| 00:08.15 | codefreeze | codefreeze: yo? |
| 00:11.53 | snuff-work | about your new stuff hashtab :) |
| 00:12.12 | snuff-work | can i make the doxygen (aka your not doing it atm are u) ? |
| 00:19.54 | codefreeze | snuff-work: no, I'm not playing it with it right now... feel free. Let me know if I need to do anything... |
| 00:27.38 | snuff-work | k i'm putting something small together.. hopefully 1/2 of hashtab.c will be doxygened by tommorrow |
| 00:28.42 | codefreeze | sounds like a biological process... doxygenation. I picture O2 levels in a lake.... |
| 00:33.36 | snuff-work | mmm.. if i could work out how doxygen function pointers properly rather than using \arg would be nice.. |
| 00:34.08 | snuff-work | also noticed hashtab goes against the usual ast functions (only looked at 3) where 0 = failure |
| 00:37.58 | codefreeze | Send me a list of this kind of thing-- I might be able to do something about it... ;) |
| 00:39.25 | snuff-work | np.. i was just browsing it since work was boring :) i shall send u list |
| 01:44.22 | *** join/#asterisk-dev Mackes (n=root@cpe-76-180-236-119.buffalo.res.rr.com) |
| 02:23.26 | *** join/#asterisk-dev CoffeeKid (n=kirk@dsl093-224-026.slc1.dsl.speakeasy.net) |
| 02:24.04 | CoffeeKid | how would one go about making a dynamic loadable modules in asterisk? |
| 02:24.58 | CoffeeKid | i want to do something similar to the cdr_addon_mysql module, but it does some custom stuff specific to our company needs. |
| 02:28.55 | CoffeeKid | the biggest part i'm confused on is, how does asterisk load modules? do i need to add code to the core asterisk system to load a module? or is it dynamic? |
| 02:29.34 | CoffeeKid | and how do i tell asterisk when to execute the module code, right now i've just been hacking logger.c |
| 02:31.43 | Corydon76-dig | CoffeeKid: depends on /etc/asterisk/modules.conf |
| 02:32.14 | Corydon76-dig | autoload=yes makes it load everything from /usr/lib/asterisk/modules unless there's an explicit noload |
| 02:32.31 | Corydon76-dig | vice-versa if autoload=no |
| 02:32.56 | Corydon76-dig | load_module is executed on load, unload_module is executed on unload |
| 02:34.27 | Corydon76-dig | the module should register whatever it wants to run with the appropriate backend handler |
| 02:35.18 | CoffeeKid | So, if i want like the queue_log function to execute my own code, do i need to register a function to execute the module code in logger.c? |
| 02:36.38 | *** part/#asterisk-dev Mackes (n=root@cpe-76-180-236-119.buffalo.res.rr.com) |
| 02:36.47 | Corydon76-dig | Yes, basically |
| 02:37.01 | Corydon76-dig | And there is none, currently |
| 02:37.55 | Corydon76-dig | but if you want there to be, I think we could create such a backend function to do that... and I'd strongly encourage you to contribute your code |
| 02:38.33 | CoffeeKid | ast_queue_log |
| 02:38.40 | CoffeeKid | is the function i want to execute the module in. |
| 02:38.46 | CoffeeKid | yeah, i would have no problem with that.. |
| 02:39.24 | CoffeeKid | basically its just a module to check the status of the extension in the queue, and then tie that to an agent within our database, and update the status (that is then updated in a gui) |
| 02:40.38 | CoffeeKid | so i'm guessing i'd create a handler in logger.c to autoload modules.. right? |
| 02:40.47 | Corydon76-dig | No, the loader handles that |
| 02:41.10 | CoffeeKid | whats a good example of something that does that already? |
| 02:41.12 | Corydon76-dig | You create a shared linked list... and create register/unregister functions |
| 02:41.22 | Corydon76-dig | main/cdr.c, for example |
| 02:41.57 | Corydon76-dig | and the module calls the register or unregister functions in its own load_module or unload_module, respectively |
| 02:42.48 | CoffeeKid | what functions in cdr.c are a good example of what i'm trying to do? |
| 02:44.15 | CoffeeKid | ast_cdr_register and ast_cdr_unregister ? |
| 02:44.29 | Corydon76-dig | Correct |
| 02:45.10 | CoffeeKid | I guess where i'm confused is.. I know cdr_addon_mysql is used as a module like i'm wanting to do, where is the module code executed at? |
| 02:45.15 | CoffeeKid | within cdr.c |
| 02:45.19 | CoffeeKid | how is that all handled? |
| 02:45.40 | Corydon76-dig | Actually, ast_custom_function_register is the current way of doing it... it passes a pointer to a struct |
| 02:46.24 | Corydon76-dig | main/pbx.c is where that's found |
| 02:47.50 | Corydon76-dig | The nice thing about that is that if we want to add additional register elements, we don't have to change the API function |
| 02:48.05 | CoffeeKid | right, thats what i'm going for |
| 02:48.14 | CoffeeKid | i'd rather not modify the core asterisk functions if i can get away with it. |
| 02:48.25 | CoffeeKid | core asterisk code rather. |
| 02:49.23 | CoffeeKid | but since logger.c isn't setup for custom functions, would that need to be modified? |
| 02:49.55 | Corydon76-dig | Correct |
| 02:50.32 | CoffeeKid | and how would that be done? |
| 02:51.17 | CoffeeKid | sorry for sounding like noob, i'm new to the asterisk code, trying wrap my head around the module stuff :) |
| 02:51.32 | Corydon76-dig | Not a problem. Should be a good learning experience |
| 02:52.10 | CoffeeKid | true, i do so many hacks to try and work with whats there, i'm getting to the point where i should be working with the code directly, helping if possible :) |
| 02:56.11 | CoffeeKid | so back to the original idea, where does cdr.c load the cdr_addon_mysql module for example, then execute the module code? |
| 02:56.31 | Corydon76-dig | loader.c loads the module |
| 02:56.44 | Corydon76-dig | the module registers itself with cdr.c |
| 02:57.16 | Corydon76-dig | and cdr.c calls the backend function pointer when the cdr is posted |
| 02:57.31 | Corydon76-dig | You know about function pointers, right? |
| 02:57.46 | CoffeeKid | right |
| 02:58.10 | CoffeeKid | where is the function pointer called in cdr.c? |
| 02:59.06 | Corydon76-dig | Line 1000 |
| 02:59.14 | Corydon76-dig | in post_cdr |
| 02:59.21 | Corydon76-dig | i->be(cdr); |
| 03:00.20 | CoffeeKid | AST_LIST_TRAVERSE, thats interesting, how does that work? |
| 03:00.45 | Corydon76-dig | see include/asterisk/linkedlists.h |
| 03:01.34 | Corydon76-dig | We macro-ize the linked lists, to help avoid making common mistakes in list traversal |
| 03:01.42 | CoffeeKid | ah, thats a macro, i see whats going on. |
| 03:02.40 | CoffeeKid | ok, so you could add a function pointer to the function pointer list, then make logger.c execute AST_LIST_TRAVERSE where you wanted it to run the custom function |
| 03:04.08 | Corydon76-dig | Right, for every module which registered a function pointer |
| 03:04.36 | CoffeeKid | so i->be(cdr); only runs the functions that are listed to be cdr functions in the list, right? |
