15:01:08 <Joubu> #startmeeting Development IRC meeting 4 November 2020
15:01:08 <huginn> Meeting started Wed Nov  4 15:01:08 2020 UTC.  The chair is Joubu. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:08 <huginn> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:08 <huginn> The meeting name has been set to 'development_irc_meeting_4_november_2020'
15:01:17 <Joubu> #topic Introductions
15:01:26 <Joubu> #info Jonathan Druart
15:01:28 <Joubu> qa_team?
15:01:28 <wahanui> somebody said qa_team was cait Joubu marcelr kohaputti josef_moravec tcohen kidclamp khall ashimema alex_a jajm matts
15:01:29 <Joubu> rmaints?
15:01:29 <wahanui> somebody said rmaints was lukeG, aleisha, and tuxayo, I SUMMON THEE!!⚡⚡
15:01:44 <Joubu> #link https://wiki.koha-community.org/wiki/General_IRC_meeting_4_November_2020
15:01:48 * tuxayo emerges
15:01:49 <tuxayo> \o/
15:02:17 <tuxayo> #info tuxayo/Victor Grousset, France
15:02:53 <amoyano> #info Agustin Moyano, Theke Solutions
15:03:20 <oleonard> #info Owen Leonard, Athens County Public Libraries, Ohio, The Bad Place
15:03:33 <tuxayo> LOL
15:03:37 <amoyano> hehe
15:03:46 <tcohen> #info Tomas Cohen Arazi, Theke Solutions
15:04:10 <tcohen> oleonard++ # keeping it up
15:04:46 <ashimema> #info Martin Renvoize, PTFS Europe
15:04:47 <thd> #info Thomas Dukleth, Agogme, New York City
15:04:48 <ashimema> made it.. just
15:04:59 <Joubu> #topic Announcements
15:05:03 <Joubu> Anyone have something?
15:05:41 <Joubu> #info Election of 21.05 release team is "tonight"! 20 UTC
15:05:58 <oleonard> Finally an election I can enjoy
15:06:05 <cait> #info Katrin Fischer, BSZ, Konstanz
15:06:21 <thd> :)
15:06:42 <lukeG> #info Lucas Gass, ByWater Solutions
15:06:56 <Joubu> #topic Update from the Release Manager (20.11)
15:06:58 <tuxayo> Job offer: Release Maintainer for Koha. Salary: a lot of magical internet points
15:07:12 <tuxayo> One candidate is needed
15:07:13 <Joubu> back from vacation, lot of things has been PQAed, great job QA team!
15:07:22 <tcohen> what brach is missing a candidate?
15:07:29 <Joubu> at least those who QAed ;)
15:07:34 <ashimema> 2
15:07:44 <ashimema> 20.05 and 19.11
15:07:44 <wahanui> 39.16
15:08:02 <oleonard> Patience, wahanui
15:08:02 <wahanui> rumour has it Patience is a virtue.. some of the updates are slow be nessecity
15:08:04 <tuxayo> oldstable (20.05)
15:08:05 <ashimema> tuxayo has been great on oldoldstable
15:08:15 <Joubu> everybody should run master anyway
15:08:21 <tuxayo> +1
15:08:46 <tuxayo> 19.11: I candated, unless aleisha wants to continue it
15:08:55 <Joubu> we should ping someone at Catalyst
15:08:55 <tuxayo> *candidated
15:09:03 <Joubu> to know if they want to continue
15:09:05 <tcohen> I volunteer for 20.05
15:09:08 <Joubu> what about lukeG?
15:09:16 <tuxayo> tcohen++
15:10:01 <lukeG> Joubu: I would like to have a break from Rmaint, I've done 4 cycles now
15:10:01 <tcohen> I would prefer to give new members the chance
15:10:07 <ashimema> I can back you up if you want to share at all tcohen.. can't commit full time but happy to share the load
15:10:13 <ashimema> yeah.. would be nice to have new blood though.
15:10:15 <Joubu> hehe lukeG, yes, fair enough!
15:10:17 <tcohen> cool
15:10:22 <tcohen> ashimema++
15:10:28 <tcohen> I will sign for 20.05
15:11:05 <ashimema> tcohen++
15:11:06 <Joubu> kidclamp: ES topic expert?
15:11:20 <Joubu> matts: CAS/Shib topic expert?
15:11:44 <Joubu> then we are good, great team once again
15:11:46 <ashimema> I would say those ones are 'nice to have', not as vital though..
