14:02:59 #startmeeting Development IRC meeting 27 October 2021 14:02:59 Meeting started Wed Oct 27 14:02:59 2021 UTC. The chair is Joubu. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:02:59 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:02:59 The meeting name has been set to 'development_irc_meeting_27_october_2021' 14:03:03 #chair jzairo 14:03:03 Current chairs: Joubu jzairo 14:03:06 #chair cait 14:03:06 Warning: Nick not in channel: cait 14:03:06 Current chairs: Joubu cait jzairo 14:03:07 #chair cait1 14:03:07 Current chairs: Joubu cait cait1 jzairo 14:03:20 #topic Introductions 14:03:29 #info Owen Leonard, Athens County Public Libraries, Ohio, USA 14:03:34 #info Marcel de Rooy, Rijksmuseum, The Netherlands 14:03:35 #info Martin Renvoize, PTFS Europe, UK 14:03:39 #info Katrin Fischer, BSZ, Germany 14:03:42 #info Thomas Dukleth, Agogme, New York City 14:03:43 qa_team? 14:03:43 i guess qa_team is cait, dcook, amoyano, ashimema, marcelr, kohaputti, jajm, tcohen, kidclamp, khall, tuxayo, petrova, nugged 14:03:49 #info Jonathan Druart 14:03:56 #info Andrew Nugged, National Library of Finland, HELSINKI 14:04:02 The agenda is full! 14:04:05 #info Peter (Petro) Vashchuk, National Library of Finland, HELSINKI 14:04:05 #info Victor Grousset, Tuxayo International Unlimited, France 14:04:07 #link https://wiki.koha-community.org/wiki/Development_IRC_meeting_27_October_2021 14:04:34 #info David Nind, New Zealand 14:04:47 davidnind++ 14:04:56 #info Jessica Zairo, ByWater Solutions 14:05:01 3am! 14:05:29 I happened to be awake! 14:05:41 🤗 14:06:02 davidnind++ 14:06:13 keep awake david 14:06:44 jzairo: do want to chair? 14:07:01 #topic Announcements 14:07:03 * cait1 sends davidnind some hot coffee and dark chocolate 14:07:06 Anyone have something? 14:07:16 deadlines maybe? 14:07:21 thanks cait1! 14:07:22 #info Tomas Cohen Arazi, Theke Solutions 14:07:38 KohaCon: I got the confirmation that it will be in-person only :( 14:07:39 cait1: yes, after 14:07:40 sure, I will give it a shot 14:07:55 tuxayo: from who? 14:08:11 kohapakistan [AT] gmail.com 14:08:36 I think it's Sher 14:08:49 Duplicate announcements to the Koha list and neither mentions in-person only :( 14:09:10 please fwd the email to "some of us" 14:09:34 ok! 14:09:48 indeed 14:09:55 hi 14:09:56 continuing with my section then letting the meeting in other hands 14:10:04 #topic Update from the Release Manager (21.11) 14:10:09 I don't have much to say. Nothing different from the usual. 14:10:09 Perhaps someone could reach them with an offer to support a remote conference effort. 14:10:11 fridolin: hi :D 14:10:16 * Feature freeze is end of the week! 14:10:27 Release dates sent to the list: 14:10:28 Oct 29th - "Soft" feature freeze, nothing big or with high risk of side-effects will be included into the final release if not marked as Passed QA 14:10:31 Nov 3-5 - "Hard" feature freeze, nothing considered as an improvement will be pushed if not marked as Passed QA 14:10:34 Nov 10th - String freeze, draft of release notes published 14:10:37 Nov 17th - Only bug fixes considered major, critical or blocker will be pushed 14:10:39 Nov 26th - Final release 14:10:41 * Please focus on this list - https://link.infini.fr/koha_bz_rel_21_11_candidate 14:10:50 I am keeping it up-to-date, so everything there is high priority (especially the bugfixes!) 14:11:00 * Missing release notes for new enh - Check that list, if you see your name please fill them in: https://link.infini.fr/koha_bz_release-notes-needed 14:11:07 tuxayo: was any reason given? 14:11:31 * 3 big things pushed in the last week, that were part of the roadmap: bug 11175, bug 14957 and bug 19185. Thanks to everybody involved! 