| 01:15.57 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 01:15.57 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 01:16.07 | mjordan | file: two quick questions if you're around |
| 01:16.51 | Juggie | 1) no 2) sometimes, but only on sundays |
| 01:17.52 | TazzNZ | and mjordan - that must have been an awesome dinner :) |
| 01:47.35 | coreyfarrell | does anyone know why app_voicemail insist on building every time anything in the source tree changes? |
| 01:51.50 | mjordan | coreyfarrell: dunno for sure. You may want to remove <defaultenabled> |
| 01:52.08 | coreyfarrell | sorry I phrased wrong. rebuilding |
| 01:52.43 | mjordan | coreyfarrell: yeah, I suspect it is a bug in menuselect |
| 01:52.52 | coreyfarrell | that is I run make, it builds everything (including app_voicemail.so), then edit res/res_rtp_asterisk.c, make - it builds app_voicemail again |
| 01:52.54 | mjordan | if you remove <defaultenabled> from the top level MODULEINFO |
| 01:53.04 | mjordan | it stops building app_voicemail every time |
| 01:54.29 | coreyfarrell | trying it now (I removed defaultenabled from app_voicemail.c) |
| 01:55.53 | coreyfarrell | mjordan: awesome that worked. I'll get a ticket/review open to remove defaultenabled (also from res_pjsip_logger). shouldn't cause a change since both are 'core' modules? |
| 01:56.02 | mjordan | well... :-) |
| 01:56.22 | mjordan | it was added for a reason. If you don't have app_voicemail defaultenabled, you'll run into funky defaults with res_mwi_external |
| 01:56.29 | mjordan | as in, neither gets selected |
| 01:56.42 | mjordan | we added that because none of the voicemail tests were getting run on the build agents |
| 01:56.50 | coreyfarrell | lame |
| 01:56.58 | mjordan | I know, right? |
| 01:57.17 | mjordan | I'd still file a bug |
| 01:57.38 | mjordan | menuselect (or something in there, at any rate) should be preventing it from getting built each time |
| 01:57.40 | coreyfarrell | I don't like the build-time conflict's.. we really need to overhaul the module loader so it can support run-time conflicts |
| 01:57.45 | mjordan | agreed |
| 01:57.53 | mjordan | amongst many other improvements |
| 01:58.39 | coreyfarrell | I've been having idea's in about how to replace loader.c for asterisk 14 |
| 01:59.46 | mjordan | looks for a wiki page, brb |
| 02:01.09 | mjordan | coreyfarrell: can you see this page - https://wiki.asterisk.org/wiki/display/~dlee/Asterisk+Module+Loading+Improvements |
| 02:01.22 | coreyfarrell | looking at it now |
| 02:05.00 | coreyfarrell | some good stuff. I've been thinking about things to make the load/unload simpler, and to replace OPTIONAL_API. |
| 02:05.07 | coreyfarrell | the documentation is on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying 'Beware of the Leopard' - HA |
| 02:07.49 | coreyfarrell | I've been thinking of ways to automate standard registration objects (CLI, AMI, dialplan apps, functions, config options). so adding pointers to those items to ast_module_info, having loader.c handle the registration/cleanup.. |
| 02:09.45 | coreyfarrell | also better handling of unloading modules in general. I haven't looked at any code yet (will do that after /branches/13 exists), but I think I have a way to allow module unload defer action until the module gets 0 refs |
| 02:15.07 | mjordan | This is another one of those "all or nothing changes" - I think it'd be interesting to see a prototyped version before we apply it across the board. I'm not sure how easy that will be to do :-) |
| 02:15.28 | mjordan | but it'd be nice to build out some module graphs to test it out with, etc. |
