| 00:00.43 | *** join/#asterisk-dev mosty (n=mosty@ppp191-34.static.internode.on.net) |
| 00:02.29 | mosty | russellb: will your deadlock talk at astricon be a general intro to deadlocks, or will it be down and dirty details of common sources of deadlocks in asterisk? |
| 00:02.55 | russellb | It's a "lightning talk", so it will likely only be about 10 minutes |
| 00:03.28 | russellb | but I plan to cover both what deadlocks are from a high level, as well as areas of the code where they have most commonly occurred, and finally, what we have done to help |
| 00:03.43 | russellb | I've done that before in about 10 minutes for a group at the office :) |
| 00:04.44 | snuff-work | has murf done a 10min astobj2 talk ? ;) |
| 00:05.00 | mosty | i figured it would have to fairly high level |
| 00:05.04 | mosty | to be, even |
| 00:05.34 | russellb | nods |
| 00:06.06 | russellb | snuff-work: no, but I can cover astobj2 on the dev day if someone wants to hear about it |
| 00:06.38 | snuff-work | mm i would love to come.. but at $2000 AU just to get return flight.. |
| 00:08.44 | snuff-work | living down here does have advantages.. international cheap flights isn't one ;) |
| 00:08.52 | russellb | ha, I bet |
| 00:09.06 | russellb | I'd like to visit someday .. |
| 00:10.14 | snuff-work | its nice.. be a little on the expensive side for you americans these days with our dollar coming closer to parity with urs |
| 00:13.09 | russellb | I guess we need Astricon AU ... |
| 00:13.27 | russellb | Astricon Down Under :) |
| 00:13.35 | tzafrir_laptop | mvanbaak, what's wrong with using an LGPL lib in the asterisk core? Do you modify it? |
| 00:14.09 | russellb | tzafrir_laptop: nothing, legally. |
| 00:14.18 | russellb | tzafrir_laptop: it's just something that we (Digium) are trying to avoid |
| 00:14.32 | tzafrir_laptop | why? |
| 00:14.41 | *** join/#asterisk-dev mvanbaak_ (n=michiel@vanbaak.xs4all.nl) |
| 00:14.41 | *** mode/#asterisk-dev [+o mvanbaak_] by ChanServ |
| 00:15.35 | russellb | <PROTECTED> |
| 00:15.48 | tzafrir_laptop | And if there were? |
| 00:15.54 | tzafrir_laptop | What difference does it make? |
| 00:15.55 | russellb | <PROTECTED> |
| 00:16.37 | tzafrir_laptop | sounds like a certain lawyer is taking some wrong type of medications |
| 00:16.57 | tzafrir_laptop | in other words: WTF? |
| 00:17.36 | russellb | I don't expect you to care about our commercial customers. :) |
| 00:17.45 | russellb | but in this specific case, it's not like the LGPL option is better ... |
| 00:18.02 | tzafrir_laptop | I'm talking about the general case |
| 00:18.14 | tzafrir_laptop | Avoiding GPL in a GPLed project is one thing |
| 00:18.22 | tzafrir_laptop | Avoiding LGPL is much worse |
| 00:18.44 | russellb | well it's the potential difference of not having to distribute any source code, to having to set up the right things to distribute source code for at a minimum, some LGPL lib |
| 00:18.57 | russellb | it's not a _policy_ to avoid LGPL |
| 00:19.05 | russellb | in fact, we use a number of LGPL libs in Asterisk |
| 00:19.28 | russellb | but not in the asterisk core, or some other critical piece of functionality |
| 00:19.36 | russellb | it's on a evaluate on a case by case basis |
| 00:20.14 | russellb | we are by no means saying LGPL libs can not be used |
| 00:35.16 | tzanger | why wouldn't you include LGPL dependent-code in commercial products? LGPL does not require you release your code |