15:12:12 <ashimema> :)
15:12:16 <matts> Joubu, okay for me !
15:12:21 <Joubu> so, I was saying, lot of things have been PQA, I am processing them all and hopefully will be "done" (pushed or not) before end of the week
15:12:46 <Joubu> no new enhancement will be pushed after this week
15:13:04 <Joubu> I hope we will manage to fix all the bugs next week!
15:13:08 <cait> i expect quite a bit of conflicts
15:13:11 <cait> and maybe needed follow-ups
15:13:21 <cait> is there some room we could give devs there?
15:13:37 <Joubu> Then, please add the release notes to your bugs. I will add the bz tags next week
15:13:57 <Joubu> We will also need to generate the technical release notes. I will ask you for help about that as well
15:14:17 <Joubu> next is the packages, we must continue the discussion we had on the dev list 2 weeks ago
15:14:36 <cait> tcohen: there is a term for people accumulating roles :)
15:15:05 <amoyano> he
15:15:07 <Joubu> cait: it will depend on the size of the patch set. I am pushing a lot of things and I am afraid of the time we will need to stabilize all that
15:15:17 <cait> some of the bigger stuff is currently sitting
15:15:25 <ashimema> I aim to lend a hand on release notes and swatting bugs the next few weeks.. at least a couple of hours a day
15:15:36 <cait> sending ILL emails for example
15:15:56 <Joubu> I am spending 90% of my work time on the PQA bugs
15:16:03 <Joubu> cannot do more
15:16:49 <Joubu> I am doing my best to give everybody the necessary feedback to have their bugs pushed to master ASAP
15:16:56 <oleonard> Joubu++
15:16:58 <Joubu> any questions?
15:16:59 <ashimema> Joubu++ You're doing great.. it was a BIG queue
15:17:00 <tcohen> cait I don't like doing it, actually
15:17:07 <tcohen> Joubu++
15:17:09 <cait> i know, i was just wondering if there might be a way to add a little room/time as i thnk it will be a close call
15:17:10 <koha-jenkins> Project Koha_Master_U2010 build #39: STILL UNSTABLE in 45 min: https://jenkins.koha-community.org/job/Koha_Master_U2010/39/
15:17:38 <Joubu> cait: yes, it won't be strict for the things PQA already
15:17:41 <cait> sorry, if that came out wrong
15:18:04 <Joubu> nope :)
15:18:14 <Joubu> #topic Updates from the Release Maintainers
15:18:20 <cait> maybe have people check their stuff for rebasing
15:18:26 <cait> could help remove some extra work/time lost
15:19:04 <tuxayo> #info 19.05: last security release was without issue
15:19:05 <Joubu> #info If you have PQA bugs that have been FQA or needs a rebase, provide feedback ASAP to have them reach 20.11!
15:20:17 <Joubu> ho, I forget, D11 is completely broken, see bug 26893, must not be used!
15:20:17 <huginn> 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=26893 blocker, P5 - low, ---, koha-bugs, NEW , New version of JSON::Validator (D11) break our REST API routes
15:20:47 <Joubu> #info Koha under Debian 11 is broken, it must not be used until bug 26893 is fixed!
15:20:54 <Joubu> we need this fixed for 20.11
15:21:25 <Joubu> #topic Updates from the QA team
15:21:38 <tuxayo> #info about RMaint I did a first pass on the release maintenance wiki page to complete and fix the security process. Needs a second one and proofreading from aleisha and lukeG
15:22:09 <lukeG> tuxayo: i'll have a look later today
15:22:36 <tuxayo> lukeG: thanks, I'll try also to do the 2nd pass
15:22:49 <Joubu> cait: something to add for the QA team topic?
15:22:58 <cait> yes
15:23:29 <cait> I am close to 1.000 QA in one year and it's not a reason for celebration
15:23:34 <cait> I can't keep this up
15:23:38 <tuxayo> :o
15:23:48 <cait> and I need everyone on the QA team signing up to step up some more next cycle
15:23:58 <ashimema> cait++
15:24:00 <cait> at the moment the load is mostly on ashimema and me
15:24:01 <ashimema> team--
15:24:03 <cait> ashimema++
15:24:06 <cait> and we need to fix that
15:24:29 <ashimema> I agree.. we need to spread the load more next release.. it's unsustainable as it is.