| 03:04.52 | Corydon76-dig | Correct |
| 03:04.58 | CoffeeKid | that is very cool. |
| 03:06.09 | CoffeeKid | so if i create a module, then modify logger.c in the correct location, put my module in /var/lib/asterisk/modules .. it'll do what i'm after? |
| 03:06.41 | CoffeeKid | logger.c would probably need two modifications, one to handle lists in ast_log, and one to handle ast_queue_log |
| 03:06.44 | CoffeeKid | is my guess. |
| 03:07.59 | CoffeeKid | now the question is, how do i make a new list specific to those two types... say log and queue_log |
| 03:08.37 | CoffeeKid | i->be(log); i->be(queue_log); .. does that require a modification to loader.c .. or do you define that in the module itself? |
| 03:09.12 | Corydon76-dig | Well, remember that it's two different lists, with different APIs |
| 03:10.12 | CoffeeKid | so loader.c i think |
| 03:10.47 | CoffeeKid | looks like around line 232 |
| 03:11.16 | CoffeeKid | i'm on source 1.4.11 btw |
| 03:11.43 | Corydon76-dig | The only thing to note is that we don't make these kinds of changes to 1.4... only trunk... |
| 03:11.55 | CoffeeKid | right |
| 03:11.59 | CoffeeKid | that makes sense. |
| 03:12.48 | Corydon76-dig | Again, no change should be necessary to loader.c |
| 03:13.20 | CoffeeKid | ok, its dynamic enough to see new list types? |
| 03:13.53 | Corydon76-dig | The loader has nothing to do with it |
| 03:14.09 | Corydon76-dig | The loader just loads the module into memory and executes load_module |
| 03:14.25 | CoffeeKid | ah, right, but doesn't it create the list? |
| 03:14.39 | Corydon76-dig | That's all. It's the module itself which registers to logger.c... and logger.c modifies the list |
| 03:14.58 | Corydon76-dig | Nothing outside of logger.c even knows the list exists |
| 03:14.58 | CoffeeKid | ah, I see |
| 03:15.35 | Corydon76-dig | It's good compartmentalization |
| 03:16.18 | Corydon76-dig | since logger.c manages the list, if an API change is required, that's the only place you have to change code |
| 03:16.41 | Corydon76-dig | Oh, and you'll need to modify include/asterisk/logger.h to include the new register/unregister functions |
| 03:16.49 | CoffeeKid | right, makes sense |
| 03:17.16 | Corydon76-dig | That's how your module knows the API for the register function... and the structure to register |
| 03:19.08 | CoffeeKid | i see, so logger.c would need something like ast_cdr_register, but ast_logger_register |
| 03:19.16 | Corydon76-dig | Right |
| 03:19.41 | Corydon76-dig | Given that there are 3 different loggers in logger.c, you might create 3 register functions, or just 1... |
| 03:19.52 | CoffeeKid | i think it'd make more sense to make 3 |
| 03:20.02 | CoffeeKid | whats the other one.. queue_log, log, and what else? |
| 03:20.14 | Corydon76-dig | It's perfectly conceivable that if somebody is creating an external logger, they'd want to handle all 3. |
| 03:20.32 | Corydon76-dig | and if the function pointer is NULL for 2, then there's only a single backend there... |
| 03:20.33 | CoffeeKid | thats true |
| 03:20.51 | CoffeeKid | so one makes more sense maybe? |
| 03:21.14 | Corydon76-dig | Third one is event log |
| 03:21.21 | Corydon76-dig | Sure |
| 03:22.36 | CoffeeKid | so, how would i go about modifying logger.c.. do i need to register myself as a contributer and such? |
| 03:22.55 | Corydon76-dig | No, you just need to file a license agreement |
| 03:23.05 | Corydon76-dig | which you can do online at http://bugs.digium.com |
| 03:23.34 | Corydon76-dig | once it's approved, you can file your patches on the bugtracker |
| 03:24.36 | CoffeeKid | once the code i do is approved? or i am approved to make changes to the code? |
| 03:25.11 | CoffeeKid | wait, my licence agreement is approved? |
| 03:25.17 | Corydon76-dig | No, it's checked by a lawyer to be sure that you have the right to make contributions |
| 03:26.22 | CoffeeKid | ah, do they check who i work for and see if they have my ballz in a vice? |
| 03:27.50 | Corydon76-dig | I don't know what all they do, I just know the procedure |
| 03:28.43 | CoffeeKid | probably would have to ask my employer if i can make changes eh? since i develop for a private company, but want to make changes to asterisk for free. |
| 03:29.26 | Corydon76-dig | If you create and test the patches on employer time and equipment, then it's a classic case of work-for-hire |
| 03:29.53 | Corydon76-dig | and your employer holds the copyright, and can thus decide how the code is contributed |
| 03:30.38 | Corydon76-dig | Of course, if you do it on your own time and equipment, unless you have a non-compete in place with your employer, the code is yours to do with as you wish |
| 03:30.41 | CoffeeKid | i see, thats very interesting. But if i did it on my own time, and my own equipment, with my employer knowing about it and not caring, its fair game? |
| 03:30.56 | tzanger | CoffeeKid: depends on your emploment contract and where you are located |
| 03:31.02 | tzanger | some states are more sane than others |
| 03:31.03 | Corydon76-dig | Right, unless you have a non-compete |
| 03:31.29 | Corydon76-dig | and even then, if you want to know the exact details, talk to a lawyer |
| 03:31.45 | tzanger | personaly I never gave a flying fuck about those kinds of contracts |
| 03:31.45 | CoffeeKid | heh, funny how things get so complicated so fast :) |
| 03:31.54 | Corydon76-dig | This is why those license agreements are handled by a lawyer... |
| 03:32.34 | CoffeeKid | makes sense. |
| 03:32.46 | CoffeeKid | damn the law anyways |
| 03:32.55 | tzanger | now I work for myself and any contracts I do I give two prices... one where once they sign off on it and the cheque clears, I erase the data from my system, and another where they give me a perpetual, transferrable, non-exclusive license to the code I wrote. |
| 03:33.07 | tzanger | one is naturally quite a bit pricier than the other |
| 03:33.31 | tzanger | funny thing is I get lots of people super scared about their precious, precious IP being stolen ... until they see the price tag |
| 03:33.37 | tzanger | then they're not nearly as concerned |
| 03:33.50 | CoffeeKid | hehe |
| 03:33.53 | Corydon76-dig | $250/hr vs $125/hr ? |
| 03:33.55 | CoffeeKid | yeah, money talks, i see it every day :) |
| 03:34.57 | CoffeeKid | so what if i start developing this logger.c stuff and i get stuck, i come here for help from you guys right? |
| 03:35.17 | tzanger | Corydon76-dig: the erase-my-work price is generally 4x the common-sense alternative |
| 03:35.26 | tzanger | I don't quote by hour, I quote by job |
| 03:35.45 | CoffeeKid | 4x less for the erase, right? |
| 03:35.50 | Corydon76-dig | tzanger: heh |
| 03:35.59 | tzanger | uh no, 4x the cost if they want me to forget I did the work |
| 03:36.05 | CoffeeKid | ahh |
| 03:37.09 | tzanger | file: I do, yes, but my cvs server won't if I ask it not to |