| 02:15.45 | mjordan | (not to mention that this would officially nuke every 3rd party module out there) |
| 02:16.15 | coreyfarrell | yup, I had seen dlee's page in the past, which is why I assumed he or someone else at Digium was thinking about load ordering issues |
| 02:16.18 | Juggie | woudn't you know evolving be a good idea at some point |
| 02:17.24 | coreyfarrell | that's why I've been mostly thinking about what should happen after we dlopen and start looking at the ast_module_info, and how it should decide to unload a module |
| 02:18.13 | coreyfarrell | I agree it's all or nothing, but I think very worth it |
| 02:19.02 | coreyfarrell | BTW addons/chan_ooh323 causes asterisk to exit (in my test of media_formats-reviewed-trunk) |
| 02:19.22 | coreyfarrell | weird thing is it actually exits, it didn't cause a crash |
| 02:20.06 | Juggie | coreyfarrell, while you are in there... fix output from modules :) |
| 02:27.06 | *** join/#asterisk-dev coreyf1513 (~cfarrell@2001:4830:117f:2:d466:5130:3a9d:8ef) |
| 02:27.24 | coreyf1513 | mjordan: so refcounter.py just locked up my desktop |
| 02:27.33 | mjordan | coreyfarrell: we were really hoping we'd get to it before 13. Unfortunately... other things were more important (like, not crashing) |
| 02:27.38 | mjordan | wow |
| 02:27.43 | mjordan | that's impressive |
| 02:28.10 | coreyf1513 | BTW the last I saw was Juggie telling me to fix output from modules |
| 02:28.16 | coreyf1513 | not sure if I missed anything |
| 02:28.19 | mjordan | that's the last thing |
| 02:28.23 | mjordan | I'm digging through res_rtp_asterisk |
| 02:28.31 | mjordan | so I'm a bit in & out |
| 02:29.17 | coreyf1513 | Juggie: I don't generally do much with logging/CLI output, but if something bothers you then you should post a ticket in JIRA with a patch :P |
| 02:31.03 | coreyf1513 | mjordan: Ctrl-Alt-F1 FINALLY switched to console, hoping that I can recover my system by killing refcounter.py |
| 02:31.22 | coreyf1513 | it's definitely still running, I think it saturated my RAM |
| 02:32.43 | coreyf1513 | so we're going to need to look into that :) |
| 02:32.58 | *** join/#asterisk-dev coreyfarrell (~cfarrell@2001:4830:117f:2:7c3b:9bec:325a:17ed) |
| 02:34.08 | mjordan | apparently caching things was a bad idea |
| 02:34.40 | mjordan | and +1 for posting a patch for log output. Patches that improve messages are always nice |
| 02:35.17 | coreyf1513 | yeah my console said that refcounter.py had allocated 3.28GB ram |
| 02:35.53 | Juggie | coreyf1513, it bothered me in 1.2 and 1.4 nd 1.6 |
| 02:36.00 | Juggie | and then i got over it and dont really care ;) |
| 02:36.07 | coreyf1513 | still waiting for my GUI to repaint the first time, apparently xchat gets to use my desktop before me |
| 02:36.32 | Juggie | i dont use * anymore at work so i am meerly a spectator now. |
| 02:37.22 | Juggie | but seriously, the console does block modules and modules need to know about each form of possible output (ami/csv/cli/etc) |
| 02:37.26 | Juggie | i think someone was working on it |
| 02:37.29 | Juggie | not sure |
| 02:37.33 | coreyfarrell | awesome I got my desktop back |
| 02:38.09 | Juggie | i vaugely remember it comming up recently. |
| 02:38.38 | Juggie | id say maybe a rewrite would be better, but the last attempt that that... |
| 02:38.44 | Juggie | i need say no more :P |
| 02:44.19 | coreyfarrell | mjordan: wow, so refcounter.py reached over 3GB of RAM processing a 232MB refs file |
| 02:51.02 | *** join/#asterisk-dev mvanbaak (~michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 02:51.02 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 02:55.56 | TazzNZ | oh - can someone give me a quick estimation of the Asterisk source tree size ? |
