IRC log for #koha, 2022-05-23

All times shown according to UTC.

Time S Nick Message
05:31 m23 joined #koha
05:51 reiveune joined #koha
05:52 reiveune hello
06:02 * dcook waves
06:03 dcook Looks like Indexdata resolved the memMax issue I reported recently:
06:17 cait joined #koha
06:19 magnuse dcook++
06:20 dcook magnuse: I try to put my complaining to good use heh
06:20 magnuse hehe
06:20 magnuse good plan
06:20 dcook Wish I could figure out this faceting one though..
06:20 dcook
06:20 liliputech joined #koha
06:20 dcook Nearly the same result set but radically different faceting times
06:20 dcook There are times where I think about setting up a Zebra dev environment, but it's 4:21pm on a Monday. Today is not that day heh.
06:26 dcook Actually I feel like I used to have a Zebra docker image around somewhere..
06:27 ashimema morning
06:31 ashimema it's so refreshing to see someone still digging on Zebra and fixing bugs dcook 🙂
06:31 dcook ashimema: I think I might be a masochist...
06:32 dcook Nah, people come to me with problems, and I like fixing them
06:32 dcook Or telling someone enough information for them to fix it hehe
06:32 ashimema me coding knowledge has ended up so Koha specific.. I really aught to branch out again and start learning some now languages and styles..
06:32 ashimema wouldn't have the foggiest looking at zebra code
06:33 dcook Yeah, we all become a bit Koha-centric I think
06:33 dcook I think I started learning C so that I could understand Zebra a bit better.. and then started using C/C++ for Arduinos and such
06:33 dcook That used to help me stay awake when my son was a baby hehe
06:34 ashimema I'm certainly a perl first guy again now.. everything else is rusty.. and even my perl is a bit Koha influenced now.. not always reaching for the right tool
06:34 liliputech dcook: arduino++ (super fun!)
06:34 dcook ashimema: I hear that..
06:34 ashimema I never made it to arduino... been doing micro:bit and pi with the daughter as an alternative
06:35 liliputech ashimema: nice as well :)
06:35 dcook My son's been doing micro:bit at school and it seems cool too
06:35 dcook I used to use a pi for a few home automation projects as well..
06:35 dcook liliputech: Good to see more arduino fans
06:35 ashimema the micro:bit is so accessible for kids.. it has just enough built in to get results immediately
06:35 dcook I haven't done much in years though. Tough to find the time.
06:35 ashimema perfect for the little one's with little patience 😜
06:35 dcook hehe
06:35 dcook "It's not working dad!" Ahhhh
06:35 dcook haha
06:36 dcook Actually, the problem with the arduino is that my imagination ran out
06:36 ashimema ha, yeah.. I have that problem too
06:36 dcook Couldn't think of any easy enough ideas that were actually fun enough for him
06:36 dcook We do have a snap together electronic kit we occasionally bring out though which makes lots of noise, so that's still fun haha
06:36 ashimema imagination and time.. and the fact that pretty much anything I think of already has a cheap out of the box solution these days so time comes in and says.. just buy it.
06:36 dcook Motors and noise :D
06:36 dcook I'm too much of a miser for that I guess hehe
06:37 ashimema haha, nice
06:37 dcook Also clutter. I'm an over the top minimalist..
06:37 dcook Have you played with Rust at all?