| 03:38.38 | CoffeeKid | thats interesting tzanger, what work do you generally do? custom phone stuff i'm guessing? |
| 03:38.56 | tzanger | I generally work on the weird embedded parts |
| 03:39.06 | tzanger | interfacing to custom PBX backends and stuff |
| 03:39.22 | CoffeeKid | ah, i see |
| 03:39.29 | tzanger | I'm currently trying to figure out a *nice* way to de-interleave 256 bits |
| 03:39.42 | tzanger | I have 32 bytes coming in that are arranged in an odd way |
| 03:40.06 | tzanger | byte 0 has bits 0,32,64,96,128,160,192,224 |
| 03:40.15 | tzanger | byte 1 has bits 1,33,65,97,129,161,193,225 |
| 03:40.16 | tzanger | ... |
| 03:40.36 | tzanger | byte 31 has bits 31,63,95,127,159,191,223,255 |
| 03:40.53 | tzanger | I need to arrange that so that byte 0 has bits 0-7, 1 8-15... in normal order |
| 03:40.57 | Corydon76-dig | Sounds like you're flipping a matrix |
| 03:41.33 | tzanger | Corydon76-dig: yeah, I always sucked at matrix math |
| 03:41.52 | Corydon76-dig | Me too... explains why I'm not a hardware engineer, eh? |
| 03:42.18 | CoffeeKid | heh, thats enough math to make your head spin :) |
| 03:42.24 | Corydon76-dig | file: masochist |
| 03:42.34 | CoffeeKid | lol |
| 03:42.45 | file | why Corydon, I didn't know you felt that way! |
| 03:43.05 | CoffeeKid | I enjoyed math until calculus, then i tried to forget everything except + - * / |
| 03:43.27 | Corydon76-dig | file: yes, you did... remember IP4IT? ;-) |
| 03:43.41 | file | I wasn't there! |
| 03:43.54 | Corydon76-dig | Oh, San Jose, then? |
| 03:44.03 | file | better. |
| 03:44.19 | Corydon76-dig | It all blurs together |
| 03:51.37 | CoffeeKid | Corydon76-dig, explain to me all the steps i need to go through to submit a patch :) |
| 03:52.39 | Corydon76-dig | Make it, test, submit a license agreement, submit the patch, suffer through the maintainers giving you criticism of the patch, making improvements, testing again, until it gets committed |
| 03:53.48 | CoffeeKid | and you submit the patch and licence agreement to bugs.digium.com ? |
| 03:55.16 | Corydon76-dig | Correct |
| 03:56.03 | CoffeeKid | and if i get stuck while developing? :) |
| 03:57.12 | Corydon76-dig | Here |
| 03:57.42 | CoffeeKid | great! Thanks tons for your help btw |
| 03:57.52 | Corydon76-dig | yw |
| 03:57.56 | CoffeeKid | i'm gonna work on wrapping my head around more of the asterisk code. |
| 04:03.39 | Corydon76-dig | Applications and dialplan functions are among the easiest to wrap your head around... just choose smaller modules to look at |
| 04:03.51 | Corydon76-dig | i.e. not app_queue, app_voicemail, or func_odbc |
| 04:04.24 | CoffeeKid | yeah, looked at app_voicemail once, then passed out. |
| 04:06.01 | CoffeeKid | i'm primary interested int he module loading so i can modify logger.c.. gonna probably base my code from cdr.c i think. |
| 04:09.29 | CoffeeKid | how do i checkout the svn trunk? whats the url? |
| 04:09.57 | Corydon76-dig | svn co http://svn.digium.com/svn/asterisk/trunk asterisk-trunk |
| 04:10.25 | CoffeeKid | thanks :) |
| 04:36.50 | *** join/#asterisk-dev CunningPike (n=CunningP@S010600095b33697f.vc.shawcable.net) |
| 04:55.24 | snuff-work | mm the doxy errors have grown since i last checked.. time to go huntin |
| 06:54.44 | *** join/#asterisk-dev citats (n=james@mrplow.gnuinternet.com) |
| 07:00.50 | *** join/#asterisk-dev CtRiX (n=CtRiX@aretha.navynet.it) |
| 07:06.01 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:37.27 | *** join/#asterisk-dev tzafrir (n=tzafrir@62.90.10.53) [NETSPLIT VICTIM] |
| 07:37.27 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 07:47.31 | *** join/#asterisk-dev pnlarsson (n=pnlarsso@c83-248-96-159.bredband.comhem.se) |
| 07:55.24 | *** join/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 08:09.54 | *** join/#asterisk-dev Juggie (n=Juggie@CPE001601df17fb-CM001a6682d7b6.cpe.net.cable.rogers.com) |
| 08:22.15 | *** join/#asterisk-dev _ys (n=yuri@236-069.nat.mns.ru) |
| 08:39.02 | *** join/#asterisk-dev denon (n=denon@tooth.decay.org) |
| 08:48.35 | *** join/#asterisk-dev qdk (n=qdk@85.235.253.139) |
| 09:08.11 | *** join/#asterisk-dev IgorG (n=FeedomPa@195.162.32.126) |
| 09:09.24 | IgorG | Is anyone here? |
| 09:09.24 | IgorG | Current trunk again is broken: |
| 09:09.35 | IgorG | server-voip:/usr/src/voip/asterisk-trunk/asterisk-89132.orig # make menuselect |
| 09:09.35 | IgorG | make: *** makeopts: Not a directory. Stop. |
| 09:09.35 | IgorG | make: *** [menuselect/makeopts] Error 2 |
| 09:15.23 | *** join/#asterisk-dev tzafrir (n=tzafrir@62.90.10.53) |
| 09:15.23 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 09:46.38 | IgorG | current trunk again is broken: |
| 09:47.08 | IgorG | after configure i've seen following: |
| 09:47.10 | IgorG | server-voip:/usr/src/voip/asterisk-trunk/asterisk-89132.orig # make |
| 09:47.10 | IgorG | make: *** makeopts: Not a directory. Stop. |
| 09:47.10 | IgorG | make: *** [menuselect/makeopts] Error 2 |
| 09:48.27 | tzafrir | IgorG, svn info menuselect |
| 09:48.31 | tzafrir | what rev is it? |
| 09:48.49 | IgorG | r89132 |
| 09:48.55 | IgorG | oh |
| 09:48.59 | IgorG | it is asteris rev |
| 09:49.02 | IgorG | one second |
| 09:49.21 | IgorG | 222 |
| 09:50.42 | _ys | Hm, chan_unistim.c compilation error. missed "#include <sys/stat.h>" |
| 09:50.58 | _ys | in trunk |
| 09:55.58 | *** join/#asterisk-dev IgorG (n=FeedomPa@195.162.32.126) |
| 10:48.07 | jmls | top of the morning to you all |
| 10:54.55 | *** join/#asterisk-dev mbranca (n=matteo@host-210-mi.linuxserver.it) |
| 10:59.14 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:01.44 | *** join/#asterisk-dev _ys (i=ys@91.151.196.254) |
| 11:19.37 | agx | is there some patch to avoid the call drop with X-lite+video sending 3 stupid RTP packet at the beggining? (rtp.c: Unknown RTP codec 126 received from) |
| 12:01.02 | mvanbaak | disable video in x-lite |
| 12:01.06 | mvanbaak | that worked for me |
| 12:04.41 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 12:05.08 | puzzled | hi |
| 12:07.35 | jmls | lo |
| 12:12.50 | *** join/#asterisk-dev dioedu (n=dioedu@201.7.117.114) |
| 12:15.13 | agx | mvanbaak, yes that's the problem :) unluckly i'm going to do some test with mobotix camera that has both SIP and ISDN capabilities... i'll try with Ekiga or some other video phone since its the camera sending a BYE |
| 12:18.22 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.106.70) |
| 12:36.03 | *** join/#asterisk-dev tzafrir (n=tzafrir@62.90.10.53) |
| 12:36.03 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 12:41.30 | *** join/#asterisk-dev qdk (n=qdk@0x573fe57e.bynqu2.broadband.tele.dk) |
| 13:10.55 | *** part/#asterisk-dev jmls (n=jmls@62.49.235.130) |
| 13:11.41 | *** join/#asterisk-dev jmls (n=jmls@62.49.235.130) |
