IRC log for #koha, 2021-10-22

← Previous day | Today | Next day → | Search | Index

All times shown according to UTC.

Time Nick Message
00:11 koha-jenkins Project Koha_Master_D11_My8 build #670: SUCCESS in 39 min: https://jenkins.koha-community[…]ster_D11_My8/670/
00:24 Project Koha_Master_D11_MDB_Latest build #706: SUCCESS in 53 min: https://jenkins.koha-community[…]1_MDB_Latest/706/
02:21 Project Koha_20.05_U21 build #1: FAILURE in 4 min 18 sec: https://jenkins.koha-community[…]Koha_20.05_U21/1/
02:24 Project Koha_20.05_U21 build #2: STILL FAILING in 2 min 55 sec: https://jenkins.koha-community[…]Koha_20.05_U21/2/
02:28 kidclamp_ joined #koha
02:48 koha-jenkins Project Koha_20.05_U21 build #4: STILL FAILING in 23 min: https://jenkins.koha-community[…]Koha_20.05_U21/4/
03:01 Project Koha_20.05_U_Stable build #13: SUCCESS in 44 min: https://jenkins.koha-community[…]0.05_U_Stable/13/
03:08 jo_ joined #koha
03:12 koha-jenkins Yippee, build fixed!
03:12 wahanui Congratulations!
03:12 koha-jenkins Project Koha_20.05_U21 build #3: FIXED in 48 min: https://jenkins.koha-community[…]Koha_20.05_U21/3/
03:48 marie-luce joined #koha
04:30 lmstrand joined #koha
06:35 reiveune joined #koha
06:35 reiveune hello
06:35 wahanui hola, reiveune
06:40 cait joined #koha
06:41 cait1 joined #koha
06:47 alex_a joined #koha
06:47 alex_a Bonjour
06:53 marcelr joined #koha
06:53 marcelr hi #koha
06:54 is bugzilla so slow today or just my laptop ?
06:56 cait1 appears to be sloe
06:56 slow
06:56 Joubu slow here as well
06:57 marcelr yeah terrible
06:57 magnuse \o/
06:58 marcelr hi magnuse
06:59 magnuse hiya marcelr
07:00 lds joined #koha
07:09 cait1 hm Friday... yeah, there is a possibillity
07:09 but not sure which level of lockdown they are at the moment
07:09 Joubu ashimema: can you add 29241 to you QA stack for today?
07:10 cait1 bug 29241
07:11 marcelr kick that server
07:11 Joubu wahanui is on the same server :)
07:11 marcelr hm no bugzilla day for me with this speed
07:12 lack of speed
07:15 Joubu marcelr: pick bug 14957, no need to access bugzilla for a couple of hours once it's applied
07:16 erk getting gateway timeout now :-/
07:18 marcelr couldnt apply stuff at all
07:18 i have recalls in a local branch erk
07:18 Joubu marcelr: saw bug 11175 comments 380 and 381?
07:19 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=11175 enhancement, P5 - low, ---, joonas.kylmala, Failed QA , Show the parent record's component parts in the detailed views
07:19 marcelr not sure
07:21 Joubu There was "4. Shouldn't the limit (300) be configurable?"
07:22 marcelr o yes i noted that too
07:22 it should
07:22 cait1 could this not be a follow-up?
07:22 Joubu the empty line removal in the squashed version in coming from add/remove of line of the same file
07:22 cait1 it's already pretty big
07:22 Joubu follow-up bugs are never filled
07:22 unless the main is not pushed..
07:22 cait1 i am nt sure, but does the feature iself have an on and off switch?
07:22 marcelr i think they need to be squashed but i am alone on that
07:22 Joubu hardcoding limit is just bad
07:22 marcelr right
07:23 actually the bug is not big
07:23 we should really remove all those cruft
07:24 cait1 i think 300 feels sensible for a start - but of course I am not against an option
07:24 just wondering if this could hold this one up even longer
07:25 marcelr cait it would be very easy to add it
07:25 so it is a blocker
07:25 5 mins work
07:26 if the author wants something in, he does it right away
07:27 cait1 this one has been adopted by ashimema... just feel bad because he already put a lot of hours in
07:27 (and I was not super helpful) *hides*
07:27 ashimema morning
07:27 marcelr hi ashimema
07:27 cait1 someoen using OAI sets?
07:28 i could use some help with a little debug... we are mapping on a repeatable field (something like a library code in the record) and it only appears to look at the first field... i need to fix
07:29 ashimema OK.. I give up
07:30 alexbuckley joined #koha
07:30 marcelr on what ashimema ?
07:30 ashimema he number of hours I put in rescuing peoples bugs..  I can't keep throwing time into it.. I've already spent whole weekends worth of personal time of this bug
07:30 marcelr you are so close
07:31 ashimema it's been "so close" for 3 cycles!
07:31 marcelr need to hang on just the last bit
07:31 i think this one could make it still
07:31 its small stuff
07:31 ashimema I am generally one of those who DOES do followup bugs
07:31 marcelr me too
07:31 ashimema I can't keep working on one bug forever with no chance of it ever getting in
07:31 yes
07:31 lots of small stuff
07:31 marcelr it goes in if you add 15 mins more
07:32 ashimema another 10 minutes to add a syspref
07:32 marcelr 5
07:32 :)
07:32 ashimema then another to for this
07:32 and another for that
07:32 it all adds up
07:32 I'm tired
07:32 I need to get to paying work
07:32 Joubu I will write the patch
07:32 marcelr but you are right, where is the author ?
07:32 we need some commitment from authors
07:33 ashimema they dissapeared years ago.. because it took us years to look at it.
07:33 on average it takes 2 years to get code into Koha
07:33 marcelr too long
07:33 ashimema I don't think we can justifiably expect people to stick around that long
07:33 especially if they're not seeing things moving
07:33 marcelr i give up after 6 months
07:33 ashimema yup
07:33 marcelr could have written much much more
07:34 ashimema lol
07:34 2013 this was originally authored!