15:25:03 <cait> I know queue looks good right now, but as soon as 2 people step off for a day or 2 numbers are rising really fast
15:25:37 <Joubu> cait: do you get feedback when you ask for in your weekly email?
15:25:50 <cait> most of the time I don't
15:26:14 <Joubu> I think that's the mail problem. People should show up when we need them
15:26:22 <cait> i will still keep sending htem as i got positive feedback in general
15:26:40 <ashimema> I read the weekly mails and try to attack those as a high priority.. but often I find I'm locked out of at least some of them
15:26:47 <cait> it#s very time consuming to ask individual members to do things
15:26:55 <ashimema> I find the qa emails really helpful personally
15:26:59 <cait> i know it's human... and that it works
15:27:41 <cait> so maybe we need new ideas
15:28:13 <cait> best solution would be more time and people
15:28:32 <ashimema> perhaps we could mentor some new QA people?
15:29:08 <tuxayo> I was just writing that I'm feeling conformable to try QA now ^^
15:29:08 <lukeG> I would be interested in helping/learning
15:29:09 <ashimema> if there's anyone out there at all interested in doing QA.. or anyone anyone here thinks might be good at QA then feel free to apply and I'd be happy to guide on the process/mentor people
15:29:10 <thd> The best solution for most things is more available time.
15:29:10 <Joubu> we have 2 new members for 21.05
15:29:12 <Joubu> which is good
15:29:28 <ashimema> agreed
15:29:40 <cait> yes, but we started out quite well number wise this cycle, it's hard to predict
15:29:48 <cait> I am fine if people don't have many hours to spend
15:29:48 <Joubu> alex_a: not candidate to be a QA member for 21.05?
15:30:01 <cait> but having some idea when they are available and some kind of consistency would be great
15:30:02 <ashimema> also.. the QA queue is often full of easy starter bugs.. one's that don't take hours.. you can be on the QA team and be picky about what you choose to QA.. it's a good way to start
15:30:15 <cait> for sure
15:30:20 <ashimema> and it free's up other QA team members to look at some of the more narly and difficult to QA patches
15:30:25 * cait recommends oleonard's for starters
15:30:28 <cait> :)
15:30:30 <tuxayo> ashimema: alright, sound good for me :)
15:30:35 <ashimema> I'd love to see someone nick all the easy patches and overtake me on the leaderboards ;P
15:30:38 <Joubu> to be honest, the main problem right now is the NSO queue..
15:30:50 <cait> we have a general flow issue
15:30:51 <ashimema> That is also true
15:31:21 <ashimema> I ran a youtube 'live coding' session just before this meeting.. first half was about how to pick bugs to SO and then how to SO them using sandboxes..
15:31:21 <cait> and a lot of  the thigs gettings stuck are stuck because they are not easy to test for people with sandboxes or require dev knowledge
15:31:37 <cait> so a good topic for dev meeting - we need devs on this queue too
15:31:39 <ashimema> I'd be happy to run similar things regularly if it would help onboard people
15:32:08 <lukeG> since I won't be Rmaint'ing I hope to dedicate much more time to NSO
15:32:13 <cait> I hope this doesn't come off as too negative, but if we don't talk about... we can't change it
15:32:30 <ashimema> lukeG++
15:32:42 <ashimema> totally a worthwhile thing to discuss cait
15:33:16 <cait> lukeG: sounds great :)
15:33:22 <ashimema> as for sandboxes (* OK, they're broken right now but I'm working on that).. if there's something you can't test on them yet let me know and I can work on that
15:33:36 <ashimema> mail testing does work for example.. that's a recent addition
15:33:43 <ashimema> but perhaps we need to guide people on how
15:33:52 <kohaputti> I don't know if it is really problem that stuff is stuck in NSO queue, if things are important to more than just person then people will sign-off, not every feature needs to get in my opinion, and giving feedback on all features suggested always might not be time well spent
15:34:24 <cait> kohaputti: maybe true for a part of the queue, but there are also bug fixes and things that just aren't easy to test
15:34:38 <Joubu> kohaputti: I agree, but having the queue "clean" makes things easier to pick
15:34:49 <cait> yeah, I beleive that too
15:34:55 <cait> a not so daunting queue will be more inviting
15:35:18 <tuxayo> ashimema: «I ran a youtube 'live coding' session» Do you mean you recorded it for publication?