| 02:57.57 | mjordan | hm |
| 02:58.00 | mjordan | that's suboptimal |
| 02:58.09 | mjordan | TazzNZ: I'm not sure I understand the question |
| 02:58.14 | mjordan | SLOC? |
| 02:58.48 | mjordan | about 1.5 million lines for that. |
| 02:58.49 | mjordan | https://code.asterisk.org/code/project/CR-AST |
| 02:59.16 | coreyfarrell | TazzNZ: a clean svn checkout of trunk is 106MB |
| 02:59.19 | TazzNZ | mjordan: I am about to do a svn "checkout http://svn.asterisk.org/svn/asterisk/trunk asterisk" |
| 02:59.25 | TazzNZ | ah - there we go :) |
| 02:59.53 | mjordan | (my graph foo was reasonable: 1,433,418 SLOC). Which is a terrible estimate of anything, but an interesting line graph nonetheless |
| 03:00.15 | TazzNZ | I built a "dev" VM. I allocated enough disk. All is good |
| 03:00.26 | coreyfarrell | TazzNZ: one of my checkouts is 382MB after being built |
| 03:00.41 | snuff-work | the stats section says 1.858 mill.. but it also says last commit 26 jul 10 |
| 03:00.58 | TazzNZ | coreyfarrell: cool - I have 100gig allocated....*should* just make it :D |
| 03:05.58 | coreyfarrell | time for a reboot, still running like crap since refcounter.py push me into swap space :( |
| 03:09.12 | *** join/#asterisk-dev coreyf1513 (~cfarrell@2001:4830:117f:2:809f:5ee3:4f4d:85d9) |
| 03:09.18 | TazzNZ | .. |
| 03:09.42 | TazzNZ | geepers....SSD *only* ? |
| 03:11.44 | TazzNZ | ok...that is odd - I am getting an error 500 trying to check out the svn |
| 03:11.59 | TazzNZ | svn: Server sent unexpected return value (500 Internal Server Error) in response to REPORT request for '/svn/asterisk/!svn/vcc/default' |
| 03:13.52 | mjordan | snuff-work: that's all repos under Asterisk |
| 03:14.03 | snuff-work | ah |
| 03:14.58 | snuff-work | i want to get back into doing stuff work isn't doing much phone stuff atm.. finding lack of motivation |
| 03:15.20 | mjordan | snuff-work: what are you interested in? |
| 03:15.45 | snuff-work | i do a little with wedhorn on the skinny side.. mostly him fixing me finding issues |
| 03:16.02 | mjordan | always good |
| 03:16.39 | snuff-work | i did do doco stuff earlier on (aka over a year ago).. |
| 03:17.06 | mjordan | I know! |
| 03:17.11 | mjordan | It was very appreciated :-) |
| 03:17.20 | mjordan | it pretty much formed the basis of all of the pjsip documentation |
| 03:17.52 | mjordan | if you wanted to go one step further with documentation, there's a feature I'd like to see added the XML documentation. It's less a "documentation" task however and more an enhancement to astxml |
| 03:18.10 | mjordan | if not, no worries |
| 03:19.19 | snuff-work | i dont mind that stuff |
| 03:21.04 | mjordan | snuff-work: so, one of the things that mucks up documentation a lot are examples. They don't get formatted well as paragraphs |
| 03:21.26 | mjordan | what's more, there's no way to mark a set of paragraphs as documentation, so they end up looking a lot like paragraphs of text |
| 03:21.34 | mjordan | where really, examples typically include blocks of dialplan |
| 03:21.49 | mjordan | It'd be nice to have an actual dedicated XML document element for examples |
| 03:21.54 | mjordan | So something like: |
| 03:22.35 | mjordan | <example><para>same => n,Set(CHANNEL(foo)=bar)</para><para>same => n,Set(CHANNEL(blah)=blech)</para></example> |
| 03:22.40 | mjordan | astxml would have to be updated to understand and process example elements |
| 03:22.46 | mjordan | and dump out things formatted properly on the CLI |
| 03:23.03 | TazzNZ | so...would anyone mind trying to check out the SVN trunk branch via http please ? |