| 00:35.39 | tzanger | I don't understand what burden LGPL puts on commercial code |
| 00:35.44 | tzanger | GPL sure, but LGPL?? |
| 00:35.56 | russellb | it requires you to make the LGPL code available |
| 00:36.05 | russellb | which is a minor nuisance, if anything |
| 00:36.15 | tzanger | that doesn't seem overly problematic, as you just said |
| 00:36.26 | russellb | trivial if you're already making some amount of source available |
| 00:36.32 | tzanger | you've already got the setup to make code available :-) |
| 00:36.36 | russellb | but for a company that does not make any source available, it's a bit of a pain ... |
| 00:36.43 | tzanger | russellb: fair enough I suppose |
| 00:36.49 | russellb | that's the concern |
| 00:37.07 | russellb | so it's a "avoid if we can" type of thing, but not a requirement |
| 00:39.28 | tzanger | ... you're already hosting code though (asterisk, libpri, zaptel, etc.) |
| 00:39.40 | tzanger | anyway, I was just popping by, I saw the convo and said "??" |
| 00:40.44 | mosty | tzanger, those are all dual-licensed i believe... |
| 00:42.39 | tzanger | sure, what's wrong with adding the LGPL code that you use too is the question |
| 00:42.45 | tzanger | you're already providing source |
| 00:42.56 | tzanger | so you have a 3rdparty/ directory with the LGPL code in there |
| 00:43.04 | tzanger | and any patches you apply to them |
| 00:43.09 | tzanger | I could be oversimplifying though |
| 00:44.52 | mosty | tzanger, i think the point is that some commercial digium customers may not be using L/GPL code |
| 00:46.11 | tzanger | ahh, I think I get it now |
| 00:46.48 | tzanger | digium doesn't give a shit, but if they sold a commercial version of asterisk to a customer and it required LGPL code, that third party would have to provide the LGPL code |
| 00:46.54 | tzanger | and *that* is what they're trying to avoid |
| 00:47.26 | tzanger | I figured a "Our shit's based on Asterisk, which uses some LGPL code; Digium keeps that code available at http:..." |
| 00:47.30 | tzanger | woudl be sufficient |
| 00:48.31 | mosty | i remember some debian derived distributions had problems with that idea a few years back, i can't remember how it was resolved |
| 01:14.25 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 01:14.56 | eliel | bkruse: ping |
| 01:19.29 | bkruse | eliel: pong |
| 01:20.50 | eliel | bkruse: Why do you think ...doc_get_field() (currently in pbx.c) should be in xml.c? I think it is documentation specific function (knows the documentation xml structure), and xml.c is for general usage... |
| 01:21.14 | bkruse | eliel: I completely agree |
| 01:21.20 | bkruse | mvanbaak said that, I think it should be in pbx.c |
| 01:21.32 | eliel | ok |
| 01:21.53 | bkruse | generic documentation retrieval should be from xml.c, mvanbaak may have a reason, doubt it though :P |
| 01:22.30 | eliel | I will ask him |
| 01:22.42 | bkruse | eliel: Good work so far, I like it |
| 01:23.00 | bkruse | gotta go though, I will see you tomorrow, maybe tonight, and be able to actually start some programming of my own |
| 01:23.02 | bkruse | ttysoon |
| 01:23.15 | eliel | bye |
| 01:24.40 | snuff-work | bkruse: has the final xml spec for apps doc has been decided? |
| 01:59.00 | *** join/#asterisk-dev moy (n=moy@189.169.61.29) |
| 04:10.28 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.137) |