14:11:31 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11175 enhancement, P5 - low, ---, martin.renvoize, Pushed to master , Show the parent record's component parts in the detailed views 14:11:32 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=14957 new feature, P5 - low, ---, glasklas, Pushed to master , Write protecting MARC fields based on source of import 14:11:33 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19185 enhancement, P5 - low, ---, jonathan.druart+koha, Pushed to master , Web installer and onboarding tool selenium test 14:11:49 14957 is missing some stuffs to be fully ready, have a look at the last comments and follow-up bug reports. 14:11:55 #info nick Clemens, ByWater solutions 14:11:56 cait1: not really, I will see if I can get more info and forward it to you 14:12:05 19185 is driving me crazy, I am having a hard time to make the tests pass (failing tests from Koha_Master, the full run). It seems that the "zebra srv mock" trick we have been doing for years in www/search_utf8.t isn't actually working as expected. 14:12:13 I based my last work on it assuming that... help welcome! 14:12:26 Ho, and we have a new blocker: bug 29330 14:12:26 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=29330 blocker, P5 - low, ---, koha-bugs, NEW , Koha cannot send emails with attachments 14:12:26 nice work 14:12:31 and a critical one: bug 28236 14:12:31 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=28236 critical, P5 - low, ---, johanna.raisa, Signed Off , Selecting database columns for system preferences in standard and dev installs is broken 14:12:32 #info Fridolin Somers, Biblibre 14:12:35 that's all for me 14:13:00 has anyone actually tested the new blocker? 14:13:22 I saw it float by.. but my understanding was that Email::Stuffer handles it transparently now 14:13:44 I haven't yet 14:13:49 The bug doesn't say how to reproduce the problem 14:13:50 Joubu: I rebased and checked 20447 (right now installing on production) but seems because just 2 days to FF... that won't pass? (ashimema?) (yeah that's big feature) 14:14:14 bug 20447 14:14:14 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20447 new feature, P5 - low, ---, ere.maijala, Needs Signoff , Add support for MARC holdings records 14:14:17 I'd say lets push it over the edge first thing next cycle 14:14:26 it's very close.. but it's also very big 😉 14:14:31 there is still also the bug for return-path, i m sorry we didn't get to test it yet 14:14:46 if it is a bug there is more time 14:15:07 nugged: my last comment is "Provide a full test plan and release notes please." and has been ignored, so... well.. 14:15:12 I think it's too big and too late for that one 14:15:46 ok 14:15:49 maybe for fridolin - early next release 14:15:56 to catch things 14:16:11 if anyone has time to test bug 28729 that would be awesome.. that's the return-path one cait1 was talking about. 14:16:11 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=28729 critical, P5 - low, ---, martin.renvoize, Needs Signoff , Return-path header not set in emails 14:16:14 nice 500K patch 14:16:29 cait? 14:16:30 i heard cait was going to add one for a follow-up on the receive/currnecy/preselect we talked about 14:16:31 * ashimema is trying hard to focus on Joubu's list and anything else that feels blocker like 14:16:35 yes? 14:16:58 I'm testing lots at the moment... hence spotted a few oddities this morning.. like the content blocks issue I promoited this morning 14:17:03 i am in for qa on that one ashimema 14:17:22 i just came across some authorised value problems 14:17:22 #action marcelr QA 28729 14:17:23 marcelr++ 14:17:36 marcel++ 14:17:38 marcelr "the brave" 14:17:45 not shying away from big patches ever 14:17:59 :) 14:18:04 hmm 14:18:17 beware.. the unit test is clear.. the code is clear.. but actually testing it is challenging to say the least.. 14:18:22 were you talking about another one? I can add another action :D 14:18:39 actions can be reverted too Joubu 14:18:39 moving on? 14:19:02 it relies on having a working email server and checking the results of sending an email... it's the MTA that actually sets the return path 14:19:03 fun aye. 14:19:15 sure 14:19:18 using gmail? 