07:34 8 years!
07:34 marcelr but in this case i would like to push you a bit
07:34 so close !
07:34 ashimema but thankyou for QAing
07:35 I don't mean to be grumpy
07:35 more coffee is clearly needed this morning
07:35 I'm being a git
07:35 Joubu It's Friday, it's allowed
07:35 marcelr np its the usual frustration
07:35 we all know it
07:35 Joubu (it's allowed on Mondays as well btw)
07:35 marcelr but not on Wednesday
07:36 * ashimema thinks with an LTS version now added to the mix we should really reduce the support period for normal maint releases
07:36 ashimema but that's another topic
07:36 school run.. then I'll take a look at the syspref
07:36 marcelr right you are on track again
07:36 ashimema it's going to need dealing with in 2 templates and 2 controllers
07:36 marcelr hope bugzilla works along
07:36 it is super slow\
07:37 Joubu BTW nugged is maintaining/rebasing bug 11175
07:37 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=11175 enhancement, P5 - low, ---, joonas.kylmala, Failed QA , Show the parent record's component parts in the detailed views
07:37 Joubu I am sure he won't mind adding follow-up
07:37 marcelr ashimema++
07:37 joonas is someone else?
07:37 Joubu but he was in a rush last 2 weeks (using master in production)
07:37 and he is taking some days off
07:38 marcelr master in production heh
07:38 cait1 Joonas switched jobs
07:38 so he is around still but possibly not for that one
07:38 marcelr he was qaing recently
07:38 cait1 yes
07:38 he got funding
07:38 but it's not the same role he had before
07:38 marcelr i did master in production about 12 years ago, but not longer than a few months..
07:39 cait1 I filed 29305
07:39 bug 29305
07:39 marcelr should we change the assignee?
07:39 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=29305 normal, P5 - low, ---, koha-bugs, NEW , OAI sets don't work correctly for repeatable tags
07:39 cait1 I've also been nagging QA team to look at the old bugs
07:39 I completely agree that the first feedback on a bug needs to come earlier
07:40 and subsequent feedback, but having people wait while other stuff moves through is going to make paopel unhappy for reasons
07:40 marcelr the recalls stuff for instance
07:41 ashimema It's hard
07:41 cait1 the recalls is another issue
07:42 the recalls is becuase if people seeing issues with specification
07:42 ashimema Sand post of the problem is you end up grunping to the people that are well aware of the problems and already busting a gut to try and solve it
07:42 cait1 different expectations on a recall feature
07:42 ashimema The challenge I had with recalls is that all feedback fell of death ears
07:42 It was never acted upon
07:42 cait1 and I am not sure how we can resolve such situations yet
07:42 marcelr for such large things you need consensus at the start
07:42 ashimema For some bugs I feel a line should be drawn and a fresh bug started..
07:43 Perhaps 11175 should have taken that route if I'm honest
07:43 marcelr you cannot throw code over the fence
07:43 cait1 I unerstand both sides
07:43 ashimema The final code doesn't bate much resemblance to the original
07:43 cait1 we are stuck between: this is what the customer is happy with
07:43 marcelr in case of recalls many will be happy with it
07:43 cait1 and other expectations
07:44 marcelr except ashimema because it will be harder for booking
07:44 s
07:44 :)
07:44 cait1 it's not only an issue for this bug
07:44 but this bug highlights it
07:44 ashimema I asked for two things over the years.. I felt from day one it was too much of a copy/paste affair.. it should have either been added to holds or written with completely fresh modern code
07:44 cait1 and we need to resolve - marcelr is right about that
07:45 ashimema And the other was I suggested a rename as recalls is very different for a large chunk of the world
07:45 marcelr how would you call it?
07:45 ashimema I wasn't against the feature entirely.. but felt it was incredibly confusing having it called recalls
07:46 Can't remember
07:46 I did comment some suggestions I think
07:46 marcelr one of the 700+
07:46 cait1 wiki is super slow too... seems like such a day
07:46 ashimema Honestly.. I don't remember the bug well
07:46 It left a bitter taste
07:47 marcelr i think it has improved over time
07:47 ashimema I could write a recalls from scratch.. that was always an intention
07:47 marcelr but the base surely is the same
07:47 cait1 Some homework for us:
07:47 ashimema But time
07:47 cait1 https://wiki.koha-community.or[…]talyst_IT_Recalls
07:47 https://wiki.koha-community.org/wiki/Recall_RFC
07:48 Joubu "you cannot throw code over the fence", really? who never did that? :D
07:49 marcelr Joubu: talking about large patches
07:49 not three lines
07:49 Joubu You forced me several times
07:49 marcelr never
07:49 Joubu it's a "you" plural :)
07:49 marcelr i cant recall haha
07:50 Joubu droping code is better sometimes
07:50 ashimema I have.. I know I'm guilty.. though I do think I try to respond to feedback more than most
07:50 marcelr right
07:50 Joubu hard to admit, but you gain time in the long term
07:50 marcelr less maintenance
07:50 i thought about removing C4
07:50 ashimema Lol
07:50 Joubu The hardest is when you rewrite stuffs in 3 different ways and still don't reach maste
07:51 marcelr at that point you must stop
07:51 Joubu anyway, syspref max stuffs is attached to the bug
07:51 next?