| 04:36.42 | *** join/#asterisk-dev philipp64 (n=chatzill@mail.redfish-solutions.com) |
| 05:57.32 | *** join/#asterisk-dev CunningPike (n=arodgers@d75-155-184-81.bchsia.telus.net) |
| 06:02.08 | *** join/#asterisk-dev fabled (n=fabled@xdsl-83-150-94-239.nebulazone.fi) |
| 06:09.25 | *** join/#asterisk-dev bkruse (n=bkruse@76.73.154.120) |
| 06:09.25 | *** mode/#asterisk-dev [+o bkruse] by ChanServ |
| 06:10.50 | mvanbaak | bkruse: ! |
| 06:10.57 | mvanbaak | quick, before I go to work |
| 06:11.10 | mvanbaak | I've been looking at the code again, and you guys are correct |
| 06:11.19 | mvanbaak | the _doc_get_field should not be in xml.c |
| 06:11.20 | mvanbaak | :) |
| 06:11.44 | mvanbaak | now I'm off to work |
| 06:11.46 | mvanbaak | latero |
| 06:51.47 | *** join/#asterisk-dev fabled (n=fabled@xdsl-83-150-94-239.nebulazone.fi) |
| 07:52.44 | *** join/#asterisk-dev mbranca (n=matteo@81.208.92.210) |
| 09:19.29 | *** join/#asterisk-dev mbranca (n=matteo@81.208.92.210) |
| 10:14.22 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 10:24.12 | *** join/#asterisk-dev friendly12345 (n=friendly@ppp59-167-89-100.lns2.mel6.internode.on.net) |
| 10:27.59 | *** join/#asterisk-dev _ys (i=yuri@91.151.196.254) |
| 10:53.29 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 11:37.24 | *** join/#asterisk-dev DagMoller (n=aguirre@unaffiliated/dagmoller) |
| 12:11.07 | *** join/#asterisk-dev russellb (n=russellb@asterisk/developer-and-stable-maintainer/drumkilla) |
| 12:11.07 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 12:32.20 | *** join/#asterisk-dev shido6 (n=shido6@209.114.208.192) |
| 12:34.24 | shido6 | what is PRETTY_FUNCTION ? |
| 12:34.54 | *** join/#asterisk-dev Rem| (n=remi@bg-fw2out.monmouth.com) |
| 12:40.39 | russellb | a preprocessor macro that gets replaced with the function that it is used in |
| 12:40.44 | ctooley | shido6, it does your hair and nails? |
| 12:41.06 | russellb | or that |
| 12:41.44 | ctooley | russellb, You said you got new edger a while back. Do you like it? If so, what kind is it? |
| 12:42.16 | russellb | It's decent enough, I guess. It was one of the cheaper ones; Black & Decker |
| 12:44.47 | russellb | gets the job done for me. |
| 12:44.56 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 12:44.56 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 12:45.09 | *** join/#asterisk-dev dioedu (n=dioedu@201.7.117.114) |
| 12:50.29 | ctooley | thx |
| 12:52.12 | russellb | np |
| 13:06.24 | shido6 | ? |
| 13:07.05 | shido6 | i get a reference to __PRETTY_FUNCTION__.2 () from gdb when I do a bt full |
| 13:07.09 | shido6 | on a core dump |
| 13:08.36 | russellb | neat |
| 13:11.03 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.95.130) |
| 13:22.15 | *** join/#asterisk-dev moy (n=moy@nat/ibm/x-c8fe5b3cb97b41f8) |
| 13:25.35 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 13:45.12 | russellb | tzafrir_laptop: thanks for the fix. I should have tested before committing. ;) |
| 13:55.18 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 13:59.46 | *** join/#asterisk-dev putnopvut (n=putnopvu@216.207.245.1) |
| 13:59.46 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 14:01.10 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 14:01.47 | *** join/#asterisk-dev mcargile (n=mcargile@mail.in-florida.net) |
| 14:10.27 | seanbright | i'm no afraid to say it... typedefs r0x0r |
| 14:10.32 | seanbright | s/no/not/ |
| 14:11.44 | russellb | seanbright: you would |