14:19:20 yah, really thankful you dug into the documeantions 14:19:20 av problems marcelr? 14:19:27 yes 14:19:35 we have matching problems on the fields 14:19:40 field length 14:19:53 av is 80 but authority just is on 10 still 14:20:02 maybe a few more 14:20:08 ah,. yes.. think I've spotted that a few times.. feel free to punt it my way if you come up with a decent idea on how to resolve 14:20:23 trivial dbrev 14:20:31 🙂 14:20:47 #topic Updates from the Release Maintainers 14:21:04 sorry, we need to move on, trying to make the meeting fit in 1h 14:21:11 good plan 14:21:21 nothing from rmaint I think 14:21:22 yey nothing about 20.11.x 14:21:32 Will release the minor in the next hours 14:22:11 I successfully worked out how to setup koha-testing-docker so you can send email using gmail: https://gitlab.com/-/snippets/1893788#enabling-email-for-testing 14:22:19 #info expect the next minor releases in the next hours 14:22:37 we had some issues with translations - something to check on next time maybe 14:22:38 (20.05.x at least ^^") 14:22:42 to see if things go smoothly again 14:23:09 I suggested adjusting the mail text slightly - somtimes the strings are not updated yet when string freeze is called 14:23:12 cait1: yes, I'll propose that we check that before announcing the string freeze 14:23:23 so maybe good to word it like "check on new strings in the coming days" or so 14:23:40 that would also be great :) 14:24:04 That could be another way, I'll see if we can announce when it's ready but otherwise, yes changing the email text 14:24:13 Or ask Bernardo when he is usually doing it? 14:24:58 #topic Updates from the QA team 14:25:10 I asked a long time ago, it's the 15th and the 17th but didn't ask if it's automated. 14:25:17 it's not 14:25:22 Bernardo is usually doing it around 15th yep 14:25:39 ok 14:25:53 (well, I don't think it is, I saw run in the history every month around ~15-17th) 14:26:18 cait1, QA? 14:27:21 thats an answer too 14:27:30 sorry 14:27:36 I've been pushing for the bad and old bugs 14:27:52 i'd really like to see a little mor QA action so close to freeze 14:28:21 but overall... I was really happy to see some of the old and big go in this week 14:28:25 * ashimema is trying.. and marcelr has been magnificent 14:28:27 a big thank you to those involved with that 14:28:42 yes, as usual the wrong ones listen up :) 14:28:49 58 is not that bad currently 14:28:56 i know it was a big piece of work, so thanks a lot 14:29:01 ashimema++ marcelr++ 14:29:24 i'll keep on pushing for some more old to new action next release cycle 14:29:49 (from our side can say that we not directly placing QAs but we have current master on one pilot production and (I hope) that bunch of feedbacks we did directly to Joubu helped to move as well) 14:30:13 every bit of feedback helps of course 14:30:16 yes, nugged, very useful! 14:30:23 i can only talk about QA 14:30:35 and we could use more hands/heads/exes there regularly to make things move better 14:30:40 (and that production is alive with current master (1 week old though, just installing latest today - so will be more feedback these days) 14:31:34 i didn't want that to sound too negative, but spreading the work more would really help I think 14:31:53 🙂 14:32:19 # Status of roadmap projects 14:32:20 I only see encoding error, but I hope it's a good one? 14:32:39 there are some roadmap projects PQA 14:33:39 ? 14:33:45 should you use # and topic 14:33:46 good news or complaining because it's not pushed? 14:33:57 there is something on the agenda 14:33:59 the latter 14:34:00 it has been said that the latter is the last of the list, no? 14:34:00 :-D 14:34:07 topic change? 