07:51 wahanui next is Koha bibliography in Zotero
07:51 marcelr driemaal is scheepsrecht
07:51 Joubu we can focus and push it in 2h max if we want ot
07:51 to
07:51 marcelr now youre talking Joubu
07:51 we got him in the right mode ashimema
07:52 Joubu It's on my list for 21.11
07:52 cait1 :)
07:53 ashimema Needs template hook too.. I'll look when I get back
07:54 Can't code from the phone.. haha
07:54 Joubu spoiler: 11175 and 14957 will be part of 21.11
07:54 cait1 bug 14957
07:54 wahanui hmmm... bug 14957 is rebased against master I see
07:54 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=14957 new feature, P5 - low, ---, glasklas, Signed Off , Write protecting MARC fields based on source of import
07:54 Joubu I will write the missing bit and push them even if not PQA
07:54 cait1 oooh
07:54 * ashimema almost cries at that prospect
07:54 Joubu so better if somebody is working along with me :)
07:55 ashimema Sorry Joubu, I've not managed to get back to the wire protected fields stuff.. I saw a couple of mails come through but haven't had a moment :(
07:55 Today's the day
07:55 marcelr Joubu just ping me when ready
07:56 i resume my qa
07:56 you know it will probably fqa again lol
07:57 Joubu ashimema: I found bug, author wrote a bugfix, I added a follow-up for tests, it's almost ready for push I think
07:57 except the lack of feedback
07:57 (see the last patch I obsoleted)
07:57 ashimema 🙂
07:59 Joubu marcelr: git grep ComponentParts
07:59 there is a mismatch
07:59 Components ([% ( ComponentParts.size ) %]) vs Components ([% ( ComponentParts.size || 1 ) %])
07:59 and [% IF ( ComponentParts ) %] vs [% IF ComponentParts.size %]
07:59 that does not feel correct
08:00 The discussion about UseControlNumber is still not clear to me
08:00 ashimema Damn, I've fixed that once already
08:00 Joubu ho wait
08:00 I am working on my old branch
08:00 you added some follow-ups
08:00 sorry!
08:01 ashimema Bet a lost a patch during rebase somewhere
08:01 UseControlNumber discussion?
08:01 Joubu ashimema: is the pb with UseControlNumber fixed as well?
08:02 ashimema Will reread the bug when I get back.. not sure what that one was
08:02 Joubu ashimema: I had to turn UseControlNumber on to make the query return result
08:02 ashimema Oh.. that it should work without
08:02 Joubu comment 384
08:02 > Caused by UseControlNumber=Don't use.
08:02 > It works when switched to "Use" (Thanks Martin!)
08:02 This, I think, highlights some flaws in our Host-item index.. I would have expected the search to work.. but I think that can be saved for another bug.
08:02 ok, will try again then
08:03 ashimema I 'think' it is.. but tis still a pretty specific search
08:03 Joubu it's standard ktd
08:03 ashimema Yeah.. if it doesn't I think a separate bug.. I agree, I think I improved the chances of hits ,but the index still needs some work
08:05 Joubu I confirm it does not work
08:05 Should we add a note on the syspref? only work if UseControlNumber is on?
08:05 marcelr: didn't you notice that?
08:08 ashimema Well.. it does work in sale cases if not
08:08 Some
08:08 Those test records are designed for control number links
08:09 So it's a somewhat bad test
08:09 marcelr Joubu i only start testing if the code looks good enough
08:10 ashimema The non control number looking is always going to be a little bit hit and miss as it's doing title string searches
08:14 marcelr surname      => 'my surname ' . $_,
08:14 Holds.t weird
08:15 Joubu why?
08:15 marcelr it was pasted from a loop to another location
08:15 o o o
08:15 will correct it somehwere
08:20 ashimema y
08:21 Joubu ashimema: What about: "2. ES index added to the default mappings but to the upgrade process. Is that expected?"
08:21 ashimema just got back from school run
08:21 catching up.. give me a mo 😉
08:22 Joubu marcelr, ashimema: please don't attach anything to the bug
08:22 ashimema oh.. you went back and fixed your first syspref submission  already 🙂
08:22 either that or I misread it on the mobile screen
08:23 Joubu and fixed the second submission as well :D
08:23 ashimema brill
08:23 Joubu I am going to clean the patchset a bit, indentation, removal of empty lines, etc.
08:23 so some will be squashed when it help readability
08:23 ashimema 🙂
08:24 Joubu Let me know about the ES index
08:24 alex_a joined #koha
08:24 ashimema ah, I missed that comment.. sorry
08:25 Nick did the ES stuff waaaay back
08:25 likely that's a mistake.. it should probably be in the upgrade process too
08:25 having said that.. I have no idea how you do that..
08:25 would like to learn if you can guide me?
08:26 marcelr Joubu: its your party now, let me know if i should have another look
08:28 ashimema my brain hurts
08:28 remind me how I can see what gets put in the Host-item index?
08:29 the .att, .xml, .xsl, ccl.properties stuff totally confuses me now
08:30 cait1 start witht he xml
08:30 it's all you need when you have the right index name (for Zebra at least)
08:30 didn't we say it was 773 a and t the other day
08:31 ashimema ```
08:31 995   <!--record.abs line 229: melm 773$a      Host-item-->... (full message at https://matrix.org/_matrix/med[…]aBiLNTSXJSRdagEg)
08:31 and
08:31 1003   <!--record.abs line 231: melm 773$t      Host-item-->
08:31 1004   <index_subfields tag="773" subfields="t">                                                                                                                                                                                                                                                                             1005     <target_index>Host-item:w</target_index>
08:31 1006   </index_subfields>
08:31 so.. a and t
08:31 but not combined
08:31 am I reading that right cait1?
08:32 i.e you should be able to search on a OR t but not "a t" or whatever combination
08:34 so Joubu.. assuming the cataloguers have put the 245$a of the Host record into the 773$a of the child it should get found properly
08:34 I tihnk
08:34 having said that..
08:35 we appear to also drop `/` punctuation if it exists.. (that came from the original searches.. so I'm not sure if they've ever worked if they're not working now either.
08:39 cataloguing is a pain
08:40 how 245 subfields should match up with 773 subfields outside of the $w is really hard to know..
