| 05:30.19 | *** join/#asterisk-dev infobot (ibot@c-174-52-60-165.hsd1.ut.comcast.net) |
| 05:30.19 | *** topic/#asterisk-dev is Take the March 2019 Asterisk User Survey! https://goo.gl/forms/xL1VUHRsf95saly13 -- Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Check out our blog! blogs.asterisk.org -=- Follow on Twitter at @AsteriskDev |
| 06:34.04 | *** join/#asterisk-dev jkroon (~jkroon@165.16.203.122) |
| 07:07.54 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:b749:768d:ed77:7c9a) |
| 07:13.07 | *** join/#asterisk-dev pchero_work (~pchero@87.213.247.82) |
| 13:32.49 | *** join/#asterisk-dev sahmed (~sahmed@66.64.44.146) |
| 13:38.30 | coreyfarrell | I just realized AMI Originate with Async: true provides two response packets. does anyone know of other manager actions which act the same (aside from eventlist responses)? |
| 13:38.59 | gtjoseph | not offhand |
| 14:13.48 | coreyfarrell | so I'm working on some integration testing for an asterisk project. I need to initiate an outbound SIP call, have that call send dtmf, play audio and eventually hangup when I tell it to from node.js. is this something ARI could accomplish? |
| 14:14.15 | gtjoseph | seanbright: do you p[lan on creating/updating a testsuite test for the port matching in res_pjsip_endpoint_identifier_ip? |
| 14:14.36 | file | coreyfarrell: should be able to |
| 14:24.31 | coreyfarrell | I'm new to ARI, looks like I will need to create an empty POST to /channels?endpoint=SIP/trunk/number and that'll respond with a snapshot including the generated channel id and the channel will be in some kind of holding? if I don't need events back from ARI then I shouldn't need to create a stasis app? |
| 14:25.19 | seanbright | gtjoseph: i can |
| 14:25.37 | seanbright | i don't want to, but i can |
| 14:25.52 | gtjoseph | i think it'd be appropriate. one of the existing tests could probably be modified easily. |
| 14:26.10 | seanbright | i'll pay you $1000 to create a community AWS AMI with the testsuite configured on it |
| 14:26.17 | seanbright | PM me for details |
| 14:26.18 | seanbright | heh |
| 14:26.38 | gtjoseph | i'm not that cheap |
| 14:26.54 | seanbright | that's not what i've heard |
| 14:27.06 | gtjoseph | ouch |
| 14:27.20 | seanbright | you fix ASTERISK-28507 |
| 14:27.25 | seanbright | i'll create a test |
| 14:27.51 | seanbright | i'm in the fortunate position of not really caring if the port matching stuff goes in :) |
| 14:28.50 | gtjoseph | i'll be working on the wiki stuff this afternoon |
| 14:28.56 | seanbright | sweeeeeeeeeet |
| 14:30.01 | seanbright | ok, let me see if i can get this testsuite stuff to work |
| 14:31.32 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-uubbqwfqqrlgrefw) |
| 14:31.32 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 14:34.50 | file | coreyfarrell: that is equivalent to an AMI originate |
| 14:35.06 | file | coreyfarrell: if you send it to the dialplan, then there is no need for an ARI app to have a websocket open |
| 14:36.57 | coreyfarrell | ok but if I wanted to avoid dialplan then I need an ARI app? the docs aren't clear on this, they say only endpoint is required but it sounds like you're saying one of extension or app is required? |
| 14:37.08 | file | yes. |
| 14:37.30 | coreyfarrell | ok thanks for the info |
| 14:37.46 | file | and to execute most of the ARI operations, the channel has to be in an ARI application |
| 14:42.18 | seanbright | re: ast_sorcery_changeset_create() not being great |
| 14:42.43 | seanbright | i did a quick search the other day and i believe everywhere that uses it never actually uses the list of changes, but only uses it to determine if there are changes |
| 14:43.00 | gtjoseph | could be |
| 14:43.19 | seanbright | so we should either fix it or deprecate it |
| 14:43.23 | seanbright | but that is above my pay grade |
| 14:44.09 | gtjoseph | fair enough. might wan to create an ASTERISK issue to keep track of it. |