15:35:32 <ashimema> I would hope bugwranglers can help with keeping the queue clean.. a certain amount of that role is not to test, but to just take a look at the bugs and comment on whether they're valid, a good idea, need more thought etc
15:35:41 <ashimema> yup
15:36:12 <ashimema> https://www.youtube.com/watch?v=0Q0KrFv_YnU
15:36:22 <ashimema> we need to edit it a little still.. but it's there
15:36:44 <ashimema> just need to cut off the first 5 minutes of us getting ourselves set up
15:36:44 <Joubu> #info QA team - The project is having a long standing issue about the QA queue. We should have more people involved in the process. It will especially help if you are kind of expert in a given area, let us know!
15:36:48 <Joubu> something like that..?
15:36:53 <tcohen> +1
15:37:01 <tuxayo> ashimema++ will watch, might learn a few things :D
15:37:47 <ashimema> all very informal.. but hopefully that also means it's a bit less scary for people.. shows we're all human
15:38:20 <ashimema> brill Joubu
15:38:25 <Joubu> so basically, I have nothing to suggest, cait, sorry. But hopefully amoyano and dcook will bring some fresh air :)
15:38:35 <henryb> ashimema: I liked all the deliberate mistakes :)
15:38:43 <Joubu> # #action amoyano QA all the things!
15:38:50 <cait> thx
15:38:51 <ashimema> haha.. only one was actually deliberate ;)
15:38:54 <Joubu> # #action dcook QA all the things!
15:38:56 <amoyano> he, I'll try
15:39:03 <ashimema> the others were me being human
15:39:07 <cait> ashimema: it looked planned to me .)
15:39:26 <Joubu> moving on?
15:39:36 <cait> ok
15:39:39 <ashimema> :)
15:39:44 <Joubu> #topic Actions from last meeting
15:39:52 <Joubu> #topic amoyano Will try to provide a Vue alternative to bug 15522
15:39:52 <huginn> 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=15522 enhancement, P5 - low, ---, jonathan.druart, Patch doesn't apply , New interface for revamped circulation rules
15:39:57 <Joubu> you did it!
15:40:02 <tuxayo> amoyano++
15:40:04 <amoyano> yep!
15:40:04 <cait> amoyano++
15:40:12 <ashimema> amoyano++
15:40:14 <tcohen> amoyano++
15:40:15 <Joubu> amoyano: can you tell us something about it?
15:40:29 <Joubu> What's the next step? Should we ask the list for feedback?
15:40:59 <amoyano> If you checked it out, it's not exactly a copy of the proposal made with react
15:41:06 * ashimema is testing it.. but it's slightly behind keeping on top of the bugs bug queue at the moment.. and rebasing code for Joubu... I hope to do allot more testing and review first thing next cycle
15:41:22 <tcohen> there's a UX change
15:41:29 <ashimema> so.. thanks for submitting amoyano.. apologies I've not fed back yet
15:41:31 <amoyano> sorry if that mess things up, but I though it was a good time to introduce that change
15:42:10 <tuxayo> amoyano: that's relevant no worries
15:42:25 <amoyano> I believe it's simple to read..
15:42:33 <Joubu> #info amoyano wrote an alternative for the new circulation rules interface - code is available on bug 15522
15:42:38 <ashimema> so.. next step I think is to probably wait a couple of weeks (whilst we're all super busy with release), then go on a drive to find testers.. thinking mailing list, pre-setup sandboxes etc
15:42:43 <amoyano> vue packages modules in a single file, (.vue files)
15:42:57 <tuxayo> #info feedback is needed on the code and UX changes
15:43:14 <ashimema> code does look clean to me.. in what I have managed to read so far
15:44:11 <Joubu> #action amoyano advertise 15522 and ask the list for feedback [after 20.11 is released], setup a sandbox, etc.
15:44:45 <amoyano> I'll have to watch achimema's video to setup a sandbox, he!
15:45:25 <Joubu> amoyano: nope, you will need to catch him to set it up for you and have it "eternal"
15:45:31 <Joubu> or it will be erase after X days
15:45:33 <ashimema> haha.. I can help with that.. it may be one of those bugs that needs a little more behind the scenes to get it working on a sandbox
15:45:35 <Joubu> d
15:45:45 <ashimema> indeed
15:45:48 <Joubu> next topic was about the same things
15:45:54 <Joubu> #topic ashimema Setup a sandbox with lot of circ rules (ideally from production data) to test 15522. Then ask the list for feedback.