| 14:12.20 | russellb | but anyway, asterisk and kernel coding guidelines say otherwise. So, we shall eradicate (most of) them! |
| 14:12.25 | seanbright | i know i know |
| 14:12.34 | seanbright | kernel schmernel |
| 14:12.40 | seanbright | asterisk schmasterisk |
| 14:18.24 | mvanbaak | typedef schmypedef |
| 14:22.37 | *** join/#asterisk-dev mog (n=mog@c-68-62-219-86.hsd1.al.comcast.net) |
| 14:22.38 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 14:30.36 | *** join/#asterisk-dev eliel (n=eliel@201.234.250.101) |
| 14:51.38 | seanbright | mvanbaak schmvanbaak |
| 14:58.54 | viraptor | how would one go about debugging a random memory corruption in *? I get everything from function pointers becoming 0x72657375 (which is a string "user") to ->rtp being null... is there any standard procedure / compile option for additional memory protection? |
| 14:59.46 | *** join/#asterisk-dev joetester2 (n=chatzill@216.191.34.13) |
| 15:00.02 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) |
| 15:00.02 | *** mode/#asterisk-dev [+o jpeeler] by ChanServ |
| 15:01.13 | kpfleming | viraptor: you have to run in valgrind... it's about the only way |
| 15:01.23 | kpfleming | and even that won't catch everything |
| 15:02.06 | *** join/#asterisk-dev Corydon76-lap (n=Corydon7@pdpc/supporter/bronze/Corydon76-home) |
| 15:02.06 | *** mode/#asterisk-dev [+o Corydon76-lap] by ChanServ |
| 15:02.39 | viraptor | kpfleming: cheers - I'll try |
| 15:03.01 | kpfleming | good luck |
| 15:04.53 | Qwell | there's a valgrind.txt or something (it may have been renamed in trunk) in the doc/ dir |
| 15:05.03 | Qwell | should help a little bit |
| 15:11.18 | Qwell | directionality?! |
| 15:11.28 | Qwell | tzafrir_laptop: that's the word of the day right there :) |
| 15:11.56 | seanbright | http://en.wiktionary.org/wiki/directionality |
| 15:12.53 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 15:12.54 | putnopvut | "The alien space ship was headed straight at us, so I directionalificated it the other way." |
| 15:18.03 | *** join/#asterisk-dev Pazzo (n=ugelt@reserved-225136.rol.raiffeisen.net) |
| 15:21.50 | *** join/#asterisk-dev Pazzo (n=ugelt@reserved-225136.rol.raiffeisen.net) |
| 15:22.22 | tzafrir_laptop | Qwell, that's only because you're not used to bidiretionality http://en.wikipedia.org/wiki/Bi-directional_text |
| 15:28.26 | mvanbaak | I'm going home. Will be back later today |
| 15:41.38 | *** join/#asterisk-dev denon (n=denon@tooth.decay.org) |
| 15:55.59 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 16:00.52 | mvanbaak | back :) |
| 16:27.14 | *** join/#asterisk-dev sruffell (n=sruffell@conference/asterisk/x-e65bc78050246942) |
| 16:27.14 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 16:29.52 | seanbright | is there a way (without actually knowing the specs or cracking the box) if a machine has PCI vs. PCI express? |
| 16:30.01 | Qwell | maybe dmidecode |
| 16:31.41 | seanbright | that did the trick |
| 16:31.43 | seanbright | danke |
| 16:31.48 | kpfleming | the lspci output will probably tell you as well |
| 16:31.58 | kpfleming | if there are any PCI express bridges |
| 16:32.04 | kpfleming | there could still be PCI slots though |
| 16:32.20 | seanbright | oh yeah |
| 16:32.27 | seanbright | "PCI Express x4 Port 2" |
| 16:32.28 | seanbright | etc etc |
| 16:32.31 | kpfleming | right |
| 16:32.42 | mvanbaak | yup |