08:40 * ashimema customers always us UseControlNumber for that reason
08:45 alex_a joined #koha
08:47 petrova joined #koha
08:59 ashimema right... meetings done.. I'll take a look at the follow-ups on the write protected marc fields one now Joubu
09:09 Joubu marcelr: 11175 back to you
09:13 marcelr wow
09:14 ok haveing a look dont touhc it now !
09:14 hmm typo typo
09:17 magnuse why would an item with zero renewals allowed be renewed from the opac three times in one minute?
09:23 ashimema ok.. the caching has got lost somewhere ☹️
09:24 * ashimema comments on the bug
09:24 nugged > 09:37 <Joubu> BTW nugged is maintaining/rebasing bug
09:24 Joubu: what was that about?
09:24 Joubu 11175
09:24 nugged yes,
09:24 Joubu ashimema: caching?
09:24 nugged sorry for partial question :)
09:24 Joubu nugged: no worries, we are on it, almost pushed
09:25 ashimema I'll reply on the bug
09:25 nugged ok
09:25 yes, I have some messy "check-out from hotel with two ladies" day.
09:34 * magnuse raises an wywbrow
09:34 magnuse s/wywbrow/eyebrow/
09:39 marcelr ashimema: you are not changing 11175 now i hope
09:39 ashimema nope
09:39 just commenting
09:39 marcelr good for you
09:39 ashimema frankly.. if I wrote this now it would be completely different.. I'd have used API's
09:39 but as it predates our API's by many years I didn't
09:43 Joubu still possible :D
09:44 marcelr did you test it already Joubu?
09:44 Joubu yes
09:44 marcelr ok
09:44 i do a 5min test
09:44 then
09:44 Joubu turn UseControlNumber on (I don't understand that and don't want to)
09:45 marcelr so far so good just 8 followups
09:45 haha
09:51 * ashimema understands the UseControlNumber off issues..
09:51 ashimema they predate this bug
09:51 so I suggest fixing them in another one
09:51 and asking someone who actually uses that configuration for guidance on exactly what should be searched 😉
09:52 marcelr CCL parsing error (10014) Unknown qualifier ZOOM for query: (((rcn=20586738 AND cni=OSt) OR rcn="OSt 20586738") AND (bib-level=a OR bib-level=b)) at /usr/share/koha/C4/Search.pm line 242.
09:52 nugged magnuse: very expensive ladies – those whom you need to care and support daily ;))))
10:00 marcelr Joubu my 5m test fails on zebra stuff. Could be my test setup now.
10:00 Need to cleanup
10:01 ashimema do you have the cni index marcelr?
10:02 i.e. bug 28830
10:02 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=28830 enhancement, P5 - low, ---, martin.renvoize, RESOLVED FIXED, Add CNI (Control Number Identifier) search index (MARC21)
10:02 ashimema that failure is what I'd expect if you haven't got that bug applied and reindexed since
10:03 Joubu
10:03 did you have a guide on how to do that 'update to add new index' you requested?
10:03 that cni bug probably needs the same treatment looking at it.
10:04 Joubu I was talking about the ES index
10:04 admin/searchengine/elasticsearch/mappings.yaml | 9 +++++++++:
10:05 ashimema I know
10:05 Joubu because of this change
10:05 marcelr yeah i added it and reindexed but it is just a config thing
10:05 code looks good
10:05 ashimema yeah.. the same file is changed in the CNI bug
10:05 but you said we need to do something to update as well as just changing the new install default
10:05 didn't you?
10:06 marcelr bz on full speed again
10:06 ;)
10:06 ashimema man I wish BZ wouldn't add a comment for every patch attached
10:07 Joubu ashimema: yes, we need to add to the DB what we add to the yaml
10:07 I think
10:07 ashimema ah, I see
10:07 Joubu it's search_marc_map and the 2 other tables
10:07 marcelr ashimema: squashing has advantages too
10:08 i'll test this again in pushed master next week :)
10:08 Joubu lol
10:09 that's not how it works!
10:11 marcelr use your RM power
10:11 lmstrand joined #koha
10:12 Joubu I don' understand the code in C4::XSLT
10:12 and I am trying to write test for it.. :-/
10:12 In which cases do we display the "show analytics" link?
10:12 ashimema I'm re-reading it now
10:12 I'm not sure it's correct
10:12 that I can tell you
10:13 We show it if
10:13 Joubu what's the expected behaviour?
10:13 just tell me that
10:13 and I will adjust
10:13 ashimema We're on the details display AND ShowComponentParts is DISABLED and there IS component parts
10:14 ShowComponentParts replaces the link
10:14 Joubu ok then
10:14 ashimema 'Only show the link if it will yield results and we're not already showing them inline'
10:15 that's possibly clearer?
10:16 pretty sure that's not what marcelrs follow-up does
10:16 Joubu 'If pref is on and record do not have component, show analytics link is not displayed'
10:16 'If pref is on and record has component, show analytics link is not displayed'
10:16 'If pref is off and record has component, show analytics link is displayed'
10:16 that's not passing :)
10:16 ashimema that's the right set of tests 🙂
10:16 was it passing prior to marcelr's final follow-up>
10:16 Joubu if pref is on, we should retrieve the components then
10:17 as the link will never be displayed
10:17 ashimema if the pref is on.. we don't need to
10:17 they'll be retrieved inside the controller instead
10:17 if the pref is off we need to retreive them so we know whether to show the link or not
10:17 silly thing is..
10:18 we retrieve them regardless
10:18 so performance wise there is no difference between on and off
10:18 which was what the point of the preference was.. people thought enabling inline display might be bad for performance
10:18 but you're already taking the performance hit by doing a search to see if you need to display the link
10:18 * ashimema tried to point that out a few times
10:19 Joubu who wrote that code? They were drunk
10:19 khall_ joined #koha
10:19 ashimema which bit?
10:19 the 'show_analytics_link' stuff?