| 13:39.23 | *** join/#asterisk-dev zerohalo (n=zeroHalo@pool-71-255-162-167.bstnma.east.verizon.net) |
| 14:16.03 | *** join/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 14:30.18 | *** join/#asterisk-dev ManxPower (n=manxpowe@242.sub-75-203-181.myvzw.com) |
| 14:31.10 | *** join/#asterisk-dev allankardec (n=root@201.50.67.92) |
| 14:48.52 | *** join/#asterisk-dev eliel (n=eliel@host172038.metrored.net.ar) |
| 14:57.16 | *** join/#asterisk-dev BadPacket (n=John@unaffiliated/badpacket) |
| 14:59.52 | *** join/#asterisk-dev Laureano (n=Laureano@200.59.172.38) |
| 15:02.15 | ManxPower | file: a flamethrower works better. |
| 15:02.25 | ManxPower | As does torturing the person that put it on the network. |
| 15:02.38 | ManxPower | Nobody Expects the File Inquisition! |
| 15:03.04 | file | jsmith... oh jsmith... where are you |
| 15:09.48 | *** join/#asterisk-dev miguel3239 (n=elguerom@h-68-167-124-171.cmbrmaor.covad.net) |
| 15:10.40 | *** join/#asterisk-dev jsmith (n=jsmith@68.246.137.171) |
| 15:10.40 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 15:11.54 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 15:11.54 | *** mode/#asterisk-dev [+o anthm] by ChanServ |
| 15:26.13 | eliel | Hello, cli.h is already including file linkedlist.h and in some modules we are duplicating the include asterisk/llinkedlists.h, if we remove those "duplicates" there will be a confusion on the code? or it is better to remove them? |
| 15:26.43 | *** join/#asterisk-dev fakhir (n=fakhir@unaffiliated/fakhir) |
| 15:33.38 | De_Mon | eliel do all the files that contain linkedlists.h also include cli.h? |
| 15:34.08 | De_Mon | (is it linkedlist.h or llinkedlists.h) |
| 15:34.15 | De_Mon | (or linkedlists.h) |
| 15:34.40 | eliel | De_Mon: Not all of them, but when cli.h is included we could avoid including linkedlists.h |
| 15:36.58 | eliel | Also utils.h include too many "standard" header files so we could avoid duplicating those includes |
| 15:37.05 | eliel | when utils.h is included |
| 15:54.55 | *** join/#asterisk-dev jsmith (n=jsmith@68.246.137.171) |
| 15:54.55 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 15:55.31 | *** join/#asterisk-dev zerohalo (n=zeroHalo@pool-71-255-162-167.bstnma.east.verizon.net) |
| 16:00.28 | *** join/#asterisk-dev kpfleming (i=kpflemin@nat/digium/x-ee9a72ebe0879471) |
| 16:00.28 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 16:01.30 | *** join/#asterisk-dev putnopvut (i=putnopvu@nat/digium/x-dc6f8ef3c3ead8a4) |
| 16:01.30 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 16:15.42 | Laureano | Hi there. I'm currently working in a patch for menuselect, to allow the definition of packages that could solve dependencies on different flavors of Linux. For example, to use cdr_addon_mysql you need to install libmysql on Debian. |
| 16:16.14 | Laureano | My idea is to alert the user in make menuselect how this dependency could be solved using packages. |
| 16:16.55 | Laureano | And then, the user should be able to run "ap-get, yum, or the flavor package manager" to compile a module. |
| 16:18.13 | Laureano | What approach should I use? Define new attributes on menuselect-tree for each module? defining new tags for systems flavor inside each module? etc. |
| 16:20.49 | Qwell | Laureano: distros change package names quite often |
| 16:23.42 | *** join/#asterisk-dev Corydon76-lap (i=Corydon7@pdpc/supporter/bronze/Corydon76-home) |
| 16:23.43 | *** mode/#asterisk-dev [+o Corydon76-lap] by ChanServ |
| 16:25.30 | Qwell | Laureano: perhaps that would be best achieved as something in the docs/ directory? |
| 16:31.16 | Laureano | Qwell, I will use that approach then. I will start writing a little doc with the modules and the possible packages to use. And the different flavors. Thanks for your help! |
| 16:32.02 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 16:36.24 | *** join/#asterisk-dev anonymouz333 (n=voce@201.19.106.70) |
| 16:43.09 | *** join/#asterisk-dev blitzrage (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 16:43.09 | *** mode/#asterisk-dev [+o blitzrage] by ChanServ |
| 16:53.02 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 16:53.02 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 17:23.43 | *** join/#asterisk-dev papucy (i=p@host129-51-dynamic.57-82-r.retail.telecomitalia.it) |
| 17:24.29 | *** part/#asterisk-dev papucy (i=p@host129-51-dynamic.57-82-r.retail.telecomitalia.it) |
| 17:29.04 | tzanger | Corydon76-lap: I got the interleave/deinterleave test cases doen |
| 17:29.12 | tzanger | I'm not happy with them, but they're done |
| 17:32.52 | Corydon76-lap | Did you do them in assembly for speed? ;-) |
| 17:33.44 | tzanger | haha no |
| 17:33.52 | tzanger | www.mixdown.ca/~andrew/tdm_sucks.c is the first cut of it |
| 17:34.01 | tzanger | I'm going to replace those condition checks with a programmed shift |
| 17:34.19 | Corydon76-lap | "Requested URL not found" |
| 17:34.25 | tzanger | oh |
| 17:34.28 | tzanger | tdm_sucks_andrew.c |
| 17:35.30 | Corydon76-lap | You could simplify that beginning part with a memset() |
| 17:35.34 | tzanger | yes I know |
| 17:35.49 | tzanger | this is just the test code |
| 17:35.58 | tzanger | the driver will have that memsetted |
| 17:36.00 | Corydon76-lap | and you unrolled loops for speed. ;-) |
| 17:36.11 | tzanger | not really, it's just how I thought it through |
| 17:36.20 | tzanger | I didn't want to hit the same byte over and over and over |
| 17:36.29 | tzanger | memsetted? memsat? |
| 17:36.31 | kpfleming | just compile with -froll-loops :-0 |
| 17:36.52 | tzanger | kpfleming: yeah... as I said I coudln't think it though without unrolling it once |
| 17:37.03 | tzanger | now honestly for 32 bytes I may just unroll the whole thing |
| 17:37.41 | tzanger | I'm gonna replace that |
| 17:37.43 | tzanger | dst[j + 0] |= ((src[i] & (1 << 0)) != 0) << bit |
| 17:37.59 | kpfleming | there is no need... the compiler will probably unroll it for you |
| 17:38.17 | tzanger | get rid of that test by just shifting to the proper position |
| 17:38.26 | Corydon76-lap | I suspect if you calculate each destination byte with one (rather long) line, the compiler should be able to optimize the assembly instructions quite a bit |
| 17:38.43 | tzanger | yeah |
| 17:38.54 | tzanger | I think ti may pick up on the repeated |= and help me out too |
| 17:39.33 | tzanger | I'm still working on the -fdo_what_i_say_not_what_i_do optimization directive though |
| 17:39.57 | tzanger | -fomit_bugs seems to be stalled too |
| 17:40.13 | Qwell | -fno_gcc_bugs |
| 17:40.41 | kpfleming | Qwell: silly.. use dashes... -fno-gcc-bugs |
| 17:41.12 | Qwell | int32_t dlen = 2147483647; if ((int32_t)(dlen + 8) > (int32_t)2147483647) |
| 17:41.13 | Qwell | <PROTECTED> |
| 17:41.19 | Corydon76-lap | tzanger: well, the fact that you're operating on a single byte (src cache) might also be the key |
