| 02:50.51 | *** join/#asterisk-dev spatel (~spatel@pool-96-237-230-175.bstnma.fios.verizon.net) |
| 03:01.26 | *** join/#asterisk-dev spatel (~spatel@pool-96-237-230-175.bstnma.fios.verizon.net) |
| 06:08.50 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01:e0a:d:1111:b749:768d:ed77:7c9a) |
| 07:29.08 | *** join/#asterisk-dev spatel (~spatel@pool-96-237-230-175.bstnma.fios.verizon.net) |
| 10:21.08 | *** join/#asterisk-dev spatel (~spatel@pool-96-237-230-175.bstnma.fios.verizon.net) |
| 13:15.27 | seanbright | can we talk briefly about the MoH announcement patch? |
| 13:15.38 | gtjoseph | ok |
| 13:16.15 | seanbright | so if you have an announcement that is 10 seconds long |
| 13:16.20 | seanbright | and you have this in your dialplan: |
| 13:16.30 | seanbright | same => n,MusicOnHold(default,2) |
| 13:16.35 | seanbright | same => n,MusicOnHold(default,2) |
| 13:16.55 | seanbright | it will play the first 2 seconds of the announcement each time MusicOnHold is entered |
| 13:17.05 | seanbright | that is true both before and after my patch |
| 13:17.12 | seanbright | i am just wondering if that is desirable |
| 13:17.44 | gtjoseph | that's the way i'd expect it to work. what would be the alternative? pick up where it left off? |
| 13:17.58 | seanbright | yes, treat the announcement as just another file |
| 13:18.12 | seanbright | but if you would expect the behavior i described, then we're good to go |
| 13:18.44 | gtjoseph | do we pick up where we left off for a regular file? |
| 13:18.48 | seanbright | yes |
| 13:19.00 | gtjoseph | hmmm. so we maintain state? |
| 13:19.07 | file | we do |
| 13:19.11 | seanbright | that is the reason my first patch from a few years ago got rolled back |
| 13:19.31 | seanbright | because it didn't maintain state properly |
| 13:19.54 | seanbright | funny aside - i did a "clean room" implementation of this fix and initially ended up with the exact same patch as last time |
| 13:20.16 | gtjoseph | :) I guess I can see that for the actual file but I'd expect the announcement to restart. |
| 13:20.26 | seanbright | then we're good to go |
| 13:20.46 | file | to some place warm? |
| 13:21.01 | seanbright | until we have to roll it back for some unforseen reason |
| 13:21.13 | seanbright | it's actually super comfortable outside here right now |
| 13:21.20 | seanbright | i kinda want to just go lay in the grass somewhere |
| 13:21.41 | file | it's 53F here at the moment |
| 13:22.07 | seanbright | same here apparently... |
| 13:22.16 | seanbright | which is odd |
| 13:22.19 | file | Hurricane Teddy is rolling in this afternoon |
| 13:22.22 | seanbright | cuz you know... canada |
| 13:22.29 | gtjoseph | tsearle: can you check the diversion_basic, diversion_caller_id and diversion_response tests in your review to make sure they actually work? They fail for me but it could just be me. The new tests pass. |
| 13:23.53 | tsearle | hrm...were those tests in a different directory? |
| 13:26.11 | tsearle | no, those tests are in our test suite & are passing |
| 13:26.26 | tsearle | which version, the one in the merge request or the original from upstream? |
| 13:28.04 | tsearle | sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/PZuxTUFEVbGSJvBdiuvjzXaq/message.txt > |
| 13:28.46 | tsearle | that was on Sep 14, 2020 |
| 13:28.47 | gtjoseph | The original versions pass fine. The versions in your review fail. |
| 13:29.02 | gtjoseph | if they work for you, i'll keep looking on my side. |
| 13:31.21 | tsearle | if I recall correctly the main change I made to the existing tests were to make 181 mandatory |
| 13:35.15 | tsearle | https://gerrit.asterisk.org/c/testsuite/+/14958/1/tests/channels/pjsip/diversion/diversion_basic/sipp/user1.xml |
| 13:35.31 | tsearle | yeah, only change is to make 181 mandatory |
| 13:35.44 | tsearle | and that should have already been the case with the first merged patch "forward 181" |
| 13:36.18 | gtjoseph | yeah. i see the 181 come from asterisk (2 in fact) then sipp sends a cancel. |
| 13:36.49 | gtjoseph | i'll mess with it a bit. |
| 13:36.50 | tsearle | strange, like it isn't expecting the 181 |