10:19 Tomas 😉
10:19 I moved it out of controller level into the xslt level though
10:20 to DRY it out
10:20 but I also had the inline component display stuff in there at the time too
10:20 history..
10:20 wahanui it has been said that history is only deleted if the purge archive maintenance script is run
10:20 ashimema it got moved back out as inline display caused other issues
10:21 this is why I didn't squash.. there is a huge amount of history 😉
10:21 though I should have squashed some.. which you have now
10:21 it was which bits to squash that killed me
10:24 cait joined #koha
10:29 Nikos_ joined #koha
10:29 ashimema Joubu.. perhaps for another bug.. but are you happy with the tests having mocked the search engine
10:29 Nikos_ Hello may I ask something ?
10:30 ashimema I kinda feel like we should perhaps rely on some of the core test data and actually hit zebra/elastic to check for results
10:30 Joubu erk
10:30 marcelr
10:30 ashimema Nikos_ always feel free to ask.. you don't need to ask to ask
10:30 Joubu didn't see your follow-up
10:31 ashimema: mocking it is correct to me
10:34 ashimema fair enough
10:34 khall joined #koha
10:34 marcelr Joubu you dont see it when you remove it :)
10:35 Joubu sorry, patch not attached
10:35 attaching now
10:36 marcelr, ashimema: https://bugs.koha-community.or[…]ent.cgi?id=126744
10:36 are those tests correct?
10:36 * ashimema is already reading them
10:36 Nikos_ Well, I use two ports for our koha (8080) and (8081). three years ago, I purchased an ssl certifiacte and everything was fine for both the ports. This year, somehow, I miss something and I can't add the new certification files in my conf file.
10:36 Joubu it's not what the original code was doing, and not how marcelr rewrote it
10:37 marcelr its a confusing test
10:38 still looks like it is not correct
10:38 you will fetch components unneeed
10:39 ashimema my brain is still processing
10:39 marcelr hmm i cant
10:39 need coffee
10:40 seeing you later
10:40 kellym joined #koha
10:41 Joubu it's only fetching the components if needed
10:41 ashimema `like( $xml, qr{Show analytics}, 'If pref is off and record do not have component, show analytics link is displayed' );`
10:41 Joubu I don't see an extra fetch
10:41 ashimema not sure that one is right
10:41 Joubu it's the only case where it's displayed in my tests
10:41 ashimema it shouldn't be
10:41 Joubu if we remove it there it's never displayed
10:41 ashimema if there are no linked records, the link shouldn't appear
10:42 I can take over if you like?
10:42 I understand what the final requirement is from the history
10:42 Joubu ha sorry, misread the original code
10:42 -        $variables->{show_analytics_link} = ( scalar @{$components} == 0 ) ? 0 : 1;
10:42 I read it without the == 0
10:43 ashimema 🙂
10:43 it's a mind bender
10:44 Joubu and one description was wrong
10:44 ashimema oh?
10:44 wahanui oh are there instructions?
10:44 Joubu revisited patch attached
10:44 tcohen hola #koha
10:44 ashimema to me it just looked like the like/unlike needed reversing for line 246 and 270 in the test
10:44 cool
10:45 will look at the updated one
10:47 tests look solid to me now
10:49 looks good to me Joubu 🙂
10:50 I might have nested the conditions slightly differently
10:50 but they work this was
10:50 way
10:51 * ashimema would have left the 'Details' test as an outer 'if'.. then looked up the pref and done an inner 'if'... finally if that passed I'd fire the search and use a ternary.
10:52 ashimema you're is almost the same.. but you merge the conditions into one 'if' and thus to the syspref fetch early even if it's not subsequently required.
10:52 Joubu 2 ifs without else means you can merge into 1 if :)
10:52 ashimema yes..
10:52 you can
10:52 alexbuckley joined #koha
10:53 ashimema but it does mean doing the syspref fetch early 😉
10:53 Joubu yes, but it's cached :D
10:53 ashimema not that that is an especially big hit
10:53 Joubu it's a L2 cache access
10:53 ashimema but if we're golfing 😉
10:53 Joubu but yes, you are right
10:53 ashimema I'm not worried really
10:53 Joubu L1 even
10:53 ashimema just commenting 😉
10:53 Joubu so, pushing? :D
10:53 ashimema go for it
10:54 I'm happy to iterate further on follow-up bugs if people want them..
10:54 * ashimema already has thoughts to do similar for some other area's.. like volumes
10:54 Joubu are you on 14957 at the same time?
10:54 ashimema yup
10:54 been reading through
10:55 was about to add the dependant bug about notifying the end user with your obsoleted patch
10:55 as a starting point
10:56 we should certainly iterate on that side early next cycle I feel
10:56 but otherwise, I feel it looks solid
10:56 I can run a few tests on it after lunch
10:56 talking lunch.. bbiab
10:58 Joubu same, enjoy!
11:10 khall_ joined #koha
12:06 cait1 kidclamp: thx for the follow-up!
12:07 kidclamp you are very welcome
12:08 Joubu kidclamp, tcohen, ashimema: please put 28700 on the top of your list
12:22 ashimema sure thing
12:22 question.. how long are we suggesting this LTS should be supported for..
12:24 jeremb joined #koha
12:29 Joubu at yesterday's dev meeting we decided to discuss it on the ML first, then come at one of the next dev meetings with the different options
12:30 ashimema fair enough
12:33 Joubu are we loosing bugzilla again?
12:33 ashimema feels like it
12:34 my attaches are very slow
12:34 cait1 bug 28700
12:34 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=28700 major, P5 - low, ---, stalkernoid, In Discussion , "Unblessed" method in Object.pm has blessed values for keys in some cases
12:34 cait1 oh, none for me then
12:35 fi you got a minute later, maybe you could clarify status on bug 28377 ashimema? I think it set a while without the last comment being seen, tryingto push/move
12:35 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=28377 enhancement, P5 - low, ---, tomascohen, Signed Off , Use the API to suspend/resume holds
12:35 ashimema it deals with dates
12:35 cait1 yes... i got that far :)
12:36 ashimema I'm basically not willing to PQA on any API stuff that does dates until we work through the DateUtils offset stuff
12:36 cait1 I was just not sure if it shoudl be FQA, BLOCKED by ? or similar
12:36 is there a bug for the latter?