| 17:41.28 | Qwell | ^ prints on gcc 4.1 |
| 17:41.56 | tzanger | Corydon76-lap: indeed... have to play with it and see what it outputs too |
| 17:42.08 | tzanger | I'm willing to bet there's a blackfin opcode that does this in one go :-) |
| 17:42.24 | tzanger | the dma engine on bf537 is pretty nice, it'll do op-level interleaving, but it won't interleave bits |
| 17:42.50 | kpfleming | but a small CPLD will :-) |
| 17:43.30 | tzanger | kpfleming: :-) unfortunately I don't get to dictate hardware |
| 17:43.47 | tzanger | they give me an ancient pbx and say "asteriskify this" |
| 17:46.44 | *** join/#asterisk-dev miguel3239 (n=elguerom@h-68-167-124-171.cmbrmaor.covad.net) |
| 17:52.32 | *** join/#asterisk-dev bmd (n=bmd@72.54.252.34) |
| 17:52.58 | bmd | putnopvut: you around? |
| 17:58.46 | eliel | Hello, I made a complete include's cleanup, to avoid including header files twice, I didn't added any include, I only remove the ones that weren't needed or duplicates feel free to close this issue if you think could be confusing... |
| 17:58.47 | eliel | M11219 |
| 17:58.48 | MuffinMan | [new] [Asterisk] Core-General 0011219: [patch] Avoid including not needed header files or already included. reported by eliel http://bugs.digium.com/view.php?id=11219 |
| 17:58.51 | putnopvut | bmd: yes I am. |
| 17:59.16 | bmd | oh, I sent you an email, I was just wondering how scalble MWI is when using IMAP on 1.4 |
| 17:59.27 | bmd | scalable even |
| 18:01.40 | putnopvut | I'm answering your e-mail right now... |
| 18:05.45 | tzanger | there |
| 18:05.49 | tzanger | got rid of the constant bit test |
| 18:05.50 | tzanger | dst[j + 28] |= (src[i] & (1 << 7)) >> 7 << bit |
| 18:05.57 | tzanger | there's gotta be a nicer way to do that than that though |
| 18:06.09 | tzanger | >> x << y I should be able to simplify |
| 18:06.17 | tzanger | I can't just say bit - 7 though |
| 18:06.49 | tzanger | because bit obviously ranges from 0-y7 |
| 18:12.23 | *** join/#asterisk-dev tzafrir (n=tzafrir@62.90.10.53) |
| 18:12.23 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 18:13.27 | jmls | M11189 |
| 18:13.28 | MuffinMan | [new] [Asterisk] Channels/chan_zap 0011189: chan_zap causing reset on E1 and eventually crashed asterisk reported by freon1 http://bugs.digium.com/view.php?id=11189 |
| 18:14.09 | jmls | I had something extremely similar happen on friday - but BT said that it was a fault in the exchange .. |
| 18:14.24 | jmls | cou was running @ 99% as well |
| 18:14.31 | jmls | cpu, that is .. |
| 18:16.15 | *** part/#asterisk-dev anonymouz333 (n=voce@201.19.106.70) |
| 18:16.51 | Corydon76-lap | jmls: that actually looks like a memory corruption error, for which I need to see the output of valgrind |
| 18:17.45 | *** join/#asterisk-dev triiiple (i=triple@pi.nxs.se) |
| 18:18.53 | jsmith | jmls: For what it's worth, I've been seeing something similar for close to a year now |
| 18:24.24 | jmls | BT replaced one their PRI cards in the exchange. Mind you, I also restarted asterisk. |
| 18:26.36 | tzafrir | regardless of that specific issue: is Cresl1n around? |
| 18:27.23 | *** join/#asterisk-dev joetester (n=patrick@216.191.34.13) |
| 18:28.05 | jsmith | I talked to him over the weekend regarding this bug |
| 18:28.10 | jsmith | But I haven't seen him online yet today |
| 18:37.31 | *** join/#asterisk-dev _ys (n=yuri@236-069.nat.mns.ru) |
| 18:38.33 | tzafrir | hmmm.... I can't just tell someone "please run zttest and report if it was successful" |
| 18:38.51 | tzafrir | You actually need to know what you're doing to get a decent report from it |
| 18:40.06 | tzafrir | so what do I need? a mode in which zttest will run for a limited number of cycles and give an OK/BAD answer |
| 18:40.21 | tzafrir | What exactly is "OK"? |
| 18:48.10 | _ys | Anyone can build trunk w/DEBUG_THREADLOCALS ? |
| 18:56.46 | kpfleming | _ys: it's broken, will fix |
| 18:58.04 | _ys | kpfleming missed main/cli.o in linkage |
| 18:58.39 | kpfleming | yeah, it's borked |
| 18:58.43 | kpfleming | have a meeting right now, will fix it later |
| 18:58.54 | kpfleming | it actually needs a copy of cli.c and other stuff |
| 19:03.15 | *** join/#asterisk-dev fakhir (n=fakhir@unaffiliated/fakhir) |
| 19:33.37 | *** join/#asterisk-dev agx (n=badpengu@81-174-46-174.dynamic.ngi.it) |
| 19:51.40 | De_Mon | grtrfgthy, Dictate has a play faster feature |
| 19:58.24 | De_Mon | wtf why is dictate creating raw files with file permissions 700 |
| 19:59.10 | *** join/#asterisk-dev Netgeeks (n=chris@204.11.231.198.static.etheric.net) |
| 20:02.35 | eliel | _ys: I think this will fix the compilation error |
| 20:02.37 | eliel | M11221 |
| 20:02.38 | MuffinMan | [new] [Asterisk] Core/BuildSystem 0011221: [patch] Error while linking with DEBUG_THREADLOCALS enabled reported by eliel http://bugs.digium.com/view.php?id=11221 |
| 20:03.23 | *** join/#asterisk-dev sergeus (n=asterisk@77.244.64.99) |
| 20:03.53 | kpfleming | this stuff is really a mess |
| 20:04.23 | kpfleming | none of that stuff should be necessary at all |
| 20:04.42 | kpfleming | but that is a quick fix, that is for sure |
| 20:05.00 | eliel | kpfeming: you are talking about clicompat? |
| 20:05.15 | kpfleming | yep |
| 20:05.43 | eliel | kpfleming: yes, you are right. |
| 20:06.14 | kpfleming | too much to change for now... the next time codefreeze is here in HSV we'll get this straightened out |
| 20:15.04 | *** join/#asterisk-dev VoicePulse (n=contact@unaffiliated/voicepulse) |
| 20:15.55 | kpfleming | no |
| 20:16.00 | kpfleming | don't mess with it |
| 20:16.29 | kpfleming | when you are here the next time, we need to figure out the *right* way to share all that code into utils/, without having to replicate all the stuff you don't actually need |
| 20:18.06 | kpfleming | 11 minutes and 13 seconds |
| 20:18.22 | mvanbaak | lol |
| 20:18.51 | *** join/#asterisk-dev flujan (n=flujan@200-160-115-020.static.spo.ctbc.com.br) |
| 20:18.54 | flujan | hi guys... |
| 20:19.04 | flujan | I am having seg faults with my asterisk. |
| 20:19.07 | kpfleming | the answer is no |
| 20:19.13 | flujan | I enabled the full log. |
| 20:19.19 | flujan | and started the process on a terminal. |
| 20:19.21 | flujan | http://pastie.caboo.se/116982 |
| 20:19.29 | flujan | please, could you check this out? |
| 20:19.51 | De_Mon | hrm... I thought voicemail was fixed so it used the directory umask for permissions, looks like 1.4 is still setting them manually. |
| 20:20.57 | mvanbaak | flujan: that's not a -dev question, please take it to #asterisk |
| 20:21.54 | jmls | first of all, SIP is disabled by the network provider |
| 20:21.57 | *** join/#asterisk-dev GreggB (n=GreggB@66.206.86.107) |
| 20:22.07 | jmls | and now I can't find a way of changing the jpeg compression |
| 20:22.09 | jsmith | jmls: Yuck... |
