15:01:04 #startmeeting Development IRC meeting 10 February 2021 15:01:04 Meeting started Wed Feb 10 15:01:04 2021 UTC. The chair is Joubu. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:04 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:04 The meeting name has been set to 'development_irc_meeting_10_february_2021' 15:01:09 #topic Introductions 15:01:17 #info Jonathan Druart 15:01:20 #info Martin Renvoize, PTFS Europe, UK 15:01:20 qa_team? 15:01:20 rumour has it qa_team is cait, dcook, amoyano, ashimema, marcelr, kohaputti, jajm, tcohen, josef_moravec, kidclamp, khall and tuxayo 15:01:22 rmaints? 15:01:22 rmaints is fridolin, AndrewFH and tuxayo 15:01:43 #info Tomas Cohen Arazi, Theke Solutions 15:01:54 info Nick Clemens, ByWaterSolutions 15:02:06 #info Thomas Dukleth, Agogme, New York City 15:02:47 #link https://wiki.koha-community.org/wiki/Development_IRC_meeting_10_February_2021 15:02:54 #info tuxayo/Victor Grousset, France 15:03:12 I have ton of things to tell you 15:03:17 #topic Announcements 15:03:24 Anyone have something? 15:03:29 #info Kyle Hall, ByWater Solutions 15:04:07 nope 15:04:11 #topic Update from the Release Manager (21.05) 15:04:21 we push some interesting things lately 15:04:33 1. Bug 20212 is adding all the DataTables features (we implement so far) on the parcel page 15:04:33 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20212 enhancement, P5 - low, ---, tomascohen, Pushed to master , Slowness in receiving in acquisitions 15:04:47 2. Bug 25026 is turning on RaiseError on the DB handler 15:04:47 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25026 enhancement, P5 - low, ---, jonathan.druart, Pushed to master , RaiseError must be set for the UI 15:04:56 This is a quite big change, the SQL requests that are exploding will not longer explode silently. With bug 26048, a nice 500 will be displayed to the end user (only if not running a dev install). 15:04:57 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=26048 normal, P5 - low, ---, dcook, Pushed to master , PSGI Koha does not use custom ErrorDocument pages 15:05:27 and I forgot the 3rd one 15:05:44 Then I'd like to highlight some patches I wrote in the last couple of weeks that need to be discussion/feedback/testing: 15:05:57 Bug 25078 - Rebuild the way our upgrade process is done. See comment 17 for screenshots. 15:05:57 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=25078 enhancement, P5 - low, ---, jonathan.druart, In Discussion , Gracefully stop updatedatabase/installer when a database change cannot be applied 15:06:13 There is still a lot of work to do but we need first to agree on *how* that must be implemented. 15:06:23 Bug 27526 - Start removing MARC from our items form. 15:06:23 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27526 enhancement, P5 - low, ---, jonathan.druart, In Discussion , Remove Mod/AddItemFromMarc from additem.pl 15:06:32 It's only a first step, it needs to be tested carefully to let us apply it to other places where items are created/modified. Please test! 15:06:47 Bug 27467 - To add the ability to share a "datatable view" with an URL. 15:06:47 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27467 enhancement, P5 - low, ---, jonathan.druart, In Discussion , Provide a static URL for tables 15:07:22 Bug 27344 - Rely on RabbitMQ to schedule the record indexation (ES). That will speed up several workflows (see 27341). If we don't go with RabbitMQ we have to write an async solution anyway. 15:07:22 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27344 enhancement, P5 - low, ---, jonathan.druart, In Discussion , Implement Elastic's update_index_background using Koha::BackgroundJob 15:07:41 #info be involved on 25078, 27526, 27467 and 27344 (please) 15:07:43 #info Katrin Fischer, BSZ, Germany 15:07:48 all interesting work.. all things I have on my list to work through :) 15:07:52 that's all for me, any questions? 15:08:00 reading 15:08:00 I've commented on many, but it would be great to get more eyes on them. 15:08:24 There are also things from the roadmap btw 15:08:47 https://annuel.framapad.org/p/koha_21.05_roadmap 15:08:54 that's moving behind the scene, check the pad regularly to know what's happening 15:09:17 I'd like to see some improvements to the task queue system prior to bug 27344.. 15:09:27 ha, and I am contacting "people" to get their go on removing the installer files for translated languages 15:09:42 it-IT answered yes, so far 15:09:46 I'm sold on 27344, but I feel strongly that we need to ensure that indexing doesn't get stuck behind long running batch jobs 15:10:00 some kind of prioritizing maybe? 