14:34:11 so I was going to pick the stuff from there first 14:34:22 #topic Bug 19532 (Recalls) 14:34:22 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=19532 new feature, P1 - high, ---, aleisha, Needs Signoff , Recalls for Koha 14:34:31 marcelr wanted to chat a bit about 19532 14:34:41 wont repeat my email here but wanted more feedback on it 14:34:42 everybody has the page opened? 14:34:46 "(Recalls) was on the roadmap; it is a feature that many libraries probably want; Aleisha rebased it around thousand times. It is too large to push it to 21.11 short before release. Some questioned the fundamentals too. It would only be fair if QA formulates an opinion on it as a team, and when agreed upon, new RM and QA coordinate pushing it early. As a side note, I really dislike patch sets of this size; we 14:34:52 should avoid them wherever possible" 14:35:25 think personally that we should try to adjust its current form; an other way will be harder in several aspects 14:35:28 i discover this bug 14:35:47 its 4 years ago so we cant say now rewrite it 14:35:48 If some of us are interested in it, we should help Aleisha 14:35:55 not letting her rebasing endlessly 14:35:59 right 14:36:00 I think that's marcelr's point 14:36:19 and since it is so large, asking for more opionions 14:36:23 hm 14:36:27 good thinks these feature can be turned off/on 14:36:46 once it's pushed, it's code to maintain 14:36:55 sure 14:36:56 so no, it's not because you can turn it off that you must push it, fridolin :D 14:37:06 ^^ 14:37:19 I have had a galcne at the spec 14:37:26 it doesn't quite work as academics expect it 14:37:31 but i see potential to extend 14:37:32 I think it's something for next month anyway 14:37:39 i'd like to have a closer look and can offer GUI testing 14:37:42 it's too late for 21.11 and we have plenty of things to focus on 14:37:45 but i'll need help with the code side 14:38:00 yes, this is for after release 14:38:10 we need the new RM not to push other large stuff before it ? 14:38:19 i'll look at code 14:38:30 we need the team to focus on it and know if it's how we want this feature 14:38:39 if it's not, it's our job to rework it 14:38:55 I worry that it is independent of holds as the workflows are similar, but if it is made clear how they interact it can be its own feature 14:39:07 i believe maybe it can be separate 14:39:12 if we can add an automatism too 14:39:16 like... unmoderated recalls 14:39:29 kidclamp: building on holds is a tricky thing too 14:39:36 indeed 14:39:41 holds is a scary place 14:39:43 but it is a clear point 14:39:47 thta would be nice - I feel it is a different workflow than others expect for recalls 14:39:48 somaybe it is good to have it separate 14:39:52 unmoderated is a nice compromise 14:40:11 i think it might not be int here yet, but that's a direction we could look for 14:40:18 The workflow being different than some expect isn't a reason to reject it 14:40:20 i have to study specs and docs more 14:40:32 It would be nice if it was written in a modern style.. that's my main issue.. yes it was written many moons ago.. but unless it's changed allot since.. it was a massive copy paste of holds at the time as far as I could tell 14:40:38 Project Koha_Master build #1797: STILL UNSTABLE in 1 hr 5 min: https://jenkins.koha-community.org/job/Koha_Master/1797/ 14:41:01 so maybe we could agree on making this a priority after release? 14:41:09 I think that would be great 14:41:13 fridolin volunteered for code - a second one maybe? 14:41:16 we should, or we should have disapproved much earlier 14:41:26 i am looking too 14:41:31 ok 14:41:36 so let's make that an agreed 14:41:41 I asked for a rename at some point.. I spoke to a number of interested parties and none of them recognised it as 'recalls' once I explained how it worked/what it did 14:41:44 which is a shame 14:42:00 #agreed marcelr, fridolin and cait to look at recalls for QA after 21.11 release 14:42:05 not a blocker.. but I feel naming it recalls misleads over half the stakeholders interested in it. 14:42:11 did you put a better name on the report ? 