| 14:44.17 | seanbright | okie |
| 14:53.29 | seanbright | are the instructions here up-to-date? https://github.com/asterisk/testsuite/blob/master/README.txt |
| 14:57.26 | file | not... the most |
| 14:57.41 | seanbright | i'll re-write it |
| 14:57.41 | file | the wiki is closer |
| 14:57.51 | gtjoseph | relatively |
| 14:58.22 | seanbright | i won't rewrite it, i don't know why i said that |
| 14:58.40 | file | have a cookie instead! cookies are good |
| 14:59.24 | seanbright | i did resolve to eat more cookies in 2020 |
| 14:59.34 | file | acquire some from a local cookie emporium |
| 14:59.50 | file | artisinal organic fine crafted cookies |
| 15:03.41 | seanbright | so should i always be using the master branch of the testsuite, regardless of the asterisk version? |
| 15:03.50 | gtjoseph | NO |
| 15:04.04 | gtjoseph | use the branch that matches the asterisk branch |
| 15:04.04 | seanbright | don't yell at me |
| 15:04.20 | seanbright | roger roger |
| 15:04.23 | gtjoseph | I'm sowwy |
| 15:04.28 | file | his name is George, not roger |
| 15:04.38 | seanbright | what's your vector, victor? |
| 15:04.41 | gtjoseph | <PROTECTED> |
| 15:04.55 | gtjoseph | what's are clearance Clarence? |
| 15:09.19 | Corydon76 | I'm gonna hug him, and squeeze him, and call him George. |
| 15:11.45 | seanbright | don't squeeze george, he's had 3 cups of coffee this morning |
| 15:12.36 | *** join/#asterisk-dev kharwell (uid358942@gateway/web/irccloud.com/x-ygaxazzhhykfibiq) |
| 15:12.37 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 15:13.23 | seanbright | so to be clear, i should use the 13 branch to test asterisk 13 git (not the 13.30 branch?) |
| 15:14.01 | file | oui |
| 15:14.04 | seanbright | k |
| 15:14.10 | file | realistically in practice it's generally fine |
| 15:19.23 | seanbright | and it will not work with python3? |
| 15:19.58 | file | all sorts of nope |
| 15:20.12 | coreyfarrell | some work was done towards python3 but I never had a chance to finish it. |
| 15:20.27 | seanbright | alrighty |
| 15:21.20 | coreyfarrell | some tests might work, many will not. |
| 15:21.46 | seanbright | runtests doesn't work because it's looking for 'python' |
| 15:28.43 | coreyfarrell | you can test with `python3 ./runtests.py` |
| 15:29.49 | coreyfarrell | but realistically to know that you're using python3 across the board you'd probably need to create a python -> python3 symlink in the path somewhere |
| 15:30.40 | seanbright | i installed python2 on bionic (where python3 is the default) |
| 15:30.52 | seanbright | that gave me the symlink |
| 15:33.25 | coreyfarrell | you might be able to create a symlink in /usr/local/bin (that's usually before /usr/bin in $PATH) |
| 16:26.31 | *** join/#asterisk-dev jkroon (~jkroon@165.16.203.101) |
| 16:35.12 | seanbright | gtjoseph: hopefully my -2 on the port matching stuff will prevent a merge |
| 16:35.38 | seanbright | i discovered a minor issue with it |
| 16:35.41 | seanbright | it doesn't work |
| 16:37.53 | gtjoseph | okaaaay |
| 16:38.07 | file | seanbright: and to think, I TRUSTED YOU! |
| 16:40.30 | seanbright | i know |
| 16:40.45 | seanbright | i've let everyone down... again |
| 16:41.36 | seanbright | good thing that i decided to write a test for it without anyone prompting me to... |
| 17:33.31 | *** join/#asterisk-dev sahmed (~sahmed@12.235.204.202) |
| 17:55.56 | *** join/#asterisk-dev sahmed (~sahmed@12.235.204.202) |
| 18:15.38 | kharwell | seanbright: https://issues.asterisk.org/jira/browse/ASTERISK-28680 something you might look into, or just wanted to report the bug? |
| 18:19.41 | *** join/#asterisk-dev sahmed (~sahmed@cpe-70-114-236-63.austin.res.rr.com) |
| 18:28.05 | seanbright | kharwell: just something for someone to eventually look at |
| 18:28.24 | seanbright | gtjoseph suggested i create it |
| 18:28.36 | gtjoseph | yep |
| 18:29.10 | kharwell | seanbright: ack. thanks |
| 22:32.27 | seanbright | so i'm working on my super duper binary protocol for all things asterisk |
