IRC log for #koha, 2023-02-24

All times shown according to UTC.

Time S Nick Message
00:56 marie-luce joined #koha
01:42 eythian perl -E 'my ($year) = "General_IRC_meeting_7_December_2022" =~ /(\d{4})/; say $year;'
01:43 eythian tuxayo: ^
01:58 tuxayo eythian++
01:59 tuxayo I tried this but didn't know about my ($year) so I got 1 instead of 2022
02:06 tuxayo oh, there is also /(\d{4})/ vs /\d{4}/g for this case 😯
02:08 eythian hm, I usually consider /g for looping. I guess it'd work here, but it feels a bit odd compared to an explicit capture.
02:10 tuxayo ok, noted :)
03:16 tuxayo @later tell caroline wiki updated post last general meeting. Calendar was already up to date :)
03:16 huginn tuxayo: The operation succeeded.
06:13 fridolin joined #koha
07:42 thibaud_g joined #koha
08:01 cait joined #koha
08:02 alex_ joined #koha
08:02 alex_ Bonjour
08:02 wahanui salut, alex_
08:02 lds joined #koha
08:11 cait good morning #koha
08:12 ashimema morning
08:42 cait it looks like it might still not be possible to apply patches on sandboxes
08:43 cait oh wait
08:43 cait this was tested on Biblibre ones - tehy are not fixed yet, right?
08:45 cait I'll ask them to test on PTFS-E ones next
08:48 Joubu they should be fixed
08:53 cait when did you fix? it doesn't work for the test30162 one
08:53 cait or maybe it makes a difference if you use the actions insted of spinning it up with the patch already?
08:54 cait I'll just try it with a fresh one right now
09:06 Joubu I applied the fix yesterday iirc
09:11 cait not working
09:11 wahanui not working is very boring but I'm making myself not do work (well I was until today)
09:11 cait Joubu: you can see it at mine on biblibre
09:12 cait I try to Action > apply patches > 30162
09:20 cait Joubu: say something, even if it is not nice :)
09:27 fridolin left #koha
09:30 PedroAmorim[m] hi morning
09:30 cait hi PedroAmorim[m]
09:30 cait I think I already sent eveyone else into hiding
09:42 Joubu cait: looking now
09:49 cait Jouub++
09:52 marcelr joined #koha
09:52 marcelr o/
09:58 Joubu cait: - should be fixed
09:59 cait do you want me to test?
10:00 cait guess it needs a rebase now
10:00 Joubu you can test, but it's fixed :)
10:01 cait fatal: previous rebase directory .git/rebase-apply still exists but mbox given. Patch left in /tmp/Bug-30162---XSLT-has-broken-link-for-traced-series-c3YBAo.patch Apply? [(y)es, (n)o, (i)nteractive]
10:01 cait maybe a fresh one... or needs rebasing
10:02 Joubu ashimema:[…]merge_requests/71
10:02 Joubu cait: this is a "does not apply" problem, not coming from the sandbox
10:02 ashimema I could have swarn that was already merged
10:03 cait Joubu: i know
10:04 Joubu ashimema: Couldn't have been, I submitted 2 min ago ;)
10:05 ashimema yeah
10:05 ashimema it's almost the same as a bunch of other recent fixes
10:05 ashimema that's how I missed it
10:05 ashimema it's on ptfs-e sanboxes now and merged
10:05 ashimema I tested
10:05 ashimema thanks Joubu
10:07 cait I'll forward the good news in a bit :)
10:07 ashimema do we need the same treatment for the updatedatabase ?
10:08 ashimema in the ansible create instance we run updatedatabase as the kshell user.. in the sandbox manager we run it as root
10:08 Joubu I am usually run updatedatabase as root
10:08 Joubu running*
10:08 ashimema coolios.. so I think it should work with eiher
10:10 ashimema I reckon we do need to fix 'Add git remote {{ GIT_REMOTE }} as {{ KOHA_INSTANCE }} in docker container'
10:10 ashimema no idea how often that option is used
10:15 Joubu I am using it for the erm sandbox for instance
10:15 Joubu it's useful if you want to point to a remote branch
10:15 ashimema yeah, I agree
10:15 ashimema I've not tested it since all the ktd changes.