| 16:33.14 | mvanbaak | kpfleming: have a minute ? |
| 16:33.53 | kpfleming | yep, about to head to lunch though |
| 16:34.11 | mvanbaak | ah |
| 16:34.12 | mvanbaak | k |
| 16:41.25 | codefreeze-lap | Very interesting: http://mpt.net.nz/archive/2008/08/01/free-software-usability |
| 16:44.55 | *** join/#asterisk-dev jrlodden (i=user@marsbase.lodden.com) |
| 17:04.30 | anonymouz666 | quick question: |
| 17:04.44 | anonymouz666 | what option I should enable to have the 'core show locks' |
| 17:04.55 | Corydon76-lap | DEBUG_THREADS |
| 17:05.07 | anonymouz666 | thanks. |
| 17:18.42 | shido6 | what is __PRETTY_FUNCTION__.2 () ? |
| 17:19.02 | Qwell | shido6: a macro that makes log messages pretty |
| 17:19.08 | Qwell | it shows line number, filename, etc |
| 17:19.08 | Corydon76-lap | It's a macro that expands to the current function name |
| 17:19.14 | Qwell | erm, wait |
| 17:19.19 | Qwell | wrong one - yeah, that |
| 17:19.26 | shido6 | hrmm |
| 17:20.08 | Corydon76-lap | similar in function to __FILE__ and __LINE__ |
| 17:24.23 | codefreeze-lap | Corydon76-lap: might as well have been called __UGLY_FUNCTION__ |
| 17:33.58 | eliel | Qwell: __PRETTY_FUNCTION__ only current function name.. |
| 17:34.05 | anonymouz666 | DEBUG_THREADS has any impact to performance or something like that? My idea is to compile every setup with this turned on to grab 'core show locks' when deadlock occurs |
| 17:34.26 | russellb | it has a performance hit, yes |
| 17:34.34 | russellb | it's not _huge_, but it's there |
| 17:34.46 | russellb | unless you're on a heavily loaded multi-core machine, it shouldn't mess you up |
| 17:35.40 | jsmith | If you're on a heavily loaded multi-core machine, you're more likely to *need* it turned on, at least in my book :-) |
| 17:36.00 | russellb | figures. |
| 17:39.02 | shido6 | ooooooh |
| 17:39.05 | shido6 | core show locks? |
| 17:39.08 | shido6 | really |
| 17:39.36 | russellb | welcome to 12 months ago ^_^ |
| 17:39.41 | shido6 | thanks guys |
| 17:39.51 | putnopvut | core show locks is pretty awesome |
| 17:40.30 | shido6 | 8 core box with an average of 10 users between 8 - 5 wouldnt hurt with debug_threads, right? |
| 17:40.41 | shido6 | 8 - 10 simultaneous users |
| 17:41.08 | jsmith | shido6: No, not at all |
| 17:41.13 | shido6 | woo hoo |
| 17:53.30 | *** join/#asterisk-dev bkruse (n=bkruse@conference/asterisk/x-7e360e17277046c0) |
| 17:53.31 | *** mode/#asterisk-dev [+o bkruse] by ChanServ |
| 17:57.29 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 18:04.01 | *** part/#asterisk-dev jrlodden (i=user@marsbase.lodden.com) |
| 18:06.57 | *** join/#asterisk-dev joetester2 (n=chatzill@216.191.34.13) |
| 18:07.05 | eliel | There was a team branch created to continue working on CLI permissions: http://svn.digium.com/svn/asterisk/team/eliel/cli-permissions |
| 18:10.40 | eliel | everyone in invited to test it. |
| 18:15.49 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 19:06.19 | bkruse | mvanbaak: !!!!!!!! |
| 19:09.32 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 19:12.07 | anonymouz666 | before 'core show locks' how was the process to catch the deadlock? |
| 19:17.57 | xilun | voodoo incantations |
| 19:18.21 | *** join/#asterisk-dev eliel (n=eliel@201.234.250.101) |
| 19:18.22 | jsmith | anonymouz666: Before then, it was quite difficult to track down deadlocks |
| 19:20.56 | *** join/#asterisk-dev seanbright (n=sean@asterisk/contributor-and-bug-marshal/seanbright) |