12:36 ashimema Well.. Tomas has looked at the DateUtils patches and is happy they work and understands the tests
12:36 as have I
12:37 and I don't really have anything further to add honestly
12:37 yes.. there has been for years
12:37 cait1 which are these patches?
12:37 just trying to build up the dependency here
12:37 ashimema but it's another one that no-one has really had the time to properly get to grips with QA wise
12:37 instead of trusting the tests they asked that I re-write them all.. again
12:37 which I don't trust myself to be able to do.. again
12:38 cait1 ashimema: it's all good, i jsut want a bug number :)
12:38 ashimema bug 24850
12:38 cait1 so we can see about how to resolve things
12:38 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=24850 major, P5 - low, ---, martin.renvoize, ASSIGNED , Koha::DateUtils ignores offsets in RFC3339 datetimes
12:38 cait1 thank you
12:39 it's currently assigned
12:39 with patches
12:42 ashimema I assigned it because it was getting such negative feedback.. but I can't spend any more time on it
12:42 * ashimema is tired and grumpy again
12:42 cait1 you are moody :)
12:42 * cait1 sends cookies
12:43 ashimema I'm just fed up of really nasty bugs blocking things for years
12:43 I just want to be able to move on
12:43 cait1 yes, I understand
12:43 Joubu "I'm going to take a look at refactoring the tests", we have asked you to :)
12:43 "I'm going to take a look at refactoring the tests", we haven't asked you to :)
12:43 cait1 i was wondering if Joonas or Marcel might be able to have a look
12:43 Joubu negation missing!
12:43 cait1 but i think they are not around today
12:43 ashimema dcook and marcelr both did I believe
12:44 marcelr already passed it once
12:44 cait1 ah yes, see Marcel, but not Joonas on it
12:44 Joubu I had a chat with tcohen, he told he would test some stuffs and come up with something
12:44 cait1 I am moving to FQA to highlight it a little more as needs work
12:46 ashimema the biggest problem is..
12:46 if it gets pushed now and has problems I won't be able to fix any of them because it's spent so much time languishing I have no idea what I was thinking when I wrote it
12:46 I was very confident in it
12:46 it was very well tested
12:47 Joubu it's not about being well tested, it's about how we handle dates
12:47 ashimema and it passes the existing tests too
12:47 Joubu I am sorry but cannot have a timezone discussion on Friday afternoon
12:47 cait1 I think don't use BLOCKED witout a dependency
12:47 Joubu and had this discussion several times already, with you and Tomas
12:47 khall joined #koha
12:47 ashimema it's fine
12:47 I need to walk away
12:47 Joubu but it will stay a problem if we keep 10 months between 2 discussions
12:47 * cait1 poked the wasp's nest apparently :(
12:47 ashimema yup
12:47 Joubu we need to focus on it at the same time
12:47 cait1 i was just trying to clean up honestly
12:48 ashimema yup
12:48 in short.. I've given up on it
12:48 Joubu I am all open to work on it, but only if we really discuss it, and rethink EVERYTHING from scratch
12:48 to know/understand where we are
12:48 ashimema but I also will adamantly fail anthing that touches dates until someone has taken the time to work through it.
12:48 Joubu (like the basis, are we handling things correctly in the DB? spoiler: no)
12:49 cait1 orrr..... please don't break all my reports by doing crazy things db wise please
12:49 ok?
12:49 Joubu so it's hard to think about handling TZ at the end of the chain (JS) and the beginning is broken
12:49 cait1 :)
12:49 Joubu s/and/when
12:49 cait1 whatever you do... just not that
12:49 Joubu anyway, <3
12:49 not today :)
12:50 cait1 i am ok with that
12:51 ashimema mysql basically doesn't do timezone
12:51 mwah
12:56 Joubu ashimema: we are setting a TZ :) and so assuming that the dates are stored using this TZ.
12:56 this timezone at config level is wrong. The client timezone should be used to convert from/to UTC/floating
12:57 tcohen we should convert all to UTC on the DB
12:57 Joubu (I may be wrong, but all that to say it's not trivial, and thinking JS when it's not clear at the root, it's complicated)
12:58 Dyrcona joined #koha
12:58 tcohen the worst scenario being some JS library sets a local timezone to pass the API, and it gets translated into whatever we have configured, right?
12:58 cait1 tcohen: pleas eno
12:59 it's horror for everyone dealing with the data
12:59 sorry to say that
12:59 tcohen koha-change-timezone <instance> haha
13:00 ashimema the database basically does nothing with datetime or date fields
13:00 timestamps it converts to UTC
13:00 the majority of our fields are datetimes
13:00 those are just treated as 'local' by mysql
13:00 magnuse ashimema++ Joubu++ cait++ tcohen++
13:01 ashimema it does not do any form of timezone handling.. it just assumes you've given it what you want
13:01 so we have to handle this in code
13:01 right now..
13:01 we set an 'instance' timezone
13:01 and when things come out of the database we assume we've put them in in that timezone
13:01 thus we mark them as being in that timezone
13:01 cait1 I am just against converting times because then we'd have to deal with that in all parts we do SQL, migrations etc. and that's quite the horror for me
13:02 can we not tell mysql which timezone the db is "living" in as we set it anyway?
13:02 ashimema I agree cait1
13:02 we don't have such a terrible solution now
13:02 we just don't do it in both directions
13:02 we assume that all incoming dates from the client are also in instance time
13:02 but they're not
13:02 JS libraries like to pass dates with an explicit timezone
13:03 and often like to add one if they're not explicitly given one
13:03 and the one they ad..