15:10:06 indexing jobs need their own worker in my opinion 15:10:07 ashimema: then you need several workers 15:10:12 that's a sysop job 15:10:35 except.. currently there's no way to specify what jobs a worker can take on 15:10:36 ;) 15:10:52 ashimema: but you can have several workers 15:10:59 * ashimema also wants 'delays' in the job queue system.. but that's another track entirely. 15:11:04 if you have 3 workers and 3 big things are running, yes, it will be delayed 15:11:12 indeed 15:11:14 And I understand your point ;) 15:11:54 I had a hard time getting it pushed, it's always hard for me to get back to things I need to pull strongly 15:12:03 also.. can our ES implementation cope with running multiple indexing simultaneously ;) 15:12:09 ashimema: you volunteered and I will be happy to test :) 15:12:11 we pass a 'bulk' call to the indexer 15:12:26 if we pass two with overlapping record ids what happens? 15:12:44 it's processed twice 15:13:02 I feel like we need one worker dedicated to the task.. then it can get through the queue without interuption from other jobs.. and also make sure they index runs are done in order 15:13:05 we need some sort of precedence, right? 15:13:12 it's processed twice.. that's fine 15:13:13 but.. 15:13:41 I am not saying that's the perfect way of doing it, but it's the easiest for now 15:13:43 do we send the record data changes with our call to ES.. or do we lookup the latest version from the DB with each run? 15:13:53 and we absolutely need to async this task 15:13:54 if we send it along in the job.. then you could have them run out of order 15:14:06 and thus have the second run replace the first with older data 15:14:13 exactly.. 15:14:24 so those are the questions I'm asking and reasons I'm asking them ;) 15:14:35 anywho.. it's a great way to go.. just needs a little more ;) 15:15:04 I am joining you if you want to improve it :) 15:15:20 * ashimema has soooooo many projects on the go right now.. 15:15:30 that's a general problem :) 15:15:33 but will certainly try to make some time for this one too 15:15:51 I did add a bug to clarify the warning on the about page for the message broker 15:16:05 that's my very first step at looking at the background jobs code more clearly 15:16:21 Is there a means to sequence separate worker index processes? 15:16:22 I don't like right now that we have to change lots of files to list the jobs 15:16:41 adding a job type should be as simple as adding a class file to me.. 15:16:55 you shouldn't then have to list it in various places for it to actually ever get called 15:17:03 anywho.. I digress ;) 15:17:06 lets move on 15:17:06 ashimema: let's discuss this after the meeting 15:17:13 +1 15:17:29 yes, 1 file should ineherit, and overwrite things. But then it get more complex :) 15:17:36 other questions? 15:17:38 thd.. not yet ;) 15:18:01 #topic Updates from the Release Maintainers 15:18:51 rmaints? 15:18:51 it has been said that rmaints is fridolin, AndrewFH and tuxayo 15:18:56 nothing special? 15:19:12 #topic Updates from the QA team 15:19:16 19.11: Not much, catching up with the flow of patches resulting from the Catalyst academy :D 15:19:58 Not much special for QA I think 15:20:03 Please do QA! :) 15:20:12 #info Please do QA! 15:20:24 Numbers are a little too high for my liking in the queue right now and we have some things from last year still waiting 15:20:32 as ever, cait2 has set us a challenging target to try and knock her off the top spot 15:20:40 someone help me kick her off ;) 15:20:49 you can band up on me, I really don't mind :) 15:20:58 cait2 added a new entry on the agenda, "Status of roadmap projects" 15:21:03 kohaputti++ # not shying away from the bigger scary QA's 15:21:12 yes, kohaputti++ definitely worth a shoutout 15:21:12 I will try to prepare it for the next dev meeting, asking the different groups where they are 15:21:22 #info Fridolin Somers, Biblibre, France 15:21:28 he should get extra points.. 15:21:28 i think maybe we could do some quick ones for the ones that have people present? 15:21:34 * fridolin nothing special 15:21:39 not going through all of them 15:22:18 can try 15:22:26 #topic Status of roadmap projects 15:22:34 #link https://annuel.framapad.org/p/koha_21.05_roadmap 15:22:46 #topic Improve the reliability of transfers 15:22:51 ashimema: ? 