| 20:22.24 | jmls | 5MP = 700k file ... ha ! |
| 20:22.43 | jsmith | Right! |
| 20:22.58 | mvanbaak | use png ;) |
| 20:23.07 | jmls | oh ha bloody ha. |
| 20:23.08 | jmls | ;) |
| 20:27.30 | eliel | kpfleming: I am also getting compile error if MTX_PROFILE is enabled... |
| 20:27.57 | kpfleming | no surprise |
| 20:28.07 | kpfleming | it's a mess, and i refuse to spend any time fixing what is wrong with it |
| 20:28.22 | kpfleming | those utilities should not be built by default, that is the quick fix |
| 20:28.27 | kpfleming | very few people need them |
| 20:28.49 | Corydon76-lap | eliel: why are you trying to build them? |
| 20:29.09 | eliel | Corydon76-lap: because it is there :-) |
| 20:29.35 | eliel | Corydon76-lap: no other reason |
| 20:29.37 | Corydon76-lap | If you were trying to diagnose an actual issue, you'd have a case... |
| 20:30.19 | kpfleming | they build by default |
| 20:30.29 | kpfleming | aelparse, check_expr, hashtab, etc. |
| 20:30.50 | kpfleming | none of them should build by default, and they should refuse to build in the presence of debugging options like DEBUG_THREADS and such |
| 20:34.14 | kpfleming | all of which is overkill for the six people who will ever use those tools :-) |
| 20:34.17 | mvanbaak | rm -rf utils/ |
| 20:34.45 | codefreeze | uh, lots more than 6, I fear, but exactly how many... no way to tell |
| 20:35.14 | mvanbaak | put some 'call-home' function in it |
| 20:35.17 | mvanbaak | then you will know ;) |
| 20:39.56 | De_Mon | laugh... |
| 20:39.59 | De_Mon | pbx*CLI> dialplan show extension 1234@local |
| 20:39.59 | De_Mon | Usage: core show dialplan [exten@][context] Show dialplan |
| 20:39.59 | De_Mon | pbx*CLI> core show dialplan 1234@local |
| 20:39.59 | De_Mon | No such command 'core show' (type 'help' for help) |
| 20:40.29 | De_Mon | is core show dialplan supposed to exist? |
| 20:40.32 | Qwell | uhh, oops? |
| 20:40.44 | mvanbaak | ouch |
| 20:41.42 | eliel | ups "dialplan show" |
| 20:41.46 | De_Mon | Want me to open a bug and submit a patch or can you fix it for me qwell? It's supposed to be 'Usage: dialplan show [exten@][context] |
| 20:42.06 | Qwell | De_Mon: does it actually work? |
| 20:42.30 | De_Mon | dialplan show 1234 and 1234@local work, as mentioned above 'core show' does not exist |
| 20:42.40 | Qwell | okay |
| 20:42.41 | lmadsen | M8925 |
| 20:42.42 | MuffinMan | [assigned] [Asterisk] Documentation 0008925: [post-1.4] CLI command audit reported by blitzrage http://bugs.digium.com/view.php?id=8925 |
| 20:42.44 | Qwell | does the deprecated command work? |
| 20:42.51 | lmadsen | any command updates should go to the bug |
| 20:42.54 | De_Mon | whats the depriciated command? |
| 20:42.59 | lmadsen | (no need to open a new one) |
| 20:43.03 | Qwell | show dialplan 1234@context |
| 20:43.13 | Qwell | lmadsen: this one is just a typo. the command itself is correct. |
| 20:43.20 | mvanbaak | indeed |
| 20:43.25 | De_Mon | it doesn and doesn't mention its depriciated |
| 20:43.26 | lmadsen | gotcha (I didn't read enough of the scrollback) |
| 20:43.39 | Qwell | De_Mon: it only says it once. |
| 20:43.39 | De_Mon | I'll mention both of those in the above ticket |
| 20:43.48 | Qwell | De_Mon: already fixing. this is trivial |
| 20:44.04 | De_Mon | I've never used show dialplan, and it didn't say it was depriciated |
| 20:44.12 | eliel | it is not deprecated anu more, it is just a typo on the CLI ->command definition |
| 20:44.32 | Qwell | The 'show dialplan' command is deprecated and will be removed in a future release. Please use 'dialplan show' instead. |
| 20:44.44 | De_Mon | 1.4.13 doesnt say that |
| 20:44.52 | De_Mon | probably gotten fixed already |
| 20:44.53 | Qwell | just did for me :p |
| 20:45.06 | putnopvut | De_Mon: it'll only output that the first time you use the command. |
| 20:45.17 | De_Mon | shrug, maybe I used it long long ago and don't remember |
| 20:45.22 | Qwell | fixed |
| 20:45.39 | De_Mon | I thought those warnings showed up every time you used the command once upon a time |
| 20:45.45 | Qwell | nope, just the first |
| 20:46.43 | De_Mon | how does it keep track of commands its warned you about? I'm tempted to restart asterisk just to prove to myself it really does complain about 'show dialplan' |
| 20:50.19 | wedhorn | I've been busy for the last couple of months so haven't been contributing, but chan_skinny seems to have hit limbo. Any particular reasons? |
| 20:50.31 | Corydon76-lap | wedhorn: no specs |
| 20:50.55 | mvanbaak | wedhorn: busy ? |
| 20:51.00 | Corydon76-lap | wedhorn: the whole module is derived off of what can be reverse engineered |
| 20:51.30 | Corydon76-lap | Cisco doesn't give details on their proprietary protocol |
| 20:52.38 | wedhorn | Corydon76-lap: yeah, we've been doing that and the patches having been sitting there for a couple of months |
| 20:52.47 | wedhorn | mvanbaak: work, study, exams |
| 20:55.20 | *** join/#asterisk-dev tensai (n=tensai@joanna.zmonkey.org) |
| 20:55.28 | De_Mon | when I type something like 'dialplan add 1234' and go back to add 'ext' when I hit tab it autocompletes the word and replaces my 1 with ^@ is that my terminals fault or asterisks? |
| 21:01.00 | putnopvut | codefreeze: ping |
| 21:01.41 | putnopvut | I just upgraded to the latest trunk, and there's something wrong in pbx.c regarding finding extensions. |
| 21:02.08 | codefreeze | putnopvut: mmmm, sounds juicy. tell me more! |
| 21:02.11 | putnopvut | I haven't yet put my finger on what it is yet, but apparently ast_exists_extension is indicating that extensions which do exist don't. |
| 21:02.37 | putnopvut | I know you changed some matching code in trunk lately, so I figured I'd bring it to your attention. |
| 21:02.53 | codefreeze | Yes, I'm very interested. I need the test case. |
| 21:02.58 | putnopvut | What I'm doing is simply dialing an extension from a SIP phone. |
| 21:03.03 | putnopvut | That's the entire test case. |
| 21:03.39 | codefreeze | Well, send me the dialplan. I need that context, etc, and the number you dialed. |
| 21:03.46 | putnopvut | Sure, no problem. |
| 21:03.51 | putnopvut | I'll pastebin it. |
| 21:06.26 | putnopvut | http://pastebin.ca/771171 |
| 21:07.43 | putnopvut | I turned debug up higher in pbx.c but got not additional messages to appear. |
| 21:07.49 | codefreeze | That's the whole context? Just one extension? |
| 21:08.03 | putnopvut | Sorry, no. I didn't realize you wanted the entire dialplan. |
| 21:08.37 | codefreeze | Well, the whole of that context will do. I don't think I need ALL your contexts, but they wouldn't hurt to be included. |
| 21:10.03 | putnopvut | Okay, I updated the pastebin to have the entire context included. |
| 21:10.24 | putnopvut | Sorry I'm not being more helpful at this point. I need to look through your code changes so I can understand them better. |
