Re: [SR-Users] sdpops and dynamic codec ids

2011-11-01 Thread Daniel-Constantin Mierla
For the record, this should be working fine for latest master branch, soon to be backported to 3.2, as reported by Juha after the last commits to sdpops module. Cheers, Daniel On 11/1/11 12:45 PM, Daniel-Constantin Mierla wrote: Hello, ok, I see. In this case there are different ids for same

Re: [SR-Users] sdpops and dynamic codec ids

2011-11-01 Thread Daniel-Constantin Mierla
Hello, ok, I see. In this case there are different ids for same name. The search of codec ID stops at first match on name and I think the sdp parser keep the list in reverse order (adding a= lines as first element in a list always). I will look over it. Cheers, Daniel On 11/1/11 12:35 PM, J

Re: [SR-Users] sdpops and dynamic codec ids

2011-11-01 Thread Juha Heinanen
Daniel-Constantin Mierla writes: > Try again with master and let me know if now is working fine. it is getting better, but not quite working yet. in: v=0. o=twinkle 1072425606 342738659 IN IP4 192.98.102.10. s=-. c=IN IP4 192.98.102.10. t=0 0. m=audio 8000 RTP/AVP 99 97 8 0 3 101. a=rtpmap:99 s

Re: [SR-Users] sdpops and dynamic codec ids

2011-11-01 Thread Daniel-Constantin Mierla
Hello, On 10/31/11 7:59 PM, Juha Heinanen wrote: Daniel-Constantin Mierla writes: btw, since I haven't spotted if you said it in previous emails, does keep_codecs_by_name() work now ok with codecs that have dynamic ids? no, as i mentioned, ok, I missed that, being after the report of the no-

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-31 Thread Juha Heinanen
Daniel-Constantin Mierla writes: > btw, since I haven't spotted if you said it in previous emails, does > keep_codecs_by_name() work now ok with codecs that have dynamic ids? no, as i mentioned, it doesn't work. sdp_keep_codecs_by_name("PCMU,PCMA,speex"); on Session Description Prot

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-31 Thread Daniel-Constantin Mierla
Hello, On 10/29/11 6:43 PM, Juha Heinanen wrote: Juha Heinanen writes: it is possible to call both keep_codecs_by_name and keep_codecs_by_id, since 101 does not have name: sorry, 101 does have name (telephone-event), so there is no problem. btw, since I haven't spotted if you said it in previ

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-31 Thread Daniel-Constantin Mierla
Hello, On 10/29/11 6:40 PM, Juha Heinanen wrote: daniel, it is possible to call both keep_codecs_by_name and keep_codecs_by_id, yes, it is possible. Practically calling many times functions removing same part of a SIP message is safe always. When adding results in duplicates of the content.

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-31 Thread Daniel-Constantin Mierla
Hello, there was a positive return code for the case of no-body in SIP message, like it was for SUBSCRIBE. I updated all the other functions as well, since they faced same issue. The fix is slightly different, practically checking for the sdp structure not to be null, since at some point is a

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-29 Thread Juha Heinanen
Juha Heinanen writes: > it is possible to call both keep_codecs_by_name and keep_codecs_by_id, > since 101 does not have name: sorry, 101 does have name (telephone-event), so there is no problem. -- juha ___ SIP Express Router (SER) and Kamailio (Open

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-29 Thread Juha Heinanen
daniel, it is possible to call both keep_codecs_by_name and keep_codecs_by_id, since 101 does not have name: > v=0. > o=twinkle 1548176998 229186862 IN IP4 192.98.102.10. > s=-. > c=IN IP4 192.98.102.10. > t=0 0. > m=audio 8000 RTP/AVP 99 97 8 0 3 101. > a=rtpmap:99 speex/32000. > a=rtpmap:97 spe

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-29 Thread Juha Heinanen
daniel, i fixed the crash like this: *** *** 482,495 int sdp_keep_codecs_by_name(sip_msg_t* msg, str* codecs) { str idslist; ! if(parse_sdp(msg) < 0) { LM_ERR("Unable to parse sdp\n"); return -1; } ! LM_ERR("at

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-29 Thread Juha Heinanen
Daniel-Constantin Mierla writes: > I committed on master branch the code for support of dynamic codecs ids: > > http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=511dc62e6a6ca74324f42b66a23bd9d80b377252 > > I tested and seemed ok, but could try many options, so if you can give

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-27 Thread Daniel-Constantin Mierla
Hello, I committed on master branch the code for support of dynamic codecs ids: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=511dc62e6a6ca74324f42b66a23bd9d80b377252 I tested and seemed ok, but could try many options, so if you can give it a try as well would be good bef

Re: [SR-Users] sdpops and dynamic codec ids

2011-10-22 Thread Daniel-Constantin Mierla
Hello, Right now is working only for non-dynamic codecs ids, indeed. There was a bug in the sdp parser that prevented implementing it in the first place, but when I fixed that before the 3.2.0 release, I forgot about this one. I will try to add it asap. Cheers, Daniel On 10/22/11 3:20 PM, J

[SR-Users] sdpops and dynamic codec ids

2011-10-22 Thread Juha Heinanen
i made call sdp_keep_codecs_by_name("PCMU,PCMA,speex"); on an invite request and it didn't keep speex although it was in the sdp: Session Description Protocol Session Description Protocol Version (v): 0 Owner/Creator, Session Id (o): sems 1 1 IN IP4 192.98.102.10