| 03:23.03 | mjordan | but if it was part of the XML schema, we could update the script that generates the wiki docs to use {code} or {noformat} tags |
| 03:23.28 | mjordan | TazzNZ: works for me |
| 03:23.33 | TazzNZ | hhhmmmm |
| 03:23.37 | coreyf1513 | TazzNZ: no problem here |
| 03:23.50 | TazzNZ | the universe doesn't want me to help :( |
| 03:24.03 | mjordan | svn co http://svn.asterisk.org/svn/asterisk/trunk trunk-checkout |
| 03:24.59 | mjordan | when do you get your 500 error? |
| 03:25.04 | mjordan | has it checked anything out? |
| 03:25.07 | TazzNZ | svn: Server sent unexpected return value (500 Internal Server Error) in response to REPORT request for '/svn/asterisk/!svn/vcc/default' |
| 03:25.12 | mjordan | are you on windows? |
| 03:25.18 | TazzNZ | Ubuntu 12.04 |
| 03:26.24 | mjordan | can you ping svn.asterisk.org? |
| 03:26.28 | coreyf1513 | TazzNZ: do you have some kind of proxy interfering with your HTTP? |
| 03:26.49 | TazzNZ | coreyf1513: nope - nothing |
| 03:27.02 | TazzNZ | mjordan: ping is timing out....which is odd |
| 03:27.09 | TazzNZ | I'll look into that :) |
| 03:27.10 | mjordan | well, that would explain the error you're getting |
| 03:27.30 | snuff-work | ur box is too secure =) |
| 03:27.34 | mjordan | coreyf1513: res_rtp_asterisk built. Probably will explode. |
| 03:27.52 | TazzNZ | snuff-work: yeah.....I comply to the C2 requirements of NT4 :) |
| 03:27.56 | coreyf1513 | mjordan: that's awesome.. I'm making some changes to refcounter.py to make it use less memory |
| 03:28.25 | coreyf1513 | unfortunately my change will result in calling parse_line() twice for every line |
| 03:28.35 | mjordan | cool. I guess that validates your earlier findings :-) |
| 03:28.46 | mjordan | I'm not sure it matters how long it takes to process a ref debug file |
| 03:28.51 | mjordan | I mean, hours would be bad |
| 03:29.35 | coreyf1513 | worst this will do is double the processing time, but it stores the unprocessed lines in the array's, instead of the hash's of parsed fields |
| 03:29.58 | coreyf1513 | then once it gets to the part of printing the objects it has to split each line again |
| 03:31.04 | coreyf1513 | I'm also tweaking it so if you disable output of certain object types (normal/skewed/leaks) it will not populate those lists, so normal objects will fall out of scope once we get the destructor |
| 03:31.55 | coreyf1513 | mjordan: you want to either post or email me the res_rtp_asterisk patch? |
| 03:32.05 | mjordan | sure |
| 03:32.10 | TazzNZ | there we go :) |
| 03:32.12 | mjordan | I'll e-mail it to you |
| 03:34.19 | mjordan | I haven't dealt with packetization yet. That's going to require a lot of BUGBUGs to get fixed, and adding a format_cap structure to rtp_engine. That's up for tomorrow - but this should get closer. Audio *may* work. |
| 03:34.27 | coreyf1513 | well I'm about to try refcounter.py with my changes, hopefully storing the lines whole will reduce the memory enough to not crash. |
| 03:34.41 | coreyf1513 | cool, I'll give chan_sip a smoke test |
| 03:34.54 | mjordan | off to bed, night |
| 04:43.52 | *** join/#asterisk-dev stevePearPear (~stevePear@202.166.82.164) |
| 05:23.23 | *** join/#asterisk-dev Qwell (north@pdpc/sponsor/digium/Qwell) |
| 05:23.23 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 05:38.58 | *** join/#asterisk-dev Qwell (north@pdpc/sponsor/digium/Qwell) |
| 05:38.58 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 05:54.53 | *** join/#asterisk-dev Qwell (north@pdpc/sponsor/digium/Qwell) |
| 05:54.53 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 06:10.09 | *** join/#asterisk-dev Qwell (north@pdpc/sponsor/digium/Qwell) |