15:23:16 #info Joonas has QA'd the Phase One tree :) 15:23:19 There is a huge pile of things that is hitting my QA 15:23:21 queue* 15:23:52 #info Looking for someone to QA the final bug on that tree which adds a nice 'Transfers to send' report. 15:24:14 #info Phase one adds all the methods we need for transfers to the Koha:: namespace, with clear uses and workflows for each 15:24:20 I can try and review it next week 15:24:30 #info cait to check the old transfer bugs to give the new work some serious testing once pushed 15:24:34 will be mostly off the week after, so... you know what to do 15:24:38 #info Phase two will remove the last remaining cases where the C4 methods are used and drop those methods. 15:24:49 thanks Joubu :) 15:24:59 #toic 15:25:03 #topic Performance investigation 15:25:09 ? 15:25:23 It's related to using the task queue for ES reindex 15:25:29 :) 15:25:49 I need to redo some benchmarks and see if it lower the difference in exec time 15:25:50 #info First major identification lead to the backgrounding of the ES indexer bug.. work ongoing :) 15:26:11 thx for doing the stats, it's really interesting 15:26:21 Joubu++ 15:26:27 it almost burn my laptop everytime I run the benchmark 15:26:42 I bought a new CPU fan, so can continue the benchmark, but I am a bit hesitant :D 15:26:47 stick it in a fridge 15:26:51 #topic Git repository shrinking 15:26:52 don't kill the laptop 15:27:08 lol 15:27:17 I need to start this one! I am really excited about it 15:27:30 certainly in March 15:27:46 :) 15:27:48 #topic Release tools improvements 15:28:03 ashimema and me, we need to focus on that one, maybe closer from the release, like April 15:28:11 #topic Backup splitter on a separate server 15:28:13 we have it! 15:28:15 brill 15:28:18 :) 15:28:21 awesome :) 15:28:25 https://splitter2.koha-community.org 15:28:36 still need to sync the DB from the original server however 15:28:42 #info Splitter backup is now available at https://splitter2.koha-community.org 15:28:50 ah yes 15:28:56 #link https://splitter2.koha-community.org 15:28:57 confetti! 15:28:57 o/`'`'`'`'`'`'`'`'`'`'`'` 15:29:00 I've added few commits, like the "Latest patch from" date on https://splitter2.koha-community.org/files 15:29:06 to tell us if it's up-to-date or not 15:29:18 nice 15:29:21 shoudl we add it to the wiki page yet? 15:29:29 I will and it will be done 15:29:41 ok 15:29:47 I haven't communicated on it yet as the DB is still not updated 15:30:03 (must be updated every night, I am in touch with rangi) 15:30:10 #topic Improve our "news" module 15:30:27 sec 15:30:37 There are some (private) discussions, we are in a FQA state for now 15:30:55 the idea is to start with what is submitted yet, then improve a bit more the UI 15:30:59 #info we discussed the different uses cases and probably a new naming scheme for the future, but moving forward with the move to the Koha namespace first 15:31:09 like we should rename this, from "news" to.. "additional contents"? 15:31:19 I'll try to get to look at the current submission again soon 15:31:23 more info coming in the next weeks certainly 15:31:32 +1 15:31:34 ashimema: I FQAed it last week 15:31:49 but you can fix if you don't want to wait for Josef :) 15:31:58 I Josef aware and looking at fixes..? 15:32:07 that's kinda what I meant.. take a look, see if he needs a hand 15:32:13 I sent him a ping 15:32:19 :) 15:32:22 i guess it would be appreciated 15:32:44 soooo many projects on the go 15:32:51 (I did CC you ashimema! :D) 15:32:54 but some are moving, that's nice :) 15:33:01 ah yes.. so you did 15:33:03 #topic Fix OPACHiddenItems [Bug 24403] 15:33:03 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=24403 normal, P5 - low, ---, koha-bugs, NEW , [OMNIBUS] OpacHiddenItems should hide items everywhere in the OPAC 15:33:06 tcohen: around? 15:33:40 there are curently no patches to test for this one I am aware of 15:33:41 it's ongoing, a bit slow right now but the first step has been pushed already 15:33:54 waiting for tcohen to write more things to test! 15:34:02 i think that sums it up 15:34:03 #topic Handling of errors 500 15:34:06 that's done 15:34:11 (and that's awesome) 15:34:15 should we mark it in the reaodmap? 