| 19:20.56 | *** mode/#asterisk-dev [+o seanbright] by ChanServ |
| 19:22.22 | anonymouz666 | jsmith: http://lists.digium.com/pipermail/asterisk-dev/2003-November/002384.html |
| 19:23.28 | *** join/#asterisk-dev bird_of_Luck (i=melifaro@secured.by.ipfw.ru) |
| 19:25.04 | jsmith | anonymouz666: Like I said, it was quite difficult to do before the "core show locks" code was added |
| 19:25.49 | jsmith | anonymouz666: (I've been down that road before, and it's not as simple as that email makes it appear) |
| 19:46.53 | *** join/#asterisk-dev outtolunc (n=me@adsl-76-211-231-133.dsl.pltn13.sbcglobal.net) |
| 20:01.37 | mvanbaak | bkruse !!!!! |
| 20:05.02 | mvanbaak | kpfleming: ping |
| 20:19.52 | kpfleming | hi... been in the lab getting some new servers setup |
| 20:20.04 | mvanbaak | sure, no problem |
| 20:20.17 | mvanbaak | mind if I bug you in private one more time ? |
| 20:20.35 | kpfleming | nope, go ahead |
| 20:20.53 | file | bugging kpfleming is great fun |
| 20:21.18 | mvanbaak | lol |
| 20:34.49 | *** join/#asterisk-dev korihor (n=korihor@201.211.168.130) |
| 20:41.23 | *** part/#asterisk-dev mog (n=mog@c-68-62-219-86.hsd1.al.comcast.net) |
| 20:41.29 | mvanbaak | meh |
| 20:41.43 | mvanbaak | file: we'll see. will have to try that in Phoenix |
| 20:49.27 | anonymouz666 | [2008-08-04 17:48:40] WARNING[5065] asterisk.c: No more connections allowed |
| 20:49.31 | anonymouz666 | why that could happen? |
| 20:49.41 | anonymouz666 | bizarre |
| 20:49.41 | russellb | when you have too many connections |
| 20:49.47 | russellb | (asterisk -r) |
| 20:50.32 | anonymouz666 | I am the only logged in |
| 20:50.48 | anonymouz666 | seems something got stuck... |
| 20:51.50 | anonymouz666 | the PBX is still running |
| 20:52.06 | putnopvut | anonymouz666: ps -ef | grep rasterisk |
| 20:52.55 | anonymouz666 | nothing. |
| 20:52.58 | *** join/#asterisk-dev snuff-home (n=snuffy@60-240-117-7-nsw-pppoe.tpgi.com.au) |
| 20:54.27 | anonymouz666 | the first time I saw it, that's why I got surprised |
| 21:13.30 | *** join/#asterisk-dev linuxstb (n=linuxstb@rockbox/developer/linuxstb) |
| 21:16.41 | *** join/#asterisk-dev _Vile (n=vile@freeswitch/developer/vile) [NETSPLIT VICTIM] |
| 21:16.41 | *** join/#asterisk-dev Iamnacho (i=Iamnacho@ip68-103-153-140.ks.ok.cox.net) [NETSPLIT VICTIM] |
| 21:17.28 | *** join/#asterisk-dev moy (n=moy@nat/ibm/x-9820150d11bd4b80) |
| 21:26.46 | *** join/#asterisk-dev puzzled (n=patrick@puzzled.xs4all.nl) |
| 21:42.28 | *** join/#asterisk-dev _Vile (n=vile@freeswitch/developer/vile) [NETSPLIT VICTIM] |
| 21:42.28 | *** join/#asterisk-dev Iamnacho (i=Iamnacho@ip68-103-153-140.ks.ok.cox.net) [NETSPLIT VICTIM] |
| 21:42.42 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 22:18.13 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 22:19.41 | mvanbaak | hey eliel |
| 22:20.10 | mvanbaak | you were correct, that ast_xml_doc_get_field should be in main/pbx.c |
| 22:20.11 | eliel | mvanbaak: hello! |
| 22:20.17 | mvanbaak | I removed it from the TODO |
| 22:20.25 | eliel | ok |
| 22:20.44 | mvanbaak | reminded me to not add TODO items at 3AM |
| 22:20.56 | eliel | no problem! |
| 22:21.35 | mvanbaak | we have to fix the description stuff |
| 22:21.52 | eliel | I am trying to get the content from a description element... the <para> element is a special in some way?? |
| 22:22.26 | mvanbaak | the <para> stuff was added by jsmith to make it easier to integrate the documentation into docbook |
| 22:22.37 | mvanbaak | so they can use it verbatim in TFOT3 |
| 22:23.15 | mvanbaak | I think we should just find the <description> tag, grab the text, and loop all the <para> items |
| 22:23.31 | eliel | and variablelist |
| 22:23.37 | mvanbaak | yup |
| 22:23.39 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 22:23.51 | mvanbaak | maybe |
| 22:24.12 | mvanbaak | we can also add a 'Variables' section to the 'core show application foo' |
| 22:24.28 | mvanbaak | Synopsis, Description, Variables |
| 22:24.43 | eliel | yes, that is better |
| 22:24.47 | mvanbaak | yup |
| 22:24.49 | mvanbaak | more clear |
| 22:26.03 | eliel | app_dial is the main example? |
| 22:26.34 | mvanbaak | that's the one that's totally converted |
| 22:27.11 | eliel | ok |
| 22:38.39 | *** join/#asterisk-dev jameswf-home (n=james@ip72-223-0-183.ph.ph.cox.net) |
| 22:41.46 | eliel | instead of free() xmlFree must be used... |
| 22:42.13 | mvanbaak | yup |
| 22:42.13 | eliel | I will submit a patch when finishing description parser.. |
| 22:42.25 | mvanbaak | I noticed that yesterday |
| 22:42.40 | mvanbaak | some functions return a casted xmlnode<somethnig> |
| 22:42.55 | mvanbaak | docs on libxml2 page say you have to xmlFree them |
| 22:43.10 | mvanbaak | I think I put that on this channel as well |
| 22:43.32 | mvanbaak | 01:21 <@ mvanbaak> eliel: I like what you have done :) |
| 22:43.36 | mvanbaak | 01:22 <@ mvanbaak> some code guideline misses |
| 22:43.39 | mvanbaak | 01:23 <@ mvanbaak> and some memory leaks |
| 22:43.50 | mvanbaak | 01:08 <@ mvanbaak> and it returns a casted xmlNodeListGetString |
| 22:43.53 | mvanbaak | 01:08 <@ mvanbaak> and the docs say: |
| 22:43.55 | mvanbaak | 01:08 <@ mvanbaak> Because xmlNodeListGetString allocates memory for the string it returns, you must use xmlFree to free it. |
| 22:43.58 | mvanbaak | 01:08 <@ mvanbaak> I cant find that one anywhere |
| 22:44.32 | eliel | I miss that.. :-) |
| 22:45.09 | mvanbaak | dont worry |
| 22:45.40 | mvanbaak | 01:05 <@ mvanbaak> there's some memory leaks in it |
| 22:45.45 | mvanbaak | 01:07 <@ russellb> memory management is overrated |
| 22:45.53 | mvanbaak | 01:06 <@ mvanbaak> we can work them out |
| 22:45.57 | mvanbaak | 01:08 <@ mvanbaak> russellb: amen |
| 22:47.00 | Corydon76-lap | We don't need no steenkin' memory management |
| 22:47.20 | mvanbaak | whehehehehehe |
| 22:47.43 | mvanbaak | let the GC handle that |
| 22:53.22 | *** join/#asterisk-dev snuff-work (n=snuff-wo@203.176.96.250) |
| 22:54.28 | mvanbaak | hey snuff-work |
| 22:54.37 | snuff-work | hello :) |
| 22:55.20 | snuff-work | whats the standing on docxml.. has the final dtd been decided? |
| 22:55.53 | mvanbaak | snuff-work: there's a pretty rc like file with the format |
| 22:56.02 | mvanbaak | jsmith did the last work on that |
| 22:56.12 | mvanbaak | it's not a dtd, but a description of the format |
| 22:56.42 | snuff-work | ah.. either way let me know when u want some of the old desc ported.. i'll do some in spare time |
| 22:57.13 | mvanbaak | snuff-work: eliel is working an libxml2 implementation |
| 22:57.40 | mvanbaak | app_dial.c stuff is altered by jsmith to be compatible with the docbook used for TFOT |
| 22:57.56 | snuff-work | ah k |
| 22:58.00 | mvanbaak | once that's done, and we have some working stuff, we'll ask the -dev list for comments |