| 06:10.09 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 06:10.17 | *** join/#asterisk-dev stevePearPear (~stevePear@202.166.82.164) |
| 06:10.29 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 06:10.29 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 06:32.59 | *** join/#asterisk-dev bulkorok (~Benjamin@85.183.61.47) |
| 06:35.15 | oej | Morning folks! |
| 06:43.51 | *** join/#asterisk-dev bulkorok (~Benjamin@85.183.61.47) |
| 06:55.12 | coreyfarrell | morning oej! btw nice half-joke about deprecating chan_iax2 |
| 08:01.04 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 08:53.58 | *** join/#asterisk-dev danjenkins_ (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 09:06.13 | oej | he he |
| 09:06.21 | skrusty | morning |
| 09:08.11 | *** join/#asterisk-dev tgu (~tgu@host-212-68-194-46.brutele.be) |
| 09:20.37 | *** join/#asterisk-dev danjenkins__ (~dan@nat/digium/x-jhurcifjgaooncdb) |
| 12:00.01 | *** join/#asterisk-dev Marquis42 (~mrbmw@d149-67-95-208.try.wideopenwest.com) |
| 12:22.11 | *** join/#asterisk-dev Qwell (~north@pdpc/sponsor/digium/Qwell) |
| 12:22.11 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 12:24.13 | *** join/#asterisk-dev stevePearPear (~stevePear@cm244.epsilon47.maxonline.com.sg) |
| 13:04.42 | vlad_starkov | malcolmd: Hi, are you there? |
| 13:38.28 | *** join/#asterisk-dev kharwell (~kharwell@nat/digium/x-ejrsnkcujbrzwkms) |
| 13:55.01 | malcolmd | vlad_starkov: so, audiofilter's something that was provided by the author of speex - i can't tell you a lot about what it's doing. you can see an example of its use in the converters script in the scripts sub-directory of sound_tools. |
| 13:58.46 | vlad_starkov | malcolmd: thanks |
| 13:59.42 | vlad_starkov | malcolmd: one more question: is there any best practices of preparing sound prompts and moh files for asterisk? |
| 14:00.20 | malcolmd | we have no such document |
| 14:00.44 | vlad_starkov | malcolmd: excerpt from asterisk wiki "(There is an ITU specification that states for 8KHz audio that is being compressed frequencies below a certain threshold should be removed because they make the resulting compressed audio sound worse than it should.)" |
| 14:01.14 | malcolmd | yes, i'm aware that's on the wiki, that's a statement from the gentleman responsible for providing the audiofilter application |
| 14:01.43 | vlad_starkov | malcolmd: right. Do you know where can I find him to ask this question? |
| 14:04.38 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-218-28-58.cablep.bezeqint.net) |
| 14:04.38 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 14:06.51 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 14:07.04 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:09.58 | *** join/#asterisk-dev jameswf (uid27319@gateway/web/irccloud.com/x-gpwdcipllitigack) |
| 14:14.25 | file | PON PON PON |
| 14:17.07 | vlad_starkov | malcolmd: anyways thank you for your answer on subject! |
| 14:17.24 | malcolmd | vlad_starkov: good luck :) |
| 14:17.59 | vlad_starkov | malcolmd: yep!) |
| 14:18.12 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 14:18.12 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 14:25.10 | *** join/#asterisk-dev otherwiseguy (~otherwise@69.76.221.33) |
| 14:25.31 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 14:33.30 | *** join/#asterisk-dev ipengineer (~zconkle@static-71-252-134-63.dllstx.fios.verizon.net) |
| 14:46.16 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 14:54.28 | *** join/#asterisk-dev gtjoseph (~gtj0@unaffiliated/gtj) |
| 15:11.50 | *** join/#asterisk-dev rmudgett (~rmudgett@nat/digium/x-dpeqmcjvouaadqpp) |
| 15:18.28 | *** join/#asterisk-dev Qwell (~north@pdpc/sponsor/digium/Qwell) |