10:16 ashimema however.. on the ptfs-e sandboxes I already use 'delegate_to' inside our crate-sandbox ansible runbook
10:16 ashimema that might be fixing the issues already
10:17 ashimema interesting
10:17 ashimema we do both for updatedatabase..
10:17 ashimema mwah..
10:17 ashimema I'm not going ot get distracted now
10:18 Joubu FYI I have "[LOCAL] Issue #81 - Run things as the koha user"
10:18 Joubu that does things like
10:18 ashimema me too
10:18 Joubu - name: "Update koha git repo for sandbox"
10:18 Joubu +      become: yes
10:18 Joubu +      become_user: debian
10:18 ashimema yup.. I wrote it 😜
10:18 ashimema though for our case it's a 'koha' user
10:18 ashimema rather than 'debian'
10:19 ashimema I think I recently asked for thoughts on what we should do to get that merged up
10:19 ashimema clearly it needs to work for all three of us and right now it doesn't without a bit of extra localisation
10:21 Joubu Given all the things I have in the pipe right now, I am happy with that local patch :D
10:39 cait ashimema: just rebased our branch to 22.11.03 - thx to the team :)
10:39 tcohen hola #koha o/
10:39 cait hola tcohen, que tal?
10:40 tcohen Bien! Wie geht es dir?
10:41 cait gut, danke :)
10:43 tcohen Joubu: Running [koha-shell kohadev -c "touch testing.success"]...
10:45 Joubu ?
10:45 Joubu Is there a question,
10:45 Joubu ?
10:46 tcohen that's the last line in Koha_Master_D10 before the container just dies before running the tests
10:46 tcohen does it ring any bell?
10:48 Joubu I am not seeing anything obvious
10:49 Joubu it is the very last thing it is doing
10:49 Joubu ha
10:49 Joubu no test are run
10:51 Joubu does it only happen on D10?!
10:51 tcohen sorry, when to get my first coffee of the day
10:52 tcohen there are a few tasks failing similarly
10:53 Joubu We should display the command
10:55 Joubu tcohen: ok got it
10:55 Joubu patch coming
10:58 Joubu tcohen: the last commit was wrong, see adjustement there[…]merge_requests/60
11:01 tcohen thank you
11:03 marcelr ok qa team get it under 100 now; we're down to 115
11:03 marcelr see you later
11:03 tcohen yes sir
11:03 tcohen marcelr++
11:08 PedroAmorim[m] oleonard: around?
11:08 tcohen PedroAmorim[m]: thanks for your message
11:09 PedroAmorim[m] o/
11:20 oleonard o/
11:20 tcohen \o
11:40 * PedroAmorim[m] uploaded an image: (7KiB) <[…]24%2010-39-28.png >
11:40 PedroAmorim[m] oleonard: bug 32991 - this is our current dialog in ERM, any comments?
11:40 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=32991 enhancement, P5 - low, ---, jonathan.druart+koha, Signed Off , Improve our Dialog component and remove routes for deletion
11:45 oleonard PedroAmorim[m]: My choice would be to make it more like this one:
11:46 oleonard I think the more specific button labels are helpful
11:46 oleonard (that is an opinion not backed up by evidence) :)
11:53 cait i agree
11:53 cait if the button has delete on it, it's better
11:53 cait also the access keys would be great
11:55 PedroAmorim[m] copy, thanks
12:01 * PedroAmorim[m] uploaded an image: (7KiB) <[…]24%2011-00-27.png >
12:01 PedroAmorim[m] icons: fa-check to accept and fa-remove to delete
12:01 PedroAmorim[m] what icon is more appropriate to just close a warning?
