Summary:	ASTERISK-10673: SayDigits seems to be broken in current trunk
Reporter:	Igor Goncharovsky (igorg)	Labels:
Date Opened:	2007-11-02 10:19:37	Date Closed:	2007-11-02 12:11:39
Priority:	Minor	Regression?	No
Status:	Closed/Complete	Components:	Core/PBX
Versions:		Frequency of Occurrence
Related Issues:
Environment:		Attachments:	( 0) asterisk-saydigits.diff.txt
Description:	Seems that problem is in function say_digit_str_full. When I try to debug this issue I've commented check for file existance in this way: if (fn /* && ast_fileexists(fn, NULL, NULL) > 0*/) { ... } Then all digits play normal. Seem that all functions with the same check: say_phonetic_str_full say_character_str_full
Comments:	By: Mark Michelson (mmichelson) 2007-11-02 10:53:38 I just tried the following with Asterisk trunk (revision 88212) and everything worked fine. exten => 124,1,Answer() exten => 124,n,SayDigits(124) exten => 124,n,Hangup() When I called 124, I heard "1 2 4" Could you upload console debug output along with the dialplan you used when you experienced the failure? Thank you. By: Igor Goncharovsky (igorg) 2007-11-02 11:04:35 Only at Tuesday, dialplan mostly the same: exten => 334,1,Ringing() exten => 334,n,Wait(5) exten => 334,n,Answer() exten => 334,n,Read(VAR) exten => 334,n,SayDigits(${VAR}) With verbosity and debug set to 3 console output is empty. No warnings. Sound files placed in new style. By: James Golovich (jamesgolovich) 2007-11-02 11:07:25 I also could not reproduce this.  I tried SayDigits, SayNumber, and the AGI commands and had no problem By: James Golovich (jamesgolovich) 2007-11-02 11:09:23 Using your exact example works for me, so I would suspect that the issue is with how your sound files are added. By: Jason Parker (jparker) 2007-11-02 11:14:57 I can reproduce - quite easily. This is caused by sounds dir structure, and language. I do not have /var/lib/asterisk/sounds/digits/, but I do have /var/lib/asterisk/sounds/en/digits/ I'm not sure what the right way to fix this is..  It appears we aren't passing language.  Perhaps if preflang == NULL, rather than setting it to "", we can set it to chan->language? By: Jason Parker (jparker) 2007-11-02 11:18:15 of course...looking again, we don't actually *have* a reference chan... By: James Golovich (jamesgolovich) 2007-11-02 11:31:54 The patch I just posted fixes it in this case.  I suspect all of the other usages of ast_fileexists need to get updated as well By: Digium Subversion (svnbot) 2007-11-02 11:52:24 Repository: asterisk Revision: 88284 _U  trunk/ U   trunk/main/say.c ------------------------------------------------------------------------ r88284 | qwell | 2007-11-02 11:52:23 -0500 (Fri, 02 Nov 2007) | 13 lines Merged revisions 88283 via svnmerge from https://origsvn.digium.com/svn/asterisk/branches/1.4 (closes issue ASTERISK-10673) ........ r88283 | qwell | 2007-11-02 11:51:08 -0500 (Fri, 02 Nov 2007) | 4 lines We need to make sure to specify a language to ast_fileexists, otherwise it may fail for anything besides en Issue 11147, fix discovered by both citats and myself (independently), with input from Corydon76 ........ ------------------------------------------------------------------------ By: Digium Subversion (svnbot) 2007-11-02 12:11:39 Repository: asterisk Revision: 88285 _U  team/group/CDRfix5/ U   team/group/CDRfix5/CHANGES U   team/group/CDRfix5/main/say.c ------------------------------------------------------------------------ r88285 | murf | 2007-11-02 12:11:37 -0500 (Fri, 02 Nov 2007) | 26 lines Merged revisions 88267,88284 via svnmerge from https://origsvn.digium.com/svn/asterisk/trunk ................ r88267 | tilghman | 2007-11-02 10:26:31 -0600 (Fri, 02 Nov 2007) | 2 lines Add a few bytes on LUA ................ r88284 | qwell | 2007-11-02 10:54:11 -0600 (Fri, 02 Nov 2007) | 13 lines Merged revisions 88283 via svnmerge from https://origsvn.digium.com/svn/asterisk/branches/1.4 (closes issue ASTERISK-10673) ........ r88283 | qwell | 2007-11-02 11:51:08 -0500 (Fri, 02 Nov 2007) | 4 lines We need to make sure to specify a language to ast_fileexists, otherwise it may fail for anything besides en Issue 11147, fix discovered by both citats and myself (independently), with input from Corydon76 ........ ................ ------------------------------------------------------------------------