13:03 is their local one
13:03 or..
13:03 they offset to utc
13:03 and then we ignore it
13:03 cait1 hmm
13:03 ashimema so right now.. anything coming in via the API we assume is instance time when it will likely be UTC or have a proper offset attached
13:03 tcohen the problem Joubu is trying to highlight, is that that's not what our TT templates do
13:03 ashimema that's ALL this patch does
13:04 properly read the offset and convert it to instance timezone
13:04 tcohen so we would end-up having some hybrid thing
13:04 ashimema TT assumes STRINGS.. those strings are all in instance time
13:04 ignoring browser timezones
13:05 tcohen so TT date-times are in instance time
13:05 ashimema yup
13:05 as far as I'm concerned they are
13:05 tcohen if we switched the checkout input to use a JS library using the browser timezone
13:05 I should've started saying I agree with your patch
13:05 ashimema and in the majority of cases that's OK.. because you're library users tend to be in the same timezone as your instance
13:05 tcohen just thinking out loud about the things we need to fix
13:06 ashimema the moment you get JS involved it gets confused
13:06 cait1 should we do something like... hard stuff fridays?
13:06 tcohen haha
13:06 ashimema I'm just tired guys.. that patch has blocked 3 paid developments of ours already.. for a long time
13:07 cait1 it seems like today is one, but it might be good to do that mroe often to help resolve things earlier
13:07 I can't help much with discussion, but I coudl try and moderate/keep agendas and the like if you want
13:11 alex_a_ joined #koha
13:11 Joubu I said "not today"
13:11 :D
13:12 ashimema yup
13:12 cait1 would it help to have a meeting about it (only that topic)
13:12 for devs to work thorugh?
13:12 just trying to figur out how we can make this work
13:13 * ashimema needs some beers
13:13 ashimema is it too early to start drinking?
13:13 tcohen we need a Zoom/Meet/Jitsi meeting
13:29 cait1 i'd be open for that
13:30 not helpful maybe.. but i can smile and look friendly...
13:31 ashimema I'm off to collect the kids from school shortly and a long weekend after that (out on Monday)
13:31 but could be tempted another time.
13:32 tcohen heh
13:32 'not today'
13:32 cait1 :)
13:33 not today, but hopefully soon? :)
13:33 ashimema replied to your comments on bug 11175 cait
13:33 huginn` Bug https://bugs.koha-community.or[…]_bug.cgi?id=11175 enhancement, P5 - low, ---, jonathan.druart+koha, Passed QA , Show the parent record's component parts in the detailed views
13:33 magnuse ashimema: somewhere in the world it' beer o'clock
13:34 * ashimema is contemplating a single malt at this point
13:34 magnuse ...or whiskey o'clock
13:34 ashimema hehe
13:34 tcohen it is clearly gin and tonic o'clock
13:54 alex_a joined #koha
14:03 alex_a joined #koha
14:14 tuxayo @later tell kidclamp ok to release the 27th? (see my email)
14:14 huginn` tuxayo: The operation succeeded.
14:15 kidclamp tuxayo: all good!
14:16 tuxayo kidclamp: great!
14:17 koha-jenkins Project Koha_21.05_U_Stable build #72: SUCCESS in 32 min: https://jenkins.koha-community[…]1.05_U_Stable/72/
14:25 Joubu see you on Monday, #koha!
14:27 koha-jenkins Project Koha_21.05_U16 build #78: SUCCESS in 42 min: https://jenkins.koha-community[…]oha_21.05_U16/78/
14:42 Project Koha_21.05_U21 build #2: SUCCESS in 57 min: https://jenkins.koha-community[…]Koha_21.05_U21/2/
14:44 Project Koha_21.05_U20 build #80: SUCCESS in 1 hr 0 min: https://jenkins.koha-community[…]oha_21.05_U20/80/
14:50 Project Koha_21.05_D10 build #119: SUCCESS in 1 hr 6 min: https://jenkins.koha-community[…]ha_21.05_D10/119/
14:50 reiveune bye
14:50 koha-jenkins Project Koha_21.05_D12 build #10: SUCCESS in 32 min: https://jenkins.koha-community[…]oha_21.05_D12/10/
14:50 reiveune left #koha
14:57 koha-jenkins Project Koha_21.05_U18 build #74: SUCCESS in 1 hr 12 min: https://jenkins.koha-community[…]oha_21.05_U18/74/
15:08 Project Koha_21.05_D9 build #76: SUCCESS in 41 min: https://jenkins.koha-community[…]Koha_21.05_D9/76/
15:25 Project Koha_21.05_U18 build #75: SUCCESS in 35 min: https://jenkins.koha-community[…]oha_21.05_U18/75/
15:37 Project Koha_21.05_U20 build #81: SUCCESS in 54 min: https://jenkins.koha-community[…]oha_21.05_U20/81/
15:42 Project Koha_21.05_D11 build #94: SUCCESS in 57 min: https://jenkins.koha-community[…]oha_21.05_D11/94/
15:48 Project Koha_21.05_U_Stable build #73: SUCCESS in 39 min: https://jenkins.koha-community[…]1.05_U_Stable/73/
15:55 Project Koha_21.05_D10 build #120: UNSTABLE in 1 hr 4 min: https://jenkins.koha-community[…]ha_21.05_D10/120/
15:58 Project Koha_21.05_U21 build #3: SUCCESS in 32 min: https://jenkins.koha-community[…]Koha_21.05_U21/3/