15:34:23 see bug 26048 15:34:23 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=26048 normal, P5 - low, ---, dcook, Pushed to master , PSGI Koha does not use custom ErrorDocument pages 15:34:31 it's almost done :) 15:34:39 we need to make it pretty now 15:34:57 see bug 27555, bug 27556 and bug 27557 15:34:57 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27555 enhancement, P5 - low, ---, koha-bugs, NEW , Use psgi_env from errors/* controller scripts 15:34:58 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27556 enhancement, P5 - low, ---, koha-bugs, NEW , Improve error pages 15:34:59 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=27557 enhancement, P5 - low, ---, koha-bugs, NEW , Add the ability to display the error that caused a 500 15:35:12 waiting for dcook 15:35:24 (At least I am waiting for him) 15:35:27 :) 15:35:32 #topic Developer handbook and documentation 15:35:36 this is.. stuck 15:35:43 devs don't want to write documentation 15:35:53 I think that's the conclusion 15:35:57 (we knew that already) 15:36:10 * ashimema just has too many projects on.. 15:36:21 I think there was some discussion about how to use a different tool than the wiki maybe 15:36:25 yes, that's a big one, and we don't really know where to start 15:36:36 rm wiki; mkdir wiki 15:36:47 i proposed a fresh start a whlie ago 15:36:49 I liked the advent calendar.. but I feel we took on too much too late with that 15:37:01 once i get a bit more free time i could try to do some wiki weeding at least 15:37:15 it was still very worthwhile 15:37:16 perhaps we could do a series similar to it.. but with a clearer timetable that commits us to writing stuff.. but not all at once 15:37:40 it would be nice to see these kinds of things also linked to the website somehow 15:37:44 maybe we should ask the dev list? 15:37:45 it was on social media, but harder to find there 15:38:00 on which topics you need more tech info/examples? 15:38:04 need to get at least some links on the official pages 15:38:12 i tihnk Rest API 15:38:13 it was very much a first go.. I'd love to see us do it again next year.. but with more planning and more officially 15:38:17 if we get no answer, we can forget about it :D 15:38:23 but ithink the calendar covered that? 15:38:25 maybe 15:38:33 the information in the wiki is very scattered there 15:38:37 that was the focus area of the calendar.. 15:38:39 lots of discussion things where the status is not clear 15:38:40 but I've learnt allot since 15:38:43 ashimema: it's in my calendar already :D 15:38:53 I could happily add a few more articles now I've learnt some more :) 15:38:57 awesome Joubu :) 15:39:13 it might also be time to review installation guides and such? 15:39:22 the more boring things 15:39:37 I think that's the only ones that are up-to-date 15:39:52 maybe :) 15:40:07 what about a series explaining more about the Koha objects? 15:40:10 I feel the wiki upgrade has got us stuck for a very long time.. as that's still a work in progress... and may continue to be for some time to come.. I agree.. lets move the dev book to another platform 15:40:15 i'd read it... if you start really low barrier 15:40:46 ashimema: it's yet another places 15:40:49 I sort of feel the guidelines could live in such a place too.. made clearer as an apendix to such a 'book' 15:40:54 mmm 15:41:06 I am not totally against the idea 15:41:07 i think the wik woruld work, but some features currently don't work quite right 15:41:13 and i believe we shoudl tidy up some stuff 15:41:21 indeed 15:41:26 but we will copy some info from the wiki, that might add confusion 15:41:34 anyways.. happy to help when we know a direction 15:41:37 there is tons of old rfcs that never got implemented as described or at all 15:41:39 for example 15:41:39 for example is not fixed 15:41:42 it can get really ocnfusing 15:41:57 the wiki is a mess.. 15:41:58 we need to obsolete those pages 15:42:07 yes 15:42:11 ok 15:42:14 Yes, I am stuck with a death of a former employer leaving a mountain of books and other things to address for the past few months, 15:42:17 I'm happy to help if I can 15:42:45 we will take over thd, we need it to be done. I will try to find some time during the next cycle 15:42:54 we could still start with a new installation and move stuff that is good maybe? I am not sure it's worth a big amount of effort really 15:43:08 +1 15:43:14 * ashimema like fresh 15:43:19 yes 15:43:26 that's for later :) 15:43:33 good to know we are 4 already 15:43:39 Slips: I've not had allot of time to focus on this one yet.. it's likely to get bits of movement soon, but will also carry on into the next cycle. Migrating the accounts slips to GetPrepparedLetter marks the start of me looking at this again. 