12:02 PedroAmorim[m] s/delete/cancel/
12:04 oleonard Outside of a dialog, it's fa-trash to delete, fa-remove to cancel. Inside a dialog I might use fa-check with "OK" if I'm just asking the user to confirm (a single button)
12:05 PedroAmorim[m] oleonard++
12:06 oleonard But if there is more than one option, e.g. "Continue ... Cancel" I would use fa-remove for the cancel
12:06 oleonard we have lots of inconsistency there and we need to improve.
12:10 cait i try to remember the aria thing we did tot he icons
12:10 cait does that apply here?
12:17 oleonard cait: Yes, 'aria-hidden="true"' should be added to any FA icon which "decorates" some associated text. If the icon is all by itself we don't add it.
12:21 alex_ joined #koha
12:23 cait oleonard: did you see the issue about navigation issues with tabs/arrows in OPAC?
12:23 cait I know i am jupming topics... accessibility related too
12:24 oleonard I saw it but I haven't looked into it
12:25 oleonard I'm fighting with KTD today... Not sure what's different but it doesn't want to run.
12:26 cait did you... update all the things?
12:26 oleonard I don't think I missed anything
12:27 oleonard Updated the ktd repo, did a fresh ktd pull
12:29 Joubu oleonard: seeing the saas error?
12:29 Joubu rm -rf node_modules, or yarn install on the host, then try again
12:33 PedroAmorim[m] cait: what is the issue about navigation issues with tabs/arrows in OPAC?
12:33 cait let me find the bug
12:34 PedroAmorim[m] we had an issue recently with a customer, it's fixed on that installation but I wonder if it's related
12:34 cait I am pretty sure it used to work, but you can't use tabs and errors for all the navigation now
12:34 cait bug 33012
12:34 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=33012 normal, P5 - low, ---, oleonard, NEW , Accessibility: Some navigation items in OPAC cannot be accessed by keyboard (search history, log out)
12:35 cait it seems to work all good for lists, but the  "acount" menu with search history and log out doesn't work well
12:35 tcohen cait: short RM <-> QAM call?
12:36 cait yess!
12:36 cait can i quickly run to grab icecream sfirst? just send the link, be right there
12:36 tcohen[…]QYUI4MFl5U3Fhdz09
12:37 tcohen anyone can join
12:39 oleonard You were right of course Joubu, removing node_modules fixed it
13:00 cait I'll try that now too
13:08 lukeg joined #koha
13:16 ashimema Sorry, was at lunch
13:21 cait I'llt ry that with the nodes now too
13:32 cait 15 bugs to below 100 in QA queue...
13:34 tcohen ashimema: our tests for x-koha-embed are... scarce
13:35 ashimema minimal I would say 😜
13:35 tcohen there's a test for bad x-koha-emebed attributes
13:35 ashimema there are some.. but I'm sure we could/should add more comprehensive cover
13:35 tcohen that expects a 400 and fails
13:35 ashimema does it fail?
13:35 ashimema I thought it was passing
13:35 tcohen meaning the 'extra' embeds are being processed
13:36 cait btw... there are some really really easy qa int here I can#t touch (signed off or written the patch)... just saying ;)
13:36 tcohen QA it cait
13:36 ashimema ah..
13:36 ashimema as in.. doesn't catch what it should
13:36 ashimema bugger
13:36 tcohen it does catch the bad params
13:36 ashimema I keep trying to work through that queue cait.. but you're adding so much stuff at the moment I can't keep up
13:36 tcohen I will add a positive test (embedding two things)
13:37 ashimema next cycle I reckon we need to add one of my devs to the QA team to help spread the load.. they'll need to be picky about what they QA at first though.. sticking to those sorts of ones as learning
13:37 tcohen why not adding them now?
13:37 tcohen we can amend the wiki :-P
13:37 cait ashimema: but easy and fun stuff!