| 22:58.10 | mvanbaak | if that's ok, we'll do the big converting |
| 22:58.21 | mvanbaak | no use converting stuff now and do it all over again |
| 22:58.32 | mvanbaak | app_dial has been done 3 times now already |
| 22:58.53 | mvanbaak | hhmm |
| 22:58.55 | snuff-work | i feel for jsmith then.. lot of work in app_dial |
| 22:59.46 | mvanbaak | maybe a post to the -dev list to ask to review app_dial because almost all feedback on earlier probes has been integrated |
| 23:00.11 | mvanbaak | he did it while sitting on a plain |
| 23:01.03 | snuff-work | not a bad time to do it.. specially when ur laptop has a decent battery unlike mine ;) |
| 23:01.44 | snuff-work | its now decided 15minutes is long enough |
| 23:02.00 | mvanbaak | snuff-work: he had a power inlet on his planeseat |
| 23:02.18 | snuff-work | oh nice |
| 23:02.35 | mvanbaak | I hope lufthansa has that as well |
| 23:09.36 | eliel | All the text will be within a <para> tag? |
| 23:09.47 | mvanbaak | yeah |
| 23:09.52 | mvanbaak | can be more then one |
| 23:10.08 | mvanbaak | if the description spans more then one paragraph |
| 23:10.30 | eliel | perfect |
| 23:10.58 | mvanbaak | we define the format |
| 23:11.24 | mvanbaak | so if it's '<para> is mandatory' it's like that |
| 23:15.56 | eliel | ok, I make <para> mandatory. |
| 23:16.21 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 23:16.31 | mvanbaak | nice |
| 23:16.41 | mvanbaak | that's how it should be |
| 23:16.49 | mvanbaak | according to jsmiths input |
| 23:16.53 | eliel | a bit of formatting will be needed inside it.. |
| 23:17.00 | mvanbaak | yeah |
| 23:17.07 | mvanbaak | same as Synopsis |
| 23:17.32 | tzafrir_laptop | well, aparantly the debian maintainer of tftpd decided to fix the bug with the cisco phones |
| 23:17.37 | mvanbaak | it has too many tabs in front of it in the 'core show application foo' stuff |
| 23:17.45 | tzafrir_laptop | Who cares if upstream's dead? |
| 23:17.49 | mvanbaak | tzafrir_laptop: eh ? |
| 23:18.19 | mvanbaak | oh wait |
| 23:18.31 | mvanbaak | nah |
| 23:18.35 | mvanbaak | tzafrir_laptop: eh ? |
| 23:19.06 | eliel | mvanbaak: yes, I don't know which is the best way to cleanup that. |
| 23:19.11 | eliel | I will commit my code. |
| 23:19.13 | tzafrir_laptop | http://lists.digium.com/pipermail/asterisk-users/2008-August/216217.html (or rather: the message it referred to) |
| 23:20.19 | mvanbaak | eliel: I think the xml parser should take care of it |
| 23:20.26 | mvanbaak | eliel: it should strip leading tabs |
| 23:20.46 | Qwell | tzafrir_laptop: interesting |
| 23:20.53 | eliel | mvanbaak: an open parameter out there striping the text? |
| 23:20.56 | eliel | maybe.. |
| 23:21.01 | Qwell | tzafrir_laptop: any idea what the actual change was? |
| 23:21.13 | Qwell | (or, if it was actually tested? :D) |
| 23:23.24 | mvanbaak | nice |
| 23:23.41 | mvanbaak | still not going to change my OpenBSD setup to Debian though ;) |
| 23:24.07 | eliel | done |
| 23:24.50 | mvanbaak | thank you |
| 23:24.56 | mvanbaak | checking out now |
| 23:27.42 | mvanbaak | thank you |
| 23:34.07 | eliel | len + strlen(tmptext) + 2 |
| 23:34.10 | eliel | not +1 |
| 23:34.12 | eliel | :-( |
| 23:34.45 | eliel | I added the \n after every <para> and I miss that byte.. |
| 23:56.07 | mvanbaak | off to bed here |
| 23:56.16 | mvanbaak | see you tomorrow |
| 23:56.18 | snuff-work | night mvanbaak |
| 23:56.26 | mvanbaak | latero |