15:46:01 <Joubu> so skipping this one
15:46:06 <ashimema> I did.. then I broke the sandboxes :(
15:46:15 <ashimema> so I'm back to square one and need to do it again
15:46:21 <tuxayo> :o
15:46:21 <ashimema> but yes.. wait for next cycle now
15:46:21 <Joubu> I am postponing the 2 next ones, David is working on them
15:46:31 <Joubu> #action davidnind Find and update relevant places to record Perl version required, including release notes, manual and wiki (one source of the truth) (still working on this)
15:46:41 <cait> i thin having both versions testable (GUI wise) woudl be great
15:46:44 <Joubu> hum, only one actually
15:46:44 <cait> and advertise that
15:47:08 <Joubu> yes, definitelly, we may need to pick a bit of both
15:47:15 <Joubu> #topic davidnind to add information about what to backport to the release maintainers role page (as a starting point for further notes and guidance) (initial information added - see Release maintenance page)
15:47:24 <Joubu> #link https://wiki.koha-community.org/wiki/Release_maintenance#Deciding_on_what_to_back_port
15:48:25 <tuxayo> davidnind++ this is great!
15:48:33 <tcohen> davidnind++
15:48:38 <ashimema> good call
15:49:14 <ashimema> davidnind++
15:49:27 <Joubu> #info davidnind wrote a wiki page to list the general principle for backporting - https://wiki.koha-community.org/wiki/Release_maintenance#Deciding_on_what_to_back_port
15:49:35 <Joubu> @later tell davidnind thanks David!
15:49:35 <huginn> Joubu: The operation succeeded.
15:49:53 <Joubu> #topic ashimema to write a guideline to forbidding the use of type "number"
15:49:54 <lukeG> davidnind++
15:49:57 <Joubu> #link     https://wiki.koha-community.org/wiki/Coding_Guidelines#ACC2:_Input_type_.22number.22_should_be_avoided
15:49:58 <ashimema> done :)
15:50:02 <Joubu> yup
15:50:16 <ashimema> feel free to point out amendments if anyone wants to :)
15:50:25 <ashimema> but hopefully it's fairly clear and concise
15:50:46 <amoyano> ashimema++
15:50:56 <cait> hm should we create an omnibus bug to check/fix codebase?
15:51:04 <Joubu> and we have the QA test: https://gitlab.com/koha-community/qa-test-tools/-/merge_requests/28
15:51:09 <ashimema> not a bad idea cait
15:51:11 <Joubu> so we can consider this done
15:51:14 <Joubu> thanks ashimema
15:51:15 <tuxayo> ashimema++
15:51:21 <tuxayo> didn't knew about inputmode
15:51:23 <ashimema> lots of academy bug potential there
15:51:50 <cait> and lots of things i no longer will have to change locally :)
15:52:05 <Joubu> #info New coding guideline -  Input type "number" is now forbidden, use inputmode=numeric instead. See https://wiki.koha-community.org/wiki/Coding_Guidelines#ACC2:_Input_type_.22number.22_should_be_avoided
15:52:21 <ashimema> I can take that action.. to create the omnibus bug
15:53:01 <Joubu> #action ashimema Open an omnibus bug to fix the existing occurrences of input type "number"
15:53:01 <ashimema> #action ashimema to add an omnibus bug to retroactively fix inputs for ACC2 coding guideline
15:53:04 <ashimema> :)
15:53:15 <Joubu> #topic General development discussion (trends, ideas, ...)
15:53:21 <Joubu> #topic  Discuss Bug 20410, Remove OpacGroupResults system preference and feature -- oleonard
15:53:21 <huginn> 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=20410 enhancement, P5 - low, ---, koha-bugs, NEW , Remove OpacGroupResults system preference and feature
15:53:41 <oleonard> So... Does anyone know if this feature is still used?
15:54:03 <cait> I haven't heard of a working implementation in quite a long while
15:54:15 <cait> and never seen a production one
15:54:29 <cait> do we have it on hea?
15:54:38 <oleonard> Every time the OPAC gets a big update I dutifully make a best effort to update opac-results-grouped.tt but I'd like to not do that anymore :)
15:54:44 <Joubu> yes, 9 occurrences for 1/yes
15:55:08 <cait> i thin we should announce deprecation
15:55:15 <tuxayo> The 0 value means yes? :o
15:55:17 <cait> and remove in 21.05
15:55:26 <cait> there are 9 with 1 :)
15:55:59 <Joubu> oleonard: go for it!