13:38 cait I am all for it
13:38 cait we really need some reinforcmeents here
13:38 ashimema lol
13:38 cait and taking the easy ones is fine
13:38 ashimema well.. right now they're all pretty tied up already with erm 😜
13:38 cait or starting and commenting on the difficult ones
13:38 ashimema and I don't have the time personally to bring them up to speed on the QA process as a QAer
13:40 cait just know I am all for it :)
13:40 cait whenever you/they are ready
13:41 Joubu Add PedroAmorim[m] as Vue specialist
13:41 Joubu We got 3 signoffs but were stuck waiting for QA...
13:42 ashimema which set Joubu.. it's likely on my list
13:42 ashimema erm refactoring
13:44 tcohen kidclamp is QAing
13:45 ashimema ah, awesome
13:45 ashimema that takes a set off my plate
13:45 ashimema thanks
13:45 pastebot "tcohen" at pasted "ashimema,Joubu: this passes" (36 lines) at
13:45 tcohen so I don't understand why it isn't working for ERM routes
13:48 Joubu what isn't working?
13:48 tcohen on bug 32713 you are completely removing the request validation from ERM routes
13:48 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=32713 enhancement, P5 - low, ---, koha-bugs, NEW , x-koha-embed appears to no longer properly validate
13:49 tcohen sorry
13:49 tcohen bug 32923
13:49 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=32923 normal, P5 - low, ---, jonathan.druart+koha, Passed QA , x-koha-embed must a header of collectionFormat csv
13:49 tcohen and I'm trying to figure why
13:50 Joubu you just said that it was safe to backport but you don't understand the change? :D
13:50 tcohen I'd backport without that last patch
13:50 Joubu the first part of the set is simply to fix the divergence with the rest of the spec
13:50 tcohen the rest of it is correct: i.e. moving the things we missed to the right place in the spec
13:51 Joubu it moved x-koha-embed to the enum
13:51 tcohen not having done it has some consequences
13:51 Joubu then we are fixing a bug because of the change
13:51 tcohen so it is a bug fix and needs a backport
13:51 tcohen yeah, but why the ERM routes fail like that
13:51 tcohen and others don't
13:52 tcohen that's what I pasted
13:52 Joubu wait, I can explain from scratch if you like
13:52 tcohen sure
13:52 Joubu but I think there is everything in the comments
13:52 Joubu there are 2 validation calls, and that break the header
13:52 Joubu We (Pedro and I) think it's a bug upstream, but didn't find evidences of that
13:52 tcohen it used to be a bug upstream
13:53 tcohen and we upgraded the tools when it got fixed
13:53 Joubu if you look at Koha/REST/V1/ERM/EHoldings/
13:53 tcohen that's why I'm intrigued
13:53 Joubu you see: 37     my $c = shift->openapi->valid_input or return;
13:53 Joubu then we call Koha::REST::V1::ERM::EHolding​s::Packages::EBSCO::list($c);
13:53 Joubu which will call valid_input as well
13:53 Joubu PRIOR to the change (move to enum), it was working, ie. the headers were not affected
13:54 Joubu AFTER the change, the headers are truncated (from 'foo,bar,etc' to 'foo')
13:54 tcohen ok
13:54 Joubu This is the bug, that does not exist in stable, it's caused by the previous patches
13:55 tcohen it is very specific to this implementation
13:55 tcohen and not the bug I was mis-understanding there was
13:55 tcohen I wonder if that fully qualified call is the right way of doing it, though
13:56 marie-luce joined #koha
13:58 tcohen stash_embed() is called when authenticate_api_request is called i.e. at the very beggining and before the spec is validated
13:58 tcohen so koha.embed is already set
13:58 tcohen and you *shouldn't* need to query for the headers again
13:59 Dyrcona joined #koha
13:59 tcohen Koha/REST/V1/
13:59 Joubu ok, but that does not explain why it gets truncated
14:00 Joubu "I guess I could use $c->stash('koha.embed') instead, but then it's no longer a string an I need to change the logic in several places."