| 22:33.18 | seanbright | working on CDRs first because that seems easiest |
| 22:34.57 | seanbright | some of the fields in the ast_cdr struct (duration, billsec) are 64 bit values (which seems excessive, but whatever... oh and they're signed cuz...) |
| 22:35.35 | seanbright | trying to decide if i go the easy route and just serialize them in 8 bytes, or do some kinda of varint encoding (a la protobuf) |
| 22:36.17 | gtjoseph | i'd stay away from varint-like stuff |
| 22:36.26 | seanbright | i think pretty much 99.9% of calls will have a duration that can fit in 16 bits, so 64 bits seems a big waste |
| 22:37.00 | gtjoseph | it might have just been "make em all that way" thing |
| 22:37.07 | seanbright | yeah |
| 22:37.26 | seanbright | i guess i could TLV them |
| 22:37.42 | seanbright | actually, that might not be the worst idea |
| 22:38.22 | seanbright | so 1 byte to indicate it's the duration, 1 byte to indicate the number of bytes (8, 16, 32, or 64), and then the value? |
| 22:38.50 | seanbright | or maybe i am over-complicating |
| 22:39.05 | gtjoseph | hmmm. i'm not sure of the context you're working in. |
| 22:39.14 | gtjoseph | whatever it is...KISS |
| 22:39.26 | seanbright | that's inappropriat... ohhhh |
| 22:41.09 | seanbright | i might just saturate to 32 bits (unsigned) |
| 22:41.59 | seanbright | if there are calls longer than 136 years, we can revisit |
| 22:42.22 | gtjoseph | :) |
| 22:43.15 | kharwell | what am I missing here? If you are encoding into binary can't you just write out the full N-bits no matter what size? |
| 22:43.54 | rmudgett | In network byte order of course. :) |
| 22:44.48 | snuff-work | mm.. way i see it storage is cheap.. memory is cheap.. unless its a huge speed-up in processing by 'compressing' something like that i struggle to see the point.. not like asterisk runs on a arduino or some super small micro.. |
| 22:47.12 | rmudgett | seanbright: Aren't those 64 it values in milliseconds for CDR's? |
| 22:53.54 | seanbright | kharwell: you can |
| 22:54.40 | seanbright | kharwell: but i don't think you need an integer big enough to store the number of seconds until the heat death of the universe for a phone call |
| 22:54.57 | seanbright | rmudgett: they may be, i think it's configurable |
| 22:56.58 | seanbright | rmudgett: looks like seconds |
| 22:57.19 | kharwell | seanbright: I agree, and I think it's fine to write them out into a smaller size. They'll certainly be able to "fit" when read back in. Given the data context it's probably unnecessary, but you can always add check when encoding. For instance if writing out is greater than bit size then log a truncation error |
| 22:59.07 | seanbright | start, answer, end will be harder |
| 23:00.12 | seanbright | i was thinking NTP format |
| 23:00.41 | seanbright | but the 64 bit NTP format is only good until 2036 |
| 23:10.54 | seanbright | snuff-work: sorry, missed your comment. this is intended as a network protocol, so the fewer bytes the better. |
| 23:11.10 | seanbright | but i agree, i am probably overcomplicating it |
| 23:13.37 | snuff-work | i guess if you wanted to communicate over LoRa or some other very low b/w conns.. might also makes sense.. but these days bandwidth is almost free right (at the datacenter-ish level) never mind the consumer than still gets regularly shafted over speed/reliability |
| 23:13.38 | snuff-work | =D |
| 23:15.26 | seanbright | i intend to send the data via carrier pigeon, so... |
| 23:16.54 | snuff-work | pigeon will be wearing a little rfid tag or similar right.. so it is bribed by food to the 'transfer station'.. data uploaded to its tag and then goes on its way to the other end right? |
| 23:20.21 | seanbright | that isn't RFC compliant, i prefer https://tools.ietf.org/html/rfc2549 |
| 23:20.47 | seanbright | ugh, wrong one: https://tools.ietf.org/html/rfc1149 |
| 23:22.05 | snuff-work | i see it has also an IPv6 addition as well |