| 15:18.28 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 15:30.42 | *** join/#asterisk-dev sgriepentrog (~sgriepent@nat/digium/x-keowdwwhlnqvuwzs) |
| 15:49.38 | *** join/#asterisk-dev newtonr (~newtonr@nat/digium/x-qbxpjjzebdrwdsit) |
| 15:49.38 | *** mode/#asterisk-dev [+o newtonr] by ChanServ |
| 15:50.09 | file | goose! |
| 15:59.51 | ipengineer | file: Does Marks recent work on resource list and pjsip component abstraction change anything we have talked about in the past in regards to dialog-info events? |
| 16:00.04 | file | not really |
| 16:00.58 | ipengineer | K |
| 16:06.15 | *** join/#asterisk-dev protocoldoug (~quassel@unaffiliated/protocoldoug) |
| 16:30.34 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-175-118.mci.googlefiber.net) |
| 17:00.21 | *** join/#asterisk-dev danjenkins_ (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 17:01.53 | *** join/#asterisk-dev protocoldoug (~quassel@unaffiliated/protocoldoug) |
| 17:05.47 | jameswf | malcolmd newtonr http://community.freepbx.org/t/digium-says-epm-pro-causes-phone-browser-issue/22798/3 |
| 17:27.32 | coreyfarrell | wdoekes: nice catch on ASTERISK-23790. do you want me to update my review with your patch or do you want to take over? I don't want to steal your patch without permission.. |
| 17:29.57 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 17:45.19 | *** join/#asterisk-dev danjenkins__ (~dan@cpc11-folk2-2-0-cust171.1-2.cable.virginm.net) |
| 18:07.18 | opticron | finally got everything worked out and issues addressed for https://reviewboard.asterisk.org/r/3649/ for anyone interested |
| 18:43.35 | malcolmd | jameswf: saw it |
| 18:57.42 | malcolmd | jameswf: commented, took a bit to gather details and make a sensible response |
| 19:07.46 | coreyfarrell | file: any thoughts on how to deal with 'union ast_frame_subclass'? I just attempted to make a call to chan_sip, got a segfault due to f.subclass.format == 0x1 (set via f.subclass.integer). |
| 19:08.36 | file | define "deal with" - it should be a pointer if it's a voice/video frame |
| 19:09.19 | coreyfarrell | file: channel.c / ast_queue_hangup sets subclass.integer to AST_CONTROL_HANGUP. this is the same memory used for subclass.frame |
| 19:10.01 | file | right, because in control frames it stores an integer value |
| 19:10.25 | file | so why is it crashing? what is trying to access it as a pointer? |
| 19:11.10 | coreyfarrell | http://paste.ubuntu.com/7696611/ |
| 19:11.57 | file | so ast_frdup isn't checking to see if the frame being copied is of one that contains a format |
| 19:12.05 | file | that's the bug |
| 19:12.37 | coreyfarrell | i'm not sure how it would check, since &f.subclass.integer == &f.subclass.format |
| 19:12.45 | file | frametype |
| 19:12.57 | file | different frame types use the union differently |
| 19:13.28 | coreyfarrell | right.. thanks I should have noticed that :) |
| 19:14.35 | file | it's actually sorta interesting... the way the old way worked |
| 19:14.40 | *** join/#asterisk-dev [TK]D-Fender (~chatzilla@64.235.216.2) |
| 19:15.06 | file | since format was the largest thing in the frame it just copied it entirely |
| 19:16.15 | file | so if it's not voice/video then it needs to be copied too |
| 19:16.28 | file | and now I walk home |
| 19:16.32 | file | be back in... a... FLASH |
| 19:29.33 | wdoekes | coreyfarrell: go ahead and steal it |
| 19:33.08 | file | yay home |
| 19:52.11 | *** join/#asterisk-dev ipengineer (~zconkle@static-71-252-134-63.dllstx.fios.verizon.net) |
| 19:52.31 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-170-9.mci.googlefiber.net) |
| 19:55.56 | *** join/#asterisk-dev bulkorok (~Benjamin@gw1.pinguin.ag) |