| 13:37.06 | tsearle | oh 2 181? |
| 13:37.48 | gtjoseph | yeah. |
| 13:38.11 | gtjoseph | and i think the second one causes sipp to fail. |
| 13:39.05 | tsearle | well I do has some small differences between my version and upstream, in my version I only generate 181 on I think (and not 302) but that isn't relevant to this flow |
| 13:40.02 | tsearle | but I still don't undrestand |
| 13:40.29 | tsearle | even if 181 is optional, it would still fail if it got more than one, wouldn't it? |
| 13:41.57 | gtjoseph | actually, i'm trying that now |
| 13:45.09 | gtjoseph | it's got to be a timing thing. sometimes I get 2 181, others I get 1. and it passes either way with 181 optional |
| 13:47.07 | tsearle | ok |
| 13:47.37 | tsearle | in chan sip there was an xmit and reliable_xmit |
| 13:47.50 | tsearle | could some sort of retransmit flag be sent to the pjsip stack? |
| 13:49.02 | gtjoseph | these are send almost immediately with no delay between them. also i noticed that the packets are actually different. |
| 13:50.25 | gtjoseph | the first one is 608 bytes with Server Contact and Allow headers in that order. The second is 618 bytes with the headers rearranges and the Supoported header has 2 histinfo entries. |
| 13:51.43 | tsearle | so no history-info in the first 181. |
| 13:51.49 | tsearle | odd |
| 13:52.09 | gtjoseph | yeah it's there in the first. it's doubled in the second. pastebin forthcoming |
| 13:52.36 | gtjoseph | https://pastebin.com/GUzvdbJV |
| 13:53.14 | tsearle | that is weird |
| 13:53.30 | tsearle | I feel like the diversion module must have been loaded twice? |
| 13:53.46 | seanbright | that's not possible |
| 13:53.56 | gtjoseph | correct |
| 13:54.24 | gtjoseph | but the callbacks could have been called twice. |
| 13:55.26 | tsearle | diversion_outgoing_response |
| 13:55.26 | tsearle | has to be called twice |
| 13:56.10 | tsearle | <PROTECTED> |
| 13:56.13 | tsearle | that might be the cause? |
| 13:56.53 | gtjoseph | <PROTECTED> |
| 13:57.32 | *** join/#asterisk-dev kharwell (uid358942@gateway/web/irccloud.com/x-qenmdwnxoityqqpu) |
| 13:57.32 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 13:57.58 | tsearle | that line was changed in the forward 181, withouth the AST_SIP_SESSION_BEFORE_MEDIA the diversion module would not receive 181 |
| 13:59.05 | gtjoseph | you might have to add a check to see if Supported already has histinfo in it. |
| 14:00.39 | tsearle | but that only fixed part of the issue right? |
| 14:00.48 | tsearle | you still have 181 |
| 14:01.04 | tsearle | 2 181 |
| 14:02.14 | gtjoseph | yeah but if you detect that histinfo is already there, suppress sending it again. i haven't looked at diversion in a while but you should be able to do it. |
| 14:24.04 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-zmfcjsnaxrxpttar) |
| 14:24.04 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 14:36.09 | DanJenkins[m] | gtjoseph: let me know ehen youve sent that PR for the backend and I'll get it merged |
| 14:36.36 | gtjoseph | wilco. about an hour. |
| 14:37.05 | DanJenkins[m] | ðï¸ |
| 14:37.46 | DanJenkins[m] | im currently seeing how difficult it would be to support say Azure STT too.... and then I'll go and add some other basic additions... like being able to download a transcription and an audio recording of the conference |
| 14:38.00 | gtjoseph | Nice! |
| 14:38.01 | DanJenkins[m] | not quite sure how to actually make the transcription useful yet..... |
| 14:38.25 | DanJenkins[m] | (i have a semi-different goal than you right now... I'm thinking "what would look cool for my talk on friday") |
| 14:38.36 | gtjoseph | Well, we want to use it for AstriDevCon... save to file and would be nice to show it in the chat box. |
| 14:38.54 | DanJenkins[m] | so STT already shows in a box on the side |
| 14:39.20 | gtjoseph | Ah cool. |
| 14:44.22 | DanJenkins[m] | and its progressive.... so as you speak it keeps up with you, and each individual stream user has their own stream going to google so you truly get 100% Dan said X and Y and George said A and B |
| 14:44.40 | gtjoseph | wicked! |