15:56:26 <Joubu> #action oleonard ask the list if someone use OpacGroupResults somewhere...
15:56:40 <oleonard> I don't remember the last time we announced the deprecation of a feature. Does anyone?
15:56:51 <Joubu> zebra?
15:56:51 <wahanui> hmmm... zebra is running?
15:56:55 <ashimema> I did it once
15:56:55 <Joubu> (lol)
15:57:00 <ashimema> can't remember what
15:57:02 <tuxayo> lol
15:57:07 <Joubu> we did it few time
15:57:15 <Joubu> non xslt view
15:57:16 <kohaputti> ashimema, you deprecated some C4 function
15:57:23 <kohaputti> maybe couple years ago
15:57:30 <ashimema> I wouldn't "ask if anyone is using it".. I would "Announce wwe intend to remove it" and then see if anyone complains ;)
15:57:41 <cait> yep
15:57:46 <tuxayo> I suppose dead external services don't count?
15:58:16 <Joubu> we are reaching 1h, moving on
15:58:21 <Joubu> #topic  Sign off and tests failing - question: Recently I've had several cases where the tests pass before and after the patches are applied. However, if system preferences are changed as part of the test plan and tests are run at the end then they fail. If the developer says this is fine (for whatever reason), should we accept that when signing off and then leave to QA to critique their work? Example (not
15:58:29 <Joubu> picking on this one in particular, it's the only one I can remember): Bug 22690. --davidnind
15:58:29 <huginn> 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=22690 normal, P5 - low, ---, ere.maijala, Failed QA , Merging records with many items too slow (Elasticsearch)
15:58:36 <ashimema> I've used https://metacpan.org/pod/Mojo::Util#deprecated in a few places to deprecate routines
15:58:42 <ashimema> manualcredit was one I think
15:59:01 <Joubu> The short answer is "it depends"
15:59:45 <Joubu> ideally tests should pass in whichever situation
15:59:46 <ashimema> it's a hard call.. in the best case all unit tests should be idempotent and do all their own setup
16:00:09 <Joubu> and mock "everything". Like: if you need this specific pref to be set to have the test pass, then mock the pref's value
16:00:10 <ashimema> if I hit such a situation during QA I tend to have a go at fixing it
16:00:11 <kohaputti> I'm under the impression if you remove like some default branches, etc. then maany tests will fail
16:00:17 <ashimema> exactly
16:00:28 <ashimema> correct kohaputti
16:00:29 <Joubu> but we use to accept tests if they pass on top of the sample data / default value
16:00:38 <ashimema> but for new tests it's important we try to be better ;)
16:00:53 <tuxayo> > do all their own setup
16:00:53 <tuxayo> Here, does is "just" mean that the test lacks setting the syspref in it's setup?
16:00:59 <Joubu> kohaputti: I don't think so, we took care of that
16:01:11 <Joubu> we do not longer use hardcoded data
16:01:17 <Joubu> like "branchcode=CPL"
16:01:40 <kohaputti> ok
16:01:43 <Joubu> the problem is more about the default values IMO, and so the syspref
16:01:45 <Joubu> s
16:01:50 <ashimema> also things like counting rows before and after the test.. rather than hard coding a number of rows we expect from the test data
16:01:55 <ashimema> that's a pattern now
16:01:56 <kohaputti> at least for new tests this might be a good idea
16:02:12 <Joubu> also: don't delete the data before the tests!
16:02:15 <ashimema> indeed... hense it being a judement call :)
16:02:19 <kohaputti> but existing tests, I don't know, could be potentially many many hours of work to get some bug fix, etc. in
16:02:25 <Joubu> most of the time it's the wrong way :)
16:02:48 <ashimema> yup.. QA team use your judement a bit basically ;)
16:02:53 <cait> should we have some notes about good pattersn in unit testing writing somewhere?
16:02:55 <ashimema> it comes with practice
16:02:57 <kohaputti> The current koha unit tests sound more like integration tests to me :D
16:03:05 <cait> if we intend to improve them, better make sure new people have some guidance
16:03:09 <ashimema> agree :(
16:03:19 <Joubu> kohaputti: yes, they are a bit of everything
16:03:24 <ashimema> well.. they're a real jumbled mix really
16:03:27 <ashimema> haha.. snap
16:04:03 <Joubu> ha but wait
16:04:15 <Joubu> Is David's question about searchengine?