14:00 Joubu comment 7
14:00 Joubu I didn't want to rewrite something else
14:07 tcohen I would push it as-is, but we should file a bug for upstream and have a bug ourselves to track this
14:09 cait do I need toknow something about bug 32728?
14:09 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=32728 normal, P5 - low, ---, jonathan.druart+koha, Signed Off , ERM - Search header should change to match the section you are in
14:09 cait i have applied the patch, restart_all... but not sure I see any change
14:10 lukeg joined #koha
14:10 cait my first "erm" patch
14:10 cait do i need to run something if a .ts is changed?
14:12 Joubu changing a ts??
14:12 Joubu what's your patch? :)
14:12 Joubu no, you don't anything special, just submit
14:12 cait ... in QA .)
14:12 cait i am testing 32728
14:13 Joubu yarn js:build
14:13 tcohen or yarn build
14:16 tcohen writing a regression test for the openapi plugin
14:16 tcohen wish me luck :-P
14:16 cait Joubu: ok, I'll try :)
14:18 cait Joubu: I notice we don't have a search for licenses
14:18 cait shoudl I file it?
14:18 cait or maybe there is already a bug?
14:18 cait otherwise this works well now
14:20 Joubu if you think we need it you can open a bug
14:20 tcohen wish me luck :-P
14:20 tcohen oops
14:26 pastebot "tcohen" at pasted "use Mojo::Base -strict; use Te" (117 lines) at
14:38 cait I think it woudl make sense and be sonsistent,... so I'll open one :)
15:07 cait brb
15:30 caroline Did my kohacon24 email go through? I sent it 45 minutes ago, but haven't received it yet...
15:33 Joubu look at the archive -[…]ruary/thread.html
15:33 Joubu seems not
15:34 Joubu make sure you are using the email that is registered
15:38 caroline I used the same email as usual... :/ I'll try again, hopefully not spamming anyone
16:00 bag joined #koha
16:06 pastebot "oleonard" at pasted "Should we update our title tags for translatability?" (19 lines) at
16:08 caroline oleonard, I think it would be helpful, yes
16:12 oleonard Another project :D
16:26 cait 106 - still some easy ones waiting!
16:33 cait caroline: i see it, your email
16:33 caroline cait thanks! you received it just once?
16:33 cait yes
16:33 cait so far :)
16:33 caroline good :)
16:36 oleonard When an AI chatbot can write my test plans for me, then I'll be interested
16:39 cait heh
16:43 cait someone qa for bug 32164? It's super fun (and I can't)
16:43 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=32164 enhancement, P5 - low, ---, oleonard, Signed Off , Add link to MARC modification templates from batch record modification page
16:43 fred joined #koha
16:46 cait hi fred
16:48 cait 99 :)
16:49 cait kidclamp++
17:00 cait oops
17:00 cait it looks like downlding marc from the result list in opac is broken:
17:00 cait Not a CODE reference at /kohadevbox/koha/opac/ line 93
17:00 cait hm list.
17:02 avengingchicken joined #koha
17:12 cait left #koha
17:24 lukeg is Joubu about?
17:27 lukeg @later tell Joubu if you could look at my patch and comments for Bug 32505 I would greatly appreciate it. I backported this to 22.05.x without realizing it relied on day.js
17:27 huginn lukeg: The operation succeeded.
17:32 emlam joined #koha
17:45 cait joined #koha
18:55 davidnind QA Team++
19:38 tcohen what have you all done to my queue!
19:38 tcohen haha
19:44 cait lol
21:32 cait oleonard-away: should we not have a tooltip on our 'home' icon in the breadcrumbs?
21:33 cait if i hover.. nothing happens and not sure what screenreaders would make of it
21:33 cait ok, I see it's actually a link - but maybe a tooltip/title would still be nice
21:45 cait tcohen: are you on it alredy? i could provide a follow-up...
21:50 cait follow-up uploaded on bug 31563
21:50 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=31563 normal, P5 - low, ---, katrin.fischer, Pushed to master , Numbers on claims tab not showing in translated templates

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary