IRC log for #asterisk-dev on 20110803

06:18.32*** join/#asterisk-dev infobot (~infobot@rikers.org)
06:18.32*** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- http://xkcd.com/303/
06:23.57*** join/#asterisk-dev schmidts (~schmidts@Lo20-rt01.lm33.sil.at)
06:24.00schmidtsgood morning
06:24.56*** join/#asterisk-dev Juggie (~Juggie@unaffiliated/juggie)
06:33.11*** join/#asterisk-dev asterisk-Tester (~RAMYT@210.5.215.40)
06:38.50*** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:4c10:9c6c:b679:266a)
06:41.47*** join/#asterisk-dev asterisk-Tester (~RAMYT@210.5.215.40)
06:47.42*** join/#asterisk-dev asterisk-Tester (~RAMYT@210.5.215.40)
06:59.05*** join/#asterisk-dev irroot (~irroot@pbx.distrotech.co.za)
06:59.24*** join/#asterisk-dev asterisk-Tester (~RAMYT@210.5.215.40)
07:09.09*** join/#asterisk-dev asterisk-Tester (~RAMYT@210.5.215.40)
07:19.29*** join/#asterisk-dev asterisk-Tester (~RAMYT@210.5.215.40)
08:01.23*** join/#asterisk-dev tzafrir_laptop (~tzafrir@local.xorcom.com)
08:01.23*** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ
09:35.56*** join/#asterisk-dev Ulrar (~quassel@2a01:e0b:1:136:62eb:69ff:fe8f:18a0)
11:03.45*** join/#asterisk-dev irroot (~irroot@pbx.distrotech.co.za)
11:39.55*** join/#asterisk-dev razu (~razu@2001:ad0:1:1:202:b3ff:fe36:921c)
12:04.13*** join/#asterisk-dev survery (~thomas@dslb-188-110-169-245.pools.arcor-ip.net)
12:04.22*** part/#asterisk-dev survery (~thomas@dslb-188-110-169-245.pools.arcor-ip.net)
12:17.30*** join/#asterisk-dev t_dot_zilla (~t_dot_zil@firewall-a.buf.ny.i-evolve.net)
12:46.24*** join/#asterisk-dev seraphie (~erin@75.76.38.159)
12:50.04*** join/#asterisk-dev moy (~moy@173.239.155.74)
13:36.30*** join/#asterisk-dev pabelanger (~pabelange@206-248-185-146.dsl.teksavvy.com)
13:36.30*** mode/#asterisk-dev [+o pabelanger] by ChanServ
13:57.02*** join/#asterisk-dev Deeewayne (~dwayne@75-150-14-49-Atlanta.hfc.comcastbusiness.net)
13:57.02*** mode/#asterisk-dev [+o Deeewayne] by ChanServ
14:13.14*** join/#asterisk-dev jdoe (jdoe@falseprophet.ca)
14:17.59russellbacidfoo: they're all wrappers around pthread_create.  background = smaller stack size (64k IIRC), detached = no pthread_join() needed.
14:18.26chazzamits... a russellb!
14:18.36russellbindeed
14:18.57chazzamwhen do you actually move back? or have you done that already?
14:19.16*** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg)
14:19.33russellbchazzam: i got to my new house yesterday
14:19.50russellbi'm now in "my life is in a boxes everywhere" hell
14:19.55chazzamoh nice! like the new place lots then?
14:19.57chazzamheh
14:20.09irrootlol there russellb
14:20.13chazzamjust make sure to make a fort out of them once you get all the life out
14:20.19chazzamfor at least an evening
14:20.20russellbgood call
14:20.38russellbi should do that while my wife is running errands
14:20.45chazzamyes!
14:21.00russellband then demand the password before i let her in to the house since my fort will be blocking the entrance
14:21.07chazzamthen when she comes back frantically urge her to get inside the walls?
14:21.18russellbanother good idea
14:21.51chazzammine might disrupt her enough to get her to play along, and maaaaybe not set off any anger hotspots
14:21.54chazzam=p
14:22.22chazzamthe barring entrance without a password is tricky when they're upset =/
14:23.51acidfoorussellb, ok thank you
14:24.27russellbacidfoo: "background" is a terrible name for what it is
14:24.41russellbit gets confused withd etached ...
14:25.16acidfoo;)
14:27.15acidfoorussellb, so detached_background is a small stack and no pthread_join()
14:27.58russellbcorrect
14:33.32*** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel)
14:33.32*** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ
14:39.50*** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell)
14:39.50*** mode/#asterisk-dev [+o sruffell] by ChanServ
14:54.45*** join/#asterisk-dev otherwiseguy (~terry@69.76.217.40)
15:35.00*** join/#asterisk-dev philipp64|laptop (~chatzilla@50-39-153-236.bvtn.or.frontiernet.net)
16:03.38Kobaz11:54 <beek> p3nguin: russellb now works for Red Hat
16:03.39Kobaz???
16:04.06Qwellhttp://blogs.digium.com/2011/07/18/russell-bryant-thank-you-for-seven-years-of-excellence/
16:04.12Kobazoh noes
16:08.31*** join/#asterisk-dev philippel_mac (~p_lindhei@c-67-160-11-168.hsd1.wa.comcast.net)
16:21.57JunK-Ylo guys!
16:23.07jsmithHey, it's philippel_mac
16:23.41philippel_machey jsmith  how goes it?
16:23.52jsmithCrazy busy, but good
16:24.03philippel_macalso - any chance you have any isight into a counterpart of IMPORT
16:24.18philippel_maclike a MASTER_CHANNEL() that let's you specify the channel?
16:24.44jsmithNope, sorry -- I haven't used IMPORT
16:25.14philippel_macIMPORT() or ImportVar just let you get variables from a specified channel
16:25.38jsmithI know what they do -- I just haven't used them
16:25.42philippel_macI need to set some and SHARED() is not a desirable option given some other dialplan complexity plus one of the them in CDR()
16:26.06philippel_machmm - will take a peek at trunk to see if anyone has done anything
16:26.26philippel_macand may have to dust off my C coding and add it and have yet another patch to our distro...
16:27.57JunK-Yphilippel_mac: what are you trying to do exactly?
16:29.52philippel_macwell - briefly
16:30.08philippel_macimplementing our own "one-touch" recording in freepbx
16:30.21philippel_macand given a bunch of recording policies, file naming, etc.
16:30.36philippel_macI have our own automixmon basically
16:30.54philippel_macand it currenlty uses MASTER_CHANNEL() for a bunch of bookeeping between the two sides
16:31.00philippel_macso it all stays in one place
16:31.09philippel_macincluding updating CDR() fields (our own)
16:31.25philippel_macbut when you do a park and then retreive
16:31.31philippel_macboth are their own master channel
16:31.45philippel_macso I can get the variables from the other channel with IMPORT() but can't set them
16:32.09philippel_macmoving to a SHARED() would add a lot of complexity and it would not address the CDR() issue
16:32.59philippel_macJunK-Y:  so ... I need an EXPORT() so to sepak, the counter part of IMPORT() or ImportVar()
16:33.25philippel_macconceptually the same as MASTER_CHANNEL() but where the other channel can be specified
16:37.48JunK-Ysomething like MASTER_CHANNEL(chan->name) and that would returns the master channel of chan->name ?
16:38.17philippel_macnot quite
16:38.33philippel_macI have the channel, because I have the bridged peer information, so it's more like:
16:38.49philippel_macMASTER_CHANNEL(varname,optinal_channel)
16:39.10philippel_mactoday there is just varname
16:39.21philippel_macand if it is the master channel, then it uses it's own var
16:40.16JunK-Yu want it for trunk ?
16:40.38JunK-Ybrb meeting
16:40.42philippel_macI'll use it in 1.8, would prefer it if already existed
16:40.46philippel_macthen I would back port it
16:44.48Kobazshould you lock a channel before grabbing its datastores?
16:44.52Kobazor is a ref good enough
16:45.46philippel_machmm and it looks pretty "straight forward" if I were regularly coding Asterisk function looking at MASTER_CHANNEL() it simply checks if there is a linked id and if so uses that, if not uses itself
16:46.14philippel_macso it would simply be getting the name of the passed in channel in a similar function and doing the same?
17:01.18Kobazis there a wiki page on "when to lock a channel" ?
17:04.11pabelangerKobaz: don't believe so, but sounds like an excellent idea!
17:07.23Kobaz^^ leifmadsen
17:08.41leifmadsenKobaz: yes?
17:09.44Kobazsounds like a good project
17:10.09Kobazit would help elimimate extra locking, and maybe deadlock issues
17:10.38Kobazbecause I have honestly no idea... i just lock the channel when i want to do something, but i'm sure there's situations where you don't need to lock and just need to hang on to the ref
17:22.47*** join/#asterisk-dev philipp64|laptop (~chatzilla@50-39-153-236.bvtn.or.frontiernet.net)
17:40.24Kobazhmm
17:40.48Kobazhttp://pastebin.com/SMke8jNc
17:42.55acidfoowould be nice to have parameters informations
17:43.04Kobazyeah, that's all i have
17:43.25Kobazvalgrind doesn't output params
17:55.41tzafrir_laptopsruffell, any idea when dahdi 2.5 is expected?
17:56.34sruffellthis Friday we'll probably tag the final….(so I guess that means I need to draft the announcement and get it to you)
17:56.53tzafrir_laptopright
18:15.24leifmadsenKobaz: sure, sounds like a good idea :)
18:28.29*** join/#asterisk-dev survery (~thomas@dslb-178-007-116-126.pools.arcor-ip.net)
18:28.33*** part/#asterisk-dev survery (~thomas@dslb-178-007-116-126.pools.arcor-ip.net)
18:48.54*** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming)
18:48.54*** mode/#asterisk-dev [+o kpfleming] by ChanServ
18:58.10acidfooany plan to support epoll() instead of poll() ?
18:59.02kpflemingin what?
18:59.47acidfoofrom what I understand, ast_poll() is wrapping poll()
19:01.32acidfooI was just trying to figure out in my head what was the best way to handle more than one tcp connection in a channel, actually chan skinny is starting a new thread everytime a device connect to it.
19:02.28acidfoowhat I like with epoll() is that you have an extra pointer to attach a structure.. so when you have a fd that is read to read/write you can extract the corresponding session
19:09.34*** join/#asterisk-dev acidfoo (~nib@modemcable094.94-70-69.static.videotron.ca)
19:09.52acidfoomm internet went down...
19:23.59tzafrir_laptopBTW: the command I used for the output in the menuselect review update was:
19:24.03tzafrir_laptopfor chr
19:24.03tzafrir_laptopoot in centos4 centos5 centos6 sarge etch lenny squeeze wheezy; do  schroot -c $chroot -- bash -c "./configure >/dev/null 2>&1 || exit; echo -n \"$chroot: \"; grep HAVE_ATTRIBUTE_weakref autoconfig.h"; done
19:24.26tzafrir_laptopDid I mention I like schroot?
19:40.21Kobazaw, there's no get_channel_by_regex
19:45.46seanbrightcorrect
19:52.13Qwellyet
19:54.32kpflemingacidfoo: have you grepped the source tree for epoll? some time ago file did some work to add support for it, but it's not enabled by default
20:16.53*** join/#asterisk-dev tamiel (~tamiel@ip-183.net-81-220-86.toulouse.rev.numericable.fr)
20:20.06pabelangerCorydon76-home: finally tried your patch for Mac OSX
20:20.15pabelangerbut ran into an issue when doing $ make samples
20:20.15pabelangerhttp://pastebin.com/xHPmgCc3
21:16.50*** join/#asterisk-dev tzafrir_laptop (~tzafrir@bzq-218-155-146.cablep.bezeqint.net)
21:16.50*** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ
21:17.36*** join/#asterisk-dev Corydon76-home (green@c-69-137-80-31.hsd1.tn.comcast.net)
21:17.36*** mode/#asterisk-dev [+o Corydon76-home] by ChanServ
21:21.20*** join/#asterisk-dev chazzam (~chazz@173-24-239-247.client.mchsi.com)
21:38.55*** join/#asterisk-dev luckman212 (~irc@2001:470:1f07:1225:2d40:d31d:26e0:afd3)
22:11.15*** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm)
22:12.30philippel_macquestion relating to MASTER_CHANNEL() being albe to  read/write to the linked master channel both variables and functions (if there is a master channel)
22:13.06philippel_macif one was to write a new function, or extend MASTER_CHANNEL() such that you could specify any target channel that is active vs. just the master channel
22:13.24philippel_macis there anything fundamentally different between that vs. just using the master channel?
22:13.59philippel_macin the case of MASTER_CHANNEL() I see that the code just checks if there is a linked channel, where as in this new scenario
22:14.49philippel_macit would have to lookup and get the channel pointer from the channel name but otherwise it seems like it would be no more "dangerous" then MASTER_CHANNEL() already is?
22:15.06philippel_macam I missing something???
22:25.57*** join/#asterisk-dev seraphie (~erin@75.76.38.159)
22:27.07*** join/#asterisk-dev seraphie (~erin@75.76.38.159)
23:43.26*** join/#asterisk-dev otherwiseguy (~terry@69.76.217.40)
23:54.46*** join/#asterisk-dev pabelanger (~pabelange@2607:f2c0:a000:166:beae:c5ff:fe3e:b315)
23:54.46*** mode/#asterisk-dev [+o pabelanger] by ChanServ

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.