14:42:11 🙂 14:42:34 seems I didn't.. I am dredging the brain from reviewing it years ago 😉 14:42:44 let's look at this with fresh eyes 14:42:47 what is the better name actually ? 14:43:00 bug 17015 - if we want to save old much worked on code 14:43:00 04Bug https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17015 enhancement, P5 - low, ---, maryse.simard, Failed QA , New Koha Calendar 14:43:02 i think it#s somethig like... a patrn iitiated request to get something earlier? 14:43:05 at th emoment 14:43:18 patron driven recalls? 14:43:32 i am nto done yet with docs, ran out of time an donly had a quick look 14:43:34 "stakeholders" aren't stakeholders if they're not chipping in 14:43:52 look at the bug 😉 14:43:53 oleonard: hm? 14:43:54 patron driven or unmoderated 14:44:01 i think they are moderated 14:44:08 indeed 14:44:15 https://www.library.cornell.edu/services/request/recall # searched for "library recall" 14:44:18 what we are looking for later might be unmoderated 14:44:51 or more "uatomated" from staff side 14:45:11 shall we leave it with the action/agreed and move on for now? 14:45:17 this lets the patron place the recall on their own and automatically (cronjob) shortens the due date - no staff action (moderation) 14:45:23 either way.. lets attack it as a group at the beginning of the next cycle. 14:45:32 thats the plan 14:45:37 kidclamp: ah ok - need to read up 14:45:44 yep 14:45:49 Joubu: moving on? 14:45:49 I wondered if we were using different meanings :-) 14:46:02 #topic Actions from last meeting 14:46:08 #topic ashimema initiate the advent calendar - create a pad to collect ideas + send email to the list 14:46:11 kidclamp: i missed out on my homework - that's the problem, I will catch up 14:46:28 ooh.. I forgot about the pad.. but did send an email 14:46:34 cait1 is the only one to have replied so far 14:46:38 oops 14:47:13 Shouldn't we have a place to put ideas? 14:47:22 i was not sure how technical so suggested something like "bugzilla search tips and tricks" 14:47:23 at least to know what others have in mind? 14:47:52 so pad? 14:48:06 No action, in 2 weeks it will be too late 14:48:31 we have about a month now... not a lot of time 14:48:43 but if multiple authors know what to write and how it's doable still 14:49:00 yup 14:49:03 I have a list 14:49:10 and a small group of people 14:49:25 will you get in touch? 14:49:27 but so far it's been a bit 'slack centric' 14:49:31 maybe email? 14:49:32 yup 14:49:35 slack? 14:49:35 i guess slack is expensive, but it falls under the 'just works' camp 14:49:56 let's do it! 14:50:08 I don't have access to slack, so just let me know what's up please 14:50:10 moving on? 14:50:10 slack is non-free 14:50:42 thd: yes, it's not a community resource 14:51:39 lets move on 14:51:49 I haven't been approached to I guess you don't need help 14:51:51 good :) 14:52:02 #topic LTS version 14:52:08 a fun one 14:52:14 won't fit in 1h.. 14:52:15 yes :D 14:52:27 lets make 20.05 the LTS 14:52:32 Should we skip it or discuss? 14:52:41 20.05 is gonna be dropped in 1 month 14:52:56 i think we can skip - but we shoudl agree on a next sstep 14:53:01 are we postponing by 2 weeks 14:53:08 or ... move back to mailing list? 14:53:21 ok 14:53:21 what is to be discussed? 14:53:23 i just put the questions there since I felt that the ml list thread got too quiet 14:53:26 it was stuck on the ML so decided here 14:53:29 maybe too lost in the security discussion 14:53:30 19.11 has volunteers to make it LTS 14:53:41 shoudl we start a new thread just LTS? 14:54:06 ok, quick 14:54:10 * Will a 2 year cycle work for Koha? 14:54:11 yes/no? 14:54:25 is that too short or 2 years is fine? 14:54:26 is that a lot of different to maintaining up to 3 now? 14:54:31 i mean does it win us much time 14:54:32 yes 14:54:55 as cait1 said, 2 is not much longer than what we do now 14:54:56 half a year longer guaranteed then, right 14:54:57 if we're doing an LTS I would like to see us drop oldoldstable 14:55:11 ashimema: that's question 2 or 3 14:55:19 had you in mind :) 14:55:35 ohw long is debian/ubuntu lts? 2 years? 