16:04:19 <kohaputti> yes it is
16:04:26 <Joubu> so
16:04:30 <Joubu> bug 26250
16:04:30 <huginn> 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=26250 major, P5 - low, ---, jonathan.druart, Pushed to master , Test suite does not pass if Elastic is used as search engine
16:04:43 <Joubu> the test suite must pass with searchengine set to ES
16:05:13 <Joubu> We did not setup the new job on Jenkins already
16:05:21 <Joubu> (or make it the default)
16:05:45 <kohaputti> ok but it doesn't, is it the job of the patch writer to fix it that touches some unrelated test in the same file that has the bug currently?
16:06:11 <Joubu> after 26250 the whole test suite must pass with ES set
16:06:15 <Joubu> if it's not, it's a bug
16:06:27 <Joubu> open a separate bug report and link it with 26250
16:06:59 <Joubu> talk about it later if needed, moving on
16:07:03 <Joubu> #topic Review of coding guidelines
16:07:16 <Joubu> #topic Proposal for uniform response code for duplicate errors on the API (vote)
16:07:30 <Joubu> tcohen: ? Is that you?
16:09:01 <Joubu> we cannot vote on that without someone explaining it
16:09:33 <amoyano> pity.. seemed interesting
16:09:41 <Joubu> "When a create operation fails due to duplicate object, it must return 409."
16:10:22 <Joubu> https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/409
16:10:25 <ashimema> hmm
16:10:27 <Joubu> The HTTP 409 Conflict response status code indicates a request conflict with current state of the target resource.
16:10:30 <Joubu> Conflicts are most likely to occur in response to a PUT request. For example, you may get a 409 response when uploading a file which is older than the one already on the server resulting in a version control conflict.
16:10:34 <Joubu> not exactly the same
16:11:37 <Joubu> well, if it's the code to return, let's return it
16:11:42 <Joubu> not sure we should vote for that
16:11:51 <ashimema> indeed
16:12:11 <ashimema> so more of a mention to make sure QA catch it
16:12:22 <Joubu> #topic  Proposal for related object ids on the API (vote)
16:12:36 <Joubu> "Fields containing a key for a related object should be named related_id where related is the related object name and if called via embed should be replaced by just the related object name. "
16:13:08 <Joubu> and this one makes sense, go for it tcohen!
16:13:17 <amoyano> not sure that applies in every case
16:13:28 <Joubu> example?
16:13:51 <amoyano> if you've got more than one FK to another table you couldn't do it
16:14:15 <Joubu> yes
16:14:24 <Joubu> for instance suggestion, manager, suggester, etc.
16:14:30 <Joubu> but then it's manager_id, suggester_id
16:15:00 <Joubu> is then the rule wrong?
16:15:10 <amoyano> ok, then that's ok.. I thought it ment to place the <table name>_id
16:15:11 <Joubu> "related object name" is "patron"
16:15:31 <ashimema> I think it reads good
16:15:46 <ashimema> although it not always 'Object' name either
16:15:51 <ashimema> it's more the 'relation' name
16:15:56 <Joubu> In my understanding the suggestion.managed_by will become manager_id, not patron_id
16:16:13 <ashimema> so 'manager_id' and 'manager' is the relation name.. the final object is a Koha::Patron
16:16:14 <Joubu> yes, that's it
16:16:30 <ashimema> correct
16:16:31 <Joubu> can you update the guideline please?
16:16:34 <ashimema> that's my understanding too
16:16:39 <ashimema> yup
16:16:43 <Joubu> maybe with the example
16:16:53 <amoyano> great.. scratch from records everything I said, please :)
16:17:03 <Joubu> good catch amoyano
16:17:21 <ashimema> very good catch amoyano
16:17:25 <amoyano> hehe
16:17:30 <amoyano> thanks
16:17:36 <ashimema> it brings the guidline directly above into the mix to make sure they agree with each other
16:17:50 <Joubu> #topic Set time of next meeting
16:17:59 <ashimema> I think we agree on the approach so I'll update it to be clear
16:18:52 <Joubu> #info Next meeting: 18 November 2020, 15 UTC
16:19:05 <Joubu> don't forget a carrot cake for oleonard
16:19:17 <Joubu> and, peel the carrot, peel it hard
16:19:25 <Joubu> #endmeeting