| 21:11.05 | putnopvut | http://pastebin.ca/771177 |
| 21:16.12 | codefreeze | OK, got the data. I'll dive into it here in a minute. |
| 21:16.20 | putnopvut | I appreciate it. |
| 21:18.43 | mvanbaak | nooooooooooooooooooooo, dont remove encrypted rsa keys ! |
| 21:19.07 | Corydon76-lap | You use them? |
| 21:19.12 | mvanbaak | yup |
| 21:19.32 | Corydon76-lap | So you type in a passphrase for your keys everytime you restart Asterisk? |
| 21:19.37 | mvanbaak | yup |
| 21:19.44 | jsmith | Corydon76-lap: Lots of people do... |
| 21:19.50 | Corydon76-lap | What happens if Asterisk restarts in the middle of the night? |
| 21:19.52 | jsmith | Corydon76-lap: Especially for DUNDi keys |
| 21:20.03 | jsmith | Corydon76-lap: Then you re-init your keys |
| 21:20.04 | mvanbaak | Corydon76-lap: it means something is wrong and we fix it |
| 21:20.21 | Corydon76-lap | mvanbaak: so you get woken up so you can reinitialize your keys? |
| 21:20.29 | mvanbaak | no |
| 21:20.39 | mvanbaak | we have 4 boxen in a redundant setup |
| 21:21.18 | mvanbaak | dundi will know one box is gone, and disable routes to it |
| 21:21.28 | Corydon76-lap | mvanbaak: well, you've been appointed to figure out exactly WTF openssl is doing to encrypt the keys |
| 21:21.40 | kpfleming | Corydon76-lap: so we may need to provide a 'conversion' script then |
| 21:22.01 | mvanbaak | you serious about removing it completely ? |
| 21:22.03 | Corydon76-lap | We can encrypt them via some other method, I suppose |
| 21:22.25 | Corydon76-lap | mvanbaak: If I can't replicate the behavior of openssl without using openssl, yes. |
| 21:22.34 | mvanbaak | shit |
| 21:22.56 | mvanbaak | svn delete https://svn.vanbaak.info/asterisk/keys |
| 21:23.21 | Corydon76-lap | mvanbaak: I'm not saying get rid of RSA encryption... |
| 21:23.31 | mvanbaak | I encrypt them so I can put them in my svn |
| 21:23.33 | Corydon76-lap | just the encryption of the encryption keys |
| 21:23.42 | mvanbaak | ah |
| 21:23.46 | mvanbaak | use gnupg |
| 21:24.14 | Corydon76-lap | or more specifically, the undocumented way that openssl is encrypting keys |
| 21:24.42 | Corydon76-lap | mvanbaak: we're actually looking at code which isn't GPL |
| 21:25.05 | Corydon76-lap | public domain code, I think |
| 21:25.12 | *** part/#asterisk-dev agx (n=badpengu@81-174-46-174.dynamic.ngi.it) |
| 21:25.15 | mvanbaak | hhmm |
| 21:25.38 | kpfleming | yes, it is |
| 21:26.18 | Corydon76-lap | I've been working on figuring out openssl's encryption method for several weeks and it's time to either get a new pair of eyes on it or to throw in the towel |
| 21:26.18 | mvanbaak | ok, then I should look what it's doing |
| 21:27.43 | tzafrir | mvanbaak, so why not decrypt them upon checkout? |
| 21:29.13 | mvanbaak | that's possible, but then you have to make sure you have gnupg on all them boxen with the same secret keys and stuff |
| 21:29.21 | De_Mon | is there any particular reason why the Say series of commands only work in playback and not background mode? |
| 21:29.59 | Corydon76-lap | De_Mon: no reason, really |
| 21:30.18 | *** join/#asterisk-dev callguy (n=callguy@pool-71-255-162-167.bstnma.east.verizon.net) |
| 21:30.42 | Corydon76-lap | Difficulty of coding both in the same codespace, perhaps? |
| 21:30.55 | Qwell | it wouldn't even be all that difficult, really |
| 21:31.03 | mvanbaak | Corydon76-lap: can I get 2 or 3 days to look at it and stuff ? |
| 21:31.14 | Corydon76-lap | mvanbaak: you can have a week or two |
| 21:31.21 | De_Mon | I was pondering just switching them to background instead of playback and calling it a feature -- opposed to supporting both |
| 21:31.21 | mvanbaak | cool, thanks |
| 21:31.35 | Qwell | De_Mon: add a flag to the Say* apps |
| 21:31.43 | Corydon76-lap | mvanbaak: and even if I rip out passphrase support, nothing says it couldn't be readded later |
| 21:32.06 | mvanbaak | that's true |
| 21:32.24 | mvanbaak | I'll dive into it this week/weekend |
| 21:32.29 | Corydon76-lap | Might be motivation for the right person |
| 21:34.21 | mvanbaak | tzafrir: I'll get back to your mail tomorrow |
| 21:34.33 | mvanbaak | I have to run |
| 21:34.38 | mvanbaak | latero all |
| 21:34.45 | mvanbaak | Corydon76-lap: I'll get back to you when I know more |
| 21:35.05 | Corydon76-lap | ok |
| 21:35.16 | mvanbaak | latero all ! |
| 21:51.35 | tzanger | Corydon76-lap: got it |
| 21:51.43 | tzanger | http://www.mixdown.ca/~andrew/tdm_sucks.c |
| 21:51.53 | tzanger | that should be optimizable and is relatively fast on its own too |
| 21:55.07 | De_Mon | Qwell switching those apps to use background doesn't look as simple as I thought it'd be :( |
| 22:04.26 | Corydon76-lap | tzanger: you know what's funny about your code? |
| 22:04.51 | tzanger | Corydon76-lap: what's that? |
| 22:05.01 | Corydon76-lap | You're initializing src to 0, then asking if each bit is 1. |
| 22:05.59 | Corydon76-lap | Might want to load some sample data... ;-) |
| 22:06.02 | tzanger | ? |
| 22:06.03 | tzanger | I am |
| 22:06.52 | Corydon76-lap | No, I meant random data |
| 22:07.33 | tzanger | I am loading random data |
| 22:07.42 | tzanger | for(i = 0; i < 32; i++) { |
| 22:07.42 | tzanger | data[i] = random() % 255; |
| 22:07.42 | tzanger | a1[i] = a2[i] = 0; |
| 22:07.43 | tzanger | } |
| 22:08.17 | Corydon76-lap | Oh, nevermind, you changed your url |
| 22:10.29 | Corydon76-lap | I hit reload on the old one, figuring you had updated it |
| 22:10.39 | tzanger | you had it open that long? wow |
| 22:10.55 | Corydon76-lap | multiple screens, hidden Safari |
| 22:11.47 | file | eh? |
| 22:11.53 | file | I was on the Windows PC doing fax stuff :( |
| 22:14.07 | Deeewayne | the windows PC makes josh sad |
| 22:15.10 | file | a combination of faxing and Vista makes me sad |
| 22:15.23 | Deeewayne | does users.conf make you happy ? |
| 22:15.38 | putnopvut | everything saddens me |
| 22:15.51 | Corydon76-lap | What about chocolate? |
| 22:16.09 | Deeewayne | I noticed that chan_sip has a check --> if (ast_true(hassip) || (!hassip && genhassip)) |
| 22:16.16 | Corydon76-lap | Because we can arrange for chocolate to be delivered to your office |
| 22:16.19 | Deeewayne | that line ^^ makes me sad too |
| 22:16.23 | Qwell | Deeewayne: that looks...wrong |
| 22:16.52 | Deeewayne | especially if the general section includes hassip=yes, but a individual user section includes hasmisdn=yes |
| 22:17.19 | Deeewayne | the hasmisdn=yes user will be displayed in 'sip show peers' |
| 22:17.27 | putnopvut | chocolate doesn't sadden me. |
| 22:17.30 | Qwell | Deeewayne: eh? |
| 22:17.45 | Qwell | if genhassip, they should |
| 22:17.56 | Qwell | unless hassip=no in the user section |
| 22:18.05 | Deeewayne | can something be hassip=yes and hasblah=yes at the same time ? |
| 22:18.12 | Qwell | yep, that's the whole point :) |