15:43:42 moving on 15:43:45 more than 4 even i think :) 15:43:48 ;) 15:44:09 #topic Automated updates for translated manuals 15:44:27 ashimema, cait2 , caroline: did something happen? 15:44:47 no, I'm not sure we decided anything 15:45:04 i think we still lack the technical "plan" for it 15:45:07 for the automation 15:45:15 I worked to get the pull request process working more cleanly 15:45:44 that now works.. i.e if caroline goes to the secret tools page, she can request a merge-request to get the latest translations and can merge them into the manual 15:46:04 and we might want to branch out 20.11 soon? currently no translations are built at all for 20.11 15:46:08 I have to remember to do it, which I don't :( 15:46:22 as for automating.. I need to talk to rangi or wizzyrea.. so we can do regular rsyncs 15:46:27 or use bernardo's server 15:46:27 I was trying to find how to branch earlier today 15:46:37 if the translate server has the space.. perhaps we just use that.. 15:46:38 but that's another matter 15:46:40 there was the duplication between translate.k-c and manual.k-c that we needed to address 15:46:47 the community website server is struggling for space 15:47:04 )I've had several failed pipes again today) 15:47:19 there is some visual ifference in styling, but otherweise i'd be happy if we pointed to the tanslation server for the translated ones 15:47:45 maybe need to ask bernardo, as the translation server is struggling a bit too currently 15:48:11 I'm happy to get it working.. just need to know what people want.. 15:48:44 we shoudl still regularly push the po files to another repo... having copies is good 15:48:49 I'll email bernardo and ask if we can point to his translated manuals or if he has an idea 15:49:15 okies 15:49:22 next? 15:49:22 next is Koha bibliography in Zotero 15:49:30 * ashimema wants to get out for his walk 15:49:46 moving out 15:49:58 you will discuss detail off meeting :) 15:50:04 #topic Koha Cookbook Project 15:50:22 We had a meeting, setup a git project 15:50:30 next date is March, for another meeting 15:50:36 :) 15:50:58 #topic Replace CGI scripts with Mojolicious controllers 15:51:15 I haven't heard anything about that 15:51:20 #topic Write Protecting fields in Koha from being overwritten through import 15:51:40 there are movements there 15:51:43 it just got a rebase 15:51:45 yes 15:51:45 but I haven't followed closely 15:51:58 bug 14957 15:51:58 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=14957 enhancement, P5 - low, ---, glasklas, Failed QA , Write protecting MARC fields based on source of import 15:51:58 i guess bug 14957 is rebased against master I see 15:52:03 happy to help with QA there cait.. if you wanted to double team it. 15:52:04 i'll try to get in touchwith AndrewFH 15:52:14 that soudns good 15:52:26 ashimema: stop adding stuffs to you list! 15:52:32 hehe 15:52:39 yeah he is terrible at that 15:52:40 * ashimema helps with all the things 15:52:41 your* 15:52:45 "helps" 15:52:47 and i am enabling him... 15:53:02 lol 15:53:03 we will have to limit to 4 topics next cycle 15:53:06 or 5 15:53:08 or 6 15:53:08 and once done 15:53:10 .. 15:53:11 we can add more 15:53:15 no 4 is good 15:53:23 well.. I'm not the RM anymore.. so can throw myself into code a bit more again :) 15:53:24 and oly can add one from backlog, when one is done 15:53:36 finishing? 15:53:39 yes 15:53:40 4 is good 15:53:43 #topic Actions from last meeting 15:54:00 #action amoyano advertise 15522 and ask the list for feedback [after 20.11 is released], setup a sandbox, etc. 15:54:06 will stop to postpone this one I think 15:54:06 We need a clear roadmap/backlog to pick of more strictly in general 15:54:12 good plan 15:54:15 haven't got reply 15:54:45 #topic Review of coding guidelines 15:54:50 " Terminology list - proposed addition for OPAC: use OPAC instead of opac or OPAC for online public access catalog, including in system preference names. (davidnind) " 15:54:58 I don't get it 15:55:14 maybe ... or catalog? 15:55:24 pass 15:55:26 or Opac 15:55:32 i had a discussion with him on a bug about this 15:55:34 * ashimema switches to mobile 15:55:41 the rename pref AllowOPACLists one 15:55:49 maybe we can ask to clarify 15:55:57 but i think it#s about the different spellings 15:55:59 int he prefs 15:56:17 ok, will ask him and postpone 15:56:20 use OPAC instead of Opac or opac 15:56:22 #topic Set time of next meeting 15:56:40 #info Next meeting: 24 February 2021, 20 UTC 15:56:42 is that correct? 15:57:11 #endmeeting