the user. To fix this, + processing of SDP data has been moved to after determining who the call + is coming from. + -- Asterisk would run out of RTP ports while waiting for SUBSCRIBE's to + expire even though an RTP port isn't needed in this case. This has been + fixed by releasing the ports early. -- chan_zap: - -- During a certain scenario when using flash and '#' transfers you would hear the - other person and the music they were hearing. This has been fixed. + -- During a certain scenario when using flash and '#' transfers you would + hear the other person and the music they were hearing. This has been + fixed. -- A fix for a compilation issue with gcc4 was added. -- chan_modem_bestdata: -- A fix for a compilation issue with gcc4 was added. -- format_g729: - -- Treat a 10-byte read as an end of file indication instead of an error. Some G729 encoders - like to put 10-bytes at the end to indicate this. + -- Treat a 10-byte read as an end of file indication instead of an error. + Some G729 encoders like to put 10-bytes at the end to indicate this. -- res_features: - -- During certain situations when parking a call, both endpoints would get musiconhold. - This has been fixed so the individual who parked the call will hear the digits and not - musiconhold. + -- During certain situations when parking a call, both endpoints would get + musiconhold. This has been fixed so the individual who parked the call + will hear the digits and not musiconhold. -- app_dial: - -- DIALEDPEERNUMBER is now being set, so if you attempted to use it in the past and failed - it should work now. - -- A callerid change caused many headaches, this has been reversed to the original 1.0 behavior. - -- A crash caused with the combination of the 'g' option and # transfer was fixed. + -- DIALEDPEERNUMBER is now being set, so if you attempted to use it in the + past and failed, it should work now. + -- A callerid change caused many headaches, this has been reversed to the + original 1.0 behavior. + -- A crash caused with the combination of the 'g' option and # transfer was + fixed. -- app_voicemail: - -- If two people hit the voicemail system at the same time, and were leaving a message - the second message was overwriting the first. This has been fixed so that each one - is distinct and will not overwrite eachother. + -- If two people hit the voicemail system at the same time, and were leaving + a message the second message was overwriting the first. This has been + fixed so that each one is distinct and will not overwrite eachother. -- cdr_tds: - -- If the server you were using was going down, it had the potential to bring your asterisk - server down with it. Extra stuff has been added so as to bring in more error/connection - checking. + -- If the server you were using was going down, it had the potential to + bring your asterisk server down with it. Extra stuff has been added so + as to bring in more error/connection checking. -- cdr_pgsql: -- This will now attempt to reconnect after a connection problem. -- IAXY firmware: - -- This has been updated to version 23. It includes a fix for lost registrations. + -- This has been updated to version 23. It includes a fix for lost + registrations. -- internals - -- Behavior was changed for 'show codec <number>' to make it more intuitive. (kshumard) - -- DNS failures and asterisk do not get along too well, this is not totally the case anymore. - -- Asterisk will now handle DNS failures at startup more gracefully, and won't crash and - burn. - -- Choosing to append to a wave file would render the outputted wave file corrupt. Appending - now works again. - -- If you failed to define certain keys, asterisk had the potential to crash when seeing if you had - used them. - -- Attempting to use such things as ${EXTEN:-1} gave a wrong return value. However, this was never - a documented feature... + -- Behavior was changed for 'show codec <number>' to make it more intuitive. + -- DNS failures and asterisk do not get along too well, this is not totally + the case anymore. + -- Asterisk will now handle DNS failures at startup more gracefully, and + won't crash and burn + -- Choosing to append to a wave file would render the outputted wave file + corrupt. Appending now works again. + -- If you failed to define certain keys, asterisk had the potential to crash + when seeing if you had used them. + -- Attempting to use such things as ${EXTEN:-1} gave a wrong return value. + However, this was never a documented feature... Asterisk 1.0.5 diff --git a/pbx/pbx_config.c b/pbx/pbx_config.c index 378dd078c..d67ff3185 100755 --- a/pbx/pbx_config.c +++ b/pbx/pbx_config.c @@ -1687,15 +1687,10 @@ static int pbx_load_module(void) else data = ""; } - pbx_substitute_variables_helper(NULL, ext, realext, sizeof(realext)-1); - cidmatch = strchr(ext, '/'); - if (cidmatch) { - *cidmatch = '\0'; - cidmatch++; - } - stringp=ext; - strsep(&stringp, "/"); - + pbx_substitute_variables_helper(NULL, ext, realext, sizeof(realext)-1); + stringp = realext; + ext = strsep(&stringp, "/"); + cidmatch = stringp; if (!data) data=""; while(*appl && (*appl < 33)) appl++; |