| 22:18.26 | Qwell | that check still looks broken though |
| 22:18.40 | Qwell | hassip= would be...funky |
| 22:19.21 | Qwell | that !hassip should probably be an ast_strlen_zero |
| 22:19.33 | Deeewayne | ok, if that is the point, then maybe the if statement just needs some massaging |
| 22:19.44 | Qwell | or, you could make it like genhassip is |
| 22:19.53 | Qwell | hassip = ast_true(ast_variable_retrieve(ucfg, cat, "hassip")); |
| 22:20.03 | Qwell | same with registersip |
| 22:20.37 | Qwell | then you can drop the ast_true() from the check there. if (hassip || genhassip) |
| 22:20.42 | Deeewayne | but for digital detection, if hassip=yes is in the general section, all digital trunk sections will have to include hassip=no ? |
| 22:21.15 | Qwell | yeah, but nobody should really set hassip=yes in general unless they know what they're doing |
| 22:21.46 | Deeewayne | I think hassip=yes in general is default, let me check |
| 22:22.03 | Qwell | Deeewayne: it's a gui option |
| 22:22.05 | Qwell | I think |
| 22:23.00 | Deeewayne | hassip=yes is in configs/users.conf.sample |
| 22:23.04 | Qwell | yuck, chan_iax2 has the same code, basically.. that's just wrong |
| 22:23.07 | Deeewayne | yeah |
| 22:23.32 | Deeewayne | chan_iax2 was the next topic ;-) |
| 22:26.30 | codefreeze | putnopvut: I just put that context into mine, dialed, and got the exten fine.... |
| 22:27.05 | putnopvut | Hmmm...okay. |
| 22:27.16 | putnopvut | I'll hack away over here and see if I can't figure out what I've done. |
| 22:27.26 | codefreeze | you are on trunk? what version? |
| 22:27.32 | Qwell | so yeah, with the current code... if you set hasiax=yes in [general], and hasiax=<empty>, then...nothing created |
| 22:27.35 | putnopvut | latest checkout. |
| 22:27.39 | putnopvut | From about 5 minutes ago. |
| 22:28.06 | putnopvut | I do have some modifications to my branch, but they shouldn't affect extension matching. |
| 22:28.20 | putnopvut | I'm going to do a fresh checkout and see if the problem still happens. |
| 22:29.48 | Corydon76-lap | Hmmmmm... what shall I break next? |
| 22:29.55 | codefreeze | putnopvut: another trick: core set debug 10; dialplan show <extensionname> |
| 22:30.07 | putnopvut | All righty. I'll try it out. |
| 22:30.30 | codefreeze | It should show the match_char tree; sometimes, that'll point out some gross/subtle problem to fix. |
| 22:31.04 | Deeewayne | qwell: thanks! pari is going to make a change for digital users in the gui |
| 22:31.32 | putnopvut | codefreeze: thanks very much! I'll let you know if I discover anything odd |
| 22:31.51 | codefreeze | ok, thanks much, putnopvut |
| 22:34.44 | putnopvut | codefreeze: I'm not sure how to read this trie that is output...it doesn't really make sense. |
| 22:35.17 | putnopvut | http://pastebin.ca/771327 |
| 22:35.33 | putnopvut | That's the output I see when doing "dialplan show 666@internal" |
| 22:36.55 | codefreeze | try just "dialplan show internal" and see what it says |
| 22:38.15 | putnopvut | http://pastebin.ca/771334 |
| 22:38.23 | putnopvut | That's the output of "dialplan show internal" |
| 22:39.18 | putnopvut | Ah ha! Found it! |
| 22:39.56 | putnopvut | When I remove the "context => internal" from the general section of features.conf, everything works again. |
| 22:40.39 | putnopvut | That's something I've had for a long time though, and all of a sudden it stopped working... |
| 22:40.43 | codefreeze | Interesting. That's it. There are two "internal" contexts. I'm matching in the one that has the 700 (park). |
| 22:41.08 | putnopvut | Hmm, I guess I need to reread about features.conf then. |
| 22:42.05 | codefreeze | This kinda brings the ol' multiple-contexts-of-same-name to a head, because my hashtabs can only store a single ptr. |
| 22:43.12 | codefreeze | The short term solution is for you to use a different context name than "internal" -- you also might avoid "default" |
| 22:43.38 | putnopvut | Yeah, apparently I misunderstood the purpose of the context variable in features.conf |
| 22:44.07 | putnopvut | I thought I was pointing it to an already-defined context in extensions.conf, not creating a new context. |
| 22:44.46 | codefreeze | I know with parking, you use the include => parking or whatever to include it... |
| 22:47.47 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 22:47.48 | *** mode/#asterisk-dev [+o anthm] by ChanServ |
| 22:57.16 | snuff-work | mm could we move the callgroup/pickgroup to be 64 bits instead of 32.. |
| 22:58.23 | snuff-work | from what i looked at of the code it wouldn't be too hard to change, am i right? |
| 23:03.39 | Qwell | snuff-home: You're the Solaris guy, right? :D |
| 23:04.09 | snuff-work | yes.. at least the solaris compiling guy :) |
| 23:04.15 | Qwell | M11219 |
| 23:04.16 | MuffinMan | [new] [Asterisk] Core-General 0011219: [patch] Avoid including not needed header files or already included. reported by eliel http://bugs.digium.com/view.php?id=11219 |
| 23:04.20 | Qwell | Would you be willing to test that? |
| 23:04.32 | Qwell | call it a heads up before we commit it :P |
| 23:04.46 | Qwell | mvanbaak: you too :D |
| 23:04.56 | Qwell | on <some random BSD> |
| 23:05.40 | snuff-work | mmm lot of changes.. |
| 23:09.26 | snuff-work | i shall first test if the latest trunk compiles before hand ;) |
| 23:09.32 | Qwell | heh |
| 23:09.34 | Qwell | good idea |
| 23:10.45 | snuff-work | damn it.. those stupid util directories again |
| 23:14.32 | snuff-work | 2/3rds of that patch failed when i did patch -p0 < include.diff |
| 23:14.46 | Qwell | worked here... |
| 23:14.52 | Qwell | there was just some fuzz |
| 23:17.41 | snuff-work | http://www.pastebin.ca/771391 some of my errors |
| 23:18.46 | Qwell | yeah, that's just fuzz |
| 23:18.56 | Qwell | stupid solaris.. |
| 23:18.59 | Qwell | got gpatch? :p |
| 23:19.39 | snuff-work | hmm probably |
| 23:20.17 | snuff-work | worked little better |
| 23:20.53 | codefreeze | Tell ya what. I'm going to go thru this and calm this stuff down, ok? |
| 23:21.01 | snuff-work | still bout 30 failed hunks :) |
| 23:21.22 | codefreeze | the utils stuff, I mean |
| 23:21.35 | snuff-work | goo codefreeze :) |
| 23:36.36 | codefreeze | Hmmm. only found a prob with hashtest.c & the MTX_PROF stuff. Fixed. Any others? I turned on almost compile option. |
| 23:37.00 | codefreeze | s/almost/almost every/ |
| 23:44.51 | snuff-work | mm i have all options on cept debug thread on solaris |
| 23:48.47 | codefreeze | Well, that's true, I do have a solaris machine, but its in no shape for asterisk. |
| 23:50.53 | snuff-work | mmm.. its all good it relates to pthreads options etc.. |
| 23:51.00 | snuff-work | i dont mind atm |
| 23:54.06 | snuff-work | qwell some with type arg rather than param.. was to stop it complaining ;) |
| 23:57.44 | snuff-work | mmm 2 errors so far |