16:10 Project Koha_21.05_U16 build #79: SUCCESS in 1 hr 12 min: https://jenkins.koha-community[…]oha_21.05_U16/79/
16:32 Project Koha_21.05_U20 build #82: SUCCESS in 54 min: https://jenkins.koha-community[…]oha_21.05_U20/82/
16:33 Project Koha_21.05_U16 build #80: SUCCESS in 34 min: https://jenkins.koha-community[…]oha_21.05_U16/80/
16:35 Project Koha_21.05_D10 build #121: STILL UNSTABLE in 47 min: https://jenkins.koha-community[…]ha_21.05_D10/121/
16:36 Project Koha_21.05_D11 build #95: SUCCESS in 54 min: https://jenkins.koha-community[…]oha_21.05_D11/95/
16:49 Project Koha_21.05_D12 build #11: SUCCESS in 54 min: https://jenkins.koha-community[…]oha_21.05_D12/11/
16:53 AndrewFH joined #koha
16:56 cait1 left #koha
16:59 cait joined #koha
17:06 koha-jenkins Project Koha_21.05_U20 build #83: SUCCESS in 33 min: https://jenkins.koha-community[…]oha_21.05_U20/83/
17:11 tcohen_ joined #koha
17:14 koha-jenkins Project Koha_21.05_D11 build #96: SUCCESS in 38 min: https://jenkins.koha-community[…]oha_21.05_D11/96/
17:21 Project Koha_21.05_U18 build #76: SUCCESS in 1 hr 11 min: https://jenkins.koha-community[…]oha_21.05_U18/76/
17:25 tcohen_ joined #koha
17:30 koha-jenkins Project Koha_21.05_U_Stable build #74: SUCCESS in 54 min: https://jenkins.koha-community[…]1.05_U_Stable/74/
17:31 Project Koha_21.05_D9 build #77: SUCCESS in 59 min: https://jenkins.koha-community[…]Koha_21.05_D9/77/
17:38 Project Koha_21.05_D12 build #12: SUCCESS in 31 min: https://jenkins.koha-community[…]oha_21.05_D12/12/
17:49 Yippee, build fixed!
17:49 wahanui Congratulations!
17:49 koha-jenkins Project Koha_21.05_D10 build #122: FIXED in 59 min: https://jenkins.koha-community[…]ha_21.05_D10/122/
17:54 Project Koha_21.05_U21 build #4: SUCCESS in 39 min: https://jenkins.koha-community[…]Koha_21.05_U21/4/
17:56 khall_ joined #koha
18:00 Oak joined #koha
18:10 koha-jenkins Project Koha_21.05_U20 build #84: SUCCESS in 48 min: https://jenkins.koha-community[…]oha_21.05_U20/84/
18:16 khall joined #koha
18:20 koha-jenkins Project Koha_21.05_D11 build #97: SUCCESS in 49 min: https://jenkins.koha-community[…]oha_21.05_D11/97/
18:31 khall joined #koha
18:33 IGS joined #koha
18:33 koha-jenkins Project Koha_21.05_U21 build #5: SUCCESS in 39 min: https://jenkins.koha-community[…]Koha_21.05_U21/5/
18:35 IGS I'm trying to use MARC field 942$n in my bibliographic framework.  I can't figure out what value goes in that field that causes it to suppress a record.  I read that I should use the value "1" in 942$n but that didn't seem to do anything.
18:46 caroline_crazycatlady IGS there is also a system preference you have to enable
18:47 IGS:  OpacSuppression
18:48 IGS: can you link the page where you read that you have to use 1 to hide a record? I'll make a note to add the information about the syspref there
18:50 IGS https://koha-community.org/man[…]gpreferences.html
18:52 caroline_crazycatlady oh, it's already the the syspref section
18:52 so is your OpacSuppression set to hide?
18:54 IGS It is now.  :-)
18:54 caroline_crazycatlady ok! :)
18:54 is your record hidden?
18:55 IGS yes
18:55 caroline_crazycatlady excellent!
18:55 wahanui darn tootin' it is.
19:07 AndrewFH joined #koha
19:15 fridolin joined #koha
19:15 fridolin hi there
19:42 tcohen_ joined #koha
20:16 alexbuckley joined #koha
20:56 koha-jenkins Project Koha_20.11_U18 build #134: SUCCESS in 38 min: https://jenkins.koha-community[…]ha_20.11_U18/134/
20:59 Project Koha_20.11_U20 build #144: SUCCESS in 41 min: https://jenkins.koha-community[…]ha_20.11_U20/144/
21:09 Project Koha_20.11_U21 build #2: SUCCESS in 53 min: https://jenkins.koha-community[…]Koha_20.11_U21/2/
21:11 Project Koha_20.11_U_Stable build #49: SUCCESS in 55 min: https://jenkins.koha-community[…]0.11_U_Stable/49/
21:20 Project Koha_20.11_D10 build #153: SUCCESS in 1 hr 2 min: https://jenkins.koha-community[…]ha_20.11_D10/153/
21:27 Project Koha_20.11_U16 build #124: SUCCESS in 1 hr 11 min: https://jenkins.koha-community[…]ha_20.11_U16/124/
21:31 Project Koha_20.11_D9 build #147: SUCCESS in 35 min: https://jenkins.koha-community[…]oha_20.11_D9/147/
21:37 Project Koha_20.11_U20 build #145: SUCCESS in 38 min: https://jenkins.koha-community[…]ha_20.11_U20/145/
22:00 Project Koha_20.11_D11 build #170: SUCCESS in 48 min: https://jenkins.koha-community[…]ha_20.11_D11/170/
22:07 Project Koha_20.11_D9 build #148: SUCCESS in 36 min: https://jenkins.koha-community[…]oha_20.11_D9/148/
22:08 Project Koha_20.11_U18 build #135: SUCCESS in 59 min: https://jenkins.koha-community[…]ha_20.11_U18/135/
22:15 Project Koha_20.11_D11 build #171: SUCCESS in 47 min: https://jenkins.koha-community[…]ha_20.11_D11/171/

← Previous day | Today | Next day → | Search | Index

koha1