14:55:46 6 at least 14:55:54 yup 14:56:02 oh wow 14:56:05 3y sounds better to me, but we need something to maintain it, and keep on doing it 14:56:11 someone* oops 14:56:20 indeed 14:56:20 does 3y make sense? 14:56:24 3 feels better, agree 14:56:27 3y seems reasonable 14:56:36 * How will updates work? Can you jump from LTS to a version in between? 14:56:37 20.04 LTS released 2020-04-23, maintenance until 2030-04-23 14:56:37 10y! 14:56:49 I dont' think this question make sense, we have a strong upgrade progress 14:56:53 i think that's too hard for us 14:56:54 yes 14:56:54 what would be the problem? 14:56:56 indeed 14:56:56 no issues 14:57:05 it was raised last meetiing 14:57:15 i try to remember what the point made was 14:57:29 * How many versions should be maintained apart from LTS? 14:57:30 LTS -> LTS would be a massive jump.. 14:57:37 3 years worth of DB upgrade 14:57:39 yes, i think we need to give people an out 14:57:47 that's gonna be fun for anyone doing it. 14:58:02 it does work (i just did one from 16.11) :) but people might follow LTS and want to switch 14:58:05 like those doing 3.22 to 20.11? :) 14:58:10 we have some jumps actually from 16.11 to 20.11 14:58:10 if idempotent db updates is enough, we are good 14:58:26 > LTS -> LTS would be a massive jump.. 14:58:26 A certain number of instances are upgraded like that right? 14:58:29 ok, so shouldn't be a problem 14:58:43 * How many versions should be maintained apart from LTS? 14:58:46 just something to keep in mind and maybe soemthing to test 14:58:47 (bis) 14:58:58 I used to think following the ubuntu cycle made some sense.. but now I can't for the life of me see any patturn to it. 14:59:08 heh 14:59:14 the number of versions we can maintain, depending on the rmaints available 14:59:18 personally.. I feel like the numerious stables pull resource away from QA and SO 14:59:19 at least 2, maybe 3 14:59:38 or say 4 and LTS==2 y 14:59:49 'L' 14:59:52 I've felt for a while that it would be good to have 1 or 2 longer supported versions with more of a rolling option too 14:59:57 same feeling, we should focus on stable+oldstable in my opinion 15:00:01 Ideally managing to keep only 3 RMaints in total since between half or a third of them could be doing QA instead of RMaint. To account for the QA workforce shortage. 15:00:29 fridolin: how does that work for people doing yearly upgrades? Like BibLibre 15:00:38 LTS is basically security only isn't it.? 15:00:51 yes 15:00:59 shouldn't be too problematic for the maintainer 15:01:00 what about supported platforms? 15:01:01 more for the devs 15:01:07 even that may be hard to rrebase on older versions ? 15:01:26 but it will prevent us from delivering .patch files to apply, like we did last sec release 15:01:31 * ashimema would love to see 'LTS or Rolling' 15:01:59 monthly release for selected LTS versions for whatever period we decide.. monthly rolling release between times 15:01:59 i don't know what Rolling means 15:02:31 rolling releases are the modern not quit continuous integration approach 15:02:36 heading for the canyon haha 15:02:54 what about a compromise for next ccycle? 15:02:57 i.e. we just keep releasing as soon as feature are ready.. rather than have a dev branch that can be anything up to 6 months ahead 15:02:58 we already have some gap 15:03:25 we have time to think it anyway, will postpone for next meeting 15:03:28 or even the one after 15:03:32 moving on 15:03:45 ok 15:03:45 #topic KohaCon22 15:03:50