| 19:56.51 | *** join/#asterisk-dev otherwiseguy (~otherwise@23-228-170-9.mci.googlefiber.net) |
| 20:07.11 | opticron | anyone who feels like doing a testsuite review: https://reviewboard.asterisk.org/r/3560/ |
| 20:09.00 | coreyfarrell | file: resolved the crash, now it's failing to negotiate codecs. my 11.x PBX is sending rtpmap attributes for ULAW and telephone-event, and media_formats is replying with telephone-event only. I am assuming this is something wrong with chan_sip, reviewing the diff now |
| 20:10.02 | file | chan_sip / rtp_engine |
| 20:12.59 | coreyfarrell | well I was also thinking about the replacement for ast_best_codec / ast_codec_choose.. what if ast_format_cap_get_format(cap, 0) is not audio? |
| 20:14.55 | file | that would get... even more messy |
| 20:15.24 | file | hrm |
| 20:15.32 | coreyfarrell | i'm rebuilding with an ast_debug in ast_format_cap_get_format to tell what's being returned when position==0 |
| 20:17.05 | *** join/#asterisk-dev Dovid (~Dovid@static-173-63-105-210.nwrknj.fios.verizon.net) |
| 20:19.21 | coreyfarrell | ok so it's not that.. all calls to that procedure gave ulaw |
| 20:49.36 | coreyfarrell | file: AST_FORMAT_CMP_NOT_EQUAL != 0, so if (!ast_format_cap_is_compatible_format()) {} does not do what is expected :( |
| 20:50.24 | file | let us weep for that which has been done and will be done |
| 20:53.47 | coreyfarrell | so I'm dealing with ulaw which is never AST_FORMAT_CMP_SUBSET, but for other stuff I'm not sure if I should check that it is AST_FORMAT_CMP_NOT_EQUAL or check if it's not AST_FORMAT_CMP_EQUAL. |
| 20:55.27 | file | SUBSET and EQUAL is compatible |
| 20:57.17 | coreyfarrell | thanks |
| 21:03.14 | coreyfarrell | I just had the first successful run of app_playback on a call from asterisk 11 to media_formats-reviewed-trunk! |
| 21:03.21 | file | with audio? |
| 21:03.31 | coreyfarrell | "I am sorry, this is not a valid extension" |
| 21:03.44 | coreyfarrell | fracked like hell when I ran 'core stop now', but I heard audio! |
| 21:03.54 | file | nifty |
| 21:05.18 | coreyfarrell | I think maybe functions like ast_channel_nativeformats() could be partly responsible for the fracking, I had assumed this would return ao2_bump(chan->nativeformats), but no bump.. |
| 21:08.24 | file | yeah, I initially did it that way but after seeing how that stuff was used didn't... the cleanup implications is... astounding |
| 21:09.16 | coreyfarrell | fair enough, so we assume that ast_channel is locked when running that procedure, therefore the reference held by chan->nativeformats is protected? |
| 21:09.52 | file | yes |
| 21:10.30 | coreyfarrell | works for me. oh speaking of reference stuff, I think we should do away with ast_format_copy, and just directly use ao2_bump. |
| 21:11.00 | file | if you just #define it... sure |
| 21:11.43 | coreyfarrell | ok so #define ast_format_copy and ast_format_replace as aliases to ao2_bump / ao2_replace? |
| 21:11.53 | file | sure |
| 21:12.06 | file | I assume this is for your REF_DEBUG? |
| 21:12.18 | coreyfarrell | yes. ast_format_copy is terrible for REF_DEBUG |
| 21:12.25 | file | yup yup go for it |
| 21:13.05 | coreyfarrell | on the other hand I think ast_format_copy is a bad name for the procedure, but I'll let mjordan deal with that when he does the API renames |
| 21:13.23 | coreyfarrell | cause it's not a copy, it's a reference.. |
| 21:25.52 | *** join/#asterisk-dev TazzNZ (~pieter@office-nat1.ytechops.co.nz) |
| 22:36.52 | *** part/#asterisk-dev bulkorok (~Benjamin@gw1.pinguin.ag) |
| 23:34.17 | *** part/#asterisk-dev kharwell (~kharwell@nat/digium/x-ejrsnkcujbrzwkms) |