| 14:45.15 | DanJenkins[m] | i push it into MQTT from the STT process and it makes its way down to mqtt websocket in the browser |
| 15:40.01 | *** join/#asterisk-dev CELYA_ (~Thunderbi@2a01cb0583d31d00f4d7ddd1bc0987bc.ipv6.abo.wanadoo.fr) |
| 16:03.56 | gtjoseph | DanJenkins[m]: pull request for chat created |
| 16:28.55 | *** join/#asterisk-dev bford (uid283514@gateway/web/irccloud.com/x-vepdczaarcyrcaka) |
| 16:28.55 | *** mode/#asterisk-dev [+o bford] by ChanServ |
| 16:30.53 | DanJenkins[m] | woot gtjoseph |
| 16:31.47 | gtjoseph | now about the front end... my react-fu is dismal at best. Do you have time to add a chat box that gets those events and calls sendMessage()? |
| 16:34.44 | gtjoseph | @DanJenkins[m]: ^^ |
| 16:36.40 | DanJenkins[m] | I can do that no problem - I just commented on the PR with some changes :) sorry! |
| 16:36.53 | gtjoseph | No worries and thanks. I'll take a look. |
| 16:39.26 | DanJenkins[m] | what do you think to the UI? different box to the transcription? |
| 16:39.57 | gtjoseph | I'd say yes or the conversions will get intermixes and folks will lose track |
| 16:40.20 | DanJenkins[m] | https://dan-jenkins.co.uk/dana-audioserver-transcription/ |
| 16:40.28 | DanJenkins[m] | no worries, I'll have a think |
| 16:40.40 | DanJenkins[m] | the other issue is right now the transcription appears over the top |
| 16:40.45 | DanJenkins[m] | ill have a think |
| 16:41.45 | gtjoseph | So for the display name... The issue is that if you want to use a different display name for a message than for the actual video session, you have to change the _ua display_name. |
| 16:41.59 | gtjoseph | nothing forces you to use a different name though. |
| 16:42.03 | DanJenkins[m] | right... why would you want to change it? ;) |
| 16:42.29 | gtjoseph | I don't remember. :) |
| 16:42.34 | gtjoseph | We had a reason though. |
| 16:42.47 | DanJenkins[m] | lol |
| 16:42.49 | gtjoseph | I can leave it out though. |
| 16:42.58 | gtjoseph | we can always add it if needed |
| 16:52.33 | *** join/#asterisk-dev mokolea (5448495a@84-72-73-90.dclient.hispeed.ch) |
| 17:01.20 | gtjoseph | DanJenkins[m]: pull request updated |
| 17:06.35 | DanJenkins[m] | Sweet thanks |
| 18:56.28 | *** join/#asterisk-dev AsteriskRoss (~AsteriskR@37.157.48.2) |
| 19:58.13 | seanbright | file: before your res_pjsip_session review gets merged |
| 19:58.52 | seanbright | if you move the initialization of 'session' down past the first 'return' statement, there are no other early exists from that function and you could just put the ao2_cleanup at the end of the function instead of using RAII_VAR |
| 19:59.00 | seanbright | i don't know that anyone but me cares |
| 19:59.11 | seanbright | early exits* |
| 19:59.26 | file | is fine either way |
| 20:00.05 | seanbright | ok, if that didn't blow your hair back don't worry about it |
| 20:00.46 | seanbright | we also only need 'session' for it's name, which could just be an ast_strdupa |
| 20:00.51 | seanbright | BUT |
| 20:00.59 | seanbright | wanders off mumbling to himself |
| 20:01.09 | file | wonders if his review will be changed entirely by tomorrow morning |
| 20:01.16 | seanbright | heh |
| 20:01.22 | seanbright | it's fine |
| 20:01.38 | seanbright | NI_MAXHOST! that is what i have been looking for |
| 20:01.45 | file | how random. |
| 20:01.59 | seanbright | its around the area of your patch |
| 20:33.11 | *** join/#asterisk-dev AsteriskRoss (~AsteriskR@37.157.48.2) |
| 21:59.56 | *** join/#asterisk-dev spatel (~spatel@50-252-194-73-static.hfc.comcastbusiness.net) |
| 22:05.18 | *** join/#asterisk-dev tuxd00d (sid401305@gateway/web/irccloud.com/x-dkqchqwljxuyqqki) |
| 22:05.28 | *** join/#asterisk-dev file (sid178970@asterisk/developer-and-muffin-lover/file) |
| 22:05.28 | *** mode/#asterisk-dev [+o file] by ChanServ |
| 22:06.28 | *** join/#asterisk-dev kharwell (uid358942@gateway/web/irccloud.com/x-fxzohjbcfhrnnsfb) |
| 22:06.31 | *** mode/#asterisk-dev [+o kharwell] by ChanServ |
| 23:09.52 | *** join/#asterisk-dev jkroon (~jkroon@165.16.204.107) |
| 23:44.35 | *** join/#asterisk-dev spatel (~spatel@96.237.230.175) |