06:37 dcook I tried it a bit a few years ago but very... rusty... now
06:37 ashimema most recent project I've wondered about is a climbing wall timer... a guy at the place I volunteer at did one for one of the walls but isn't keen to do a second one so we can have climbing races side by side
06:38 ashimema he went arduino and manualy soldered up a partial LED matrix then wrote all the code in C
06:38 ashimema I think I'd probably start a little higher up the stack and get some more complete components.. haha
06:38 dcook Oh there's all kinds of options for sure
06:39 ashimema Rust has been on my 'to learn' list for literally years.. feels like it's matured allot of late.. now could be the time
06:39 dcook A mate of mine gave me a Seeeduino kit when he was moving abroad
06:39 dcook That comes with this little LCD screen and I think I could download the libraries online or something
06:39 dcook So very little coding for me to do in the end for that one. That was a big hit
06:39 ashimema go too.. though Go doesn't feel as popular as it once was.. maybe I'm just not moving in Go circles any more
06:39 dcook Yeah, I was thinking about that too
06:40 ashimema cool
06:40 dcook I've written some Go programs for fun but all small time stuff
06:40 dcook Again imagination...
06:40 dcook I do some IoT work professionally, and I have a Perl-based device simulator. I'm planning on re-writing that in Go.
06:40 ashimema I haven't really wrapped my head around what Go brings to the table..
06:41 dcook use the goroutine/user-threads to make it work a bit better than an event-driven single-threaded Perl program..
06:41 dcook I think it's big thing was the goroutine/user-threads
06:41 dcook Apparently Java is doing something in that space soon..
06:41 ashimema makes sense
06:41 dcook Loom or something..
06:42 dcook
06:42 ashimema Java, oh Java.. I learnt to code on Java.. at the time it felt nice.. but I've quickly run away from it.. feels ah so heavy now
06:42 dcook ikr
06:42 ashimema did you see bug 30822 dcook?
06:42 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=30822 enhancement, P5 - low, ---, martin.renvoize, Needs Signoff , BatchCommit does not deal with indexation correctly
06:42 dcook I saw the subject line but didn't look into it
06:42 ashimema it's another of those don't queue everything bugs..
06:42 dcook Yeah I figured y'all had those cased
06:43 dcook Btw I like that "task dependencies" idea
06:43 ashimema skirting around the fact that I think both of us are a little against the idea as a concept.. indexing on the go somehow feels better..
06:43 ashimema I think I found a race condition in existing code..
06:43 dcook I wouldn't be surprised regarding the race condition
06:43 dcook As for the concept... I don't know.
06:43 ashimema `BiblioAddsAuthorities` requires up to date indexes as it uses them to search for dupes
06:44 dcook ah yes that does sound familiar
06:44 ashimema so it's fine for single calls of AddBiblio.. but as soon as you stick it in the loop you're very likely to be looking at an index that's out of date..
06:44 alex_ joined #koha
06:44 dcook Yuuup
06:44 ashimema with this patch series we only make that more likely to happen
06:44 dcook Back when I was working on OAI stuff, I quickly learned I couldn't trust the timeliness of the indexing
06:44 dcook It's a tough one though
06:44 dcook But yeah I agree with you. Much more likely to happen with these patches.
06:45 ashimema I stuck as to a way around it... well, not without introducing a task dependency stack.
06:45 alex_ Bonjour
06:45 dcook I reckon just throw all the indexing stuff into a queue, and just process it ASAP
06:45 dcook yeah I'm stuck on that one too
06:45 dcook I mean even if you had the task dependency..
06:45 ashimema but even with such a stack it's going to be a pain for bulk stuff as you'd pretty much need to rebuild indexes between each call regardless right
06:45 dcook How would you really know what's the most recent?
06:45 ashimema exactly
06:46 ashimema glad it's not just me then
06:46 dcook Nope. Not just you hehe
06:46 ashimema zebra queue is remarkably elegant
06:46 dcook I think I've just given up that hill
06:46 dcook Is it? Haha
06:46 ashimema well, it's simple and works at least.. mostly
06:46 dcook Yeah I was just about to say that heh
06:46 dcook It mostly gets the job done
06:47 dcook I've actually had a number of ideas over time about ways to optimize it even further..
06:47 dcook But
06:47 ashimema I think our task queue is stuck in a middle place at the moment.. it's either way over engineers or too underengineered for most tasks I come to want to use it.
06:47 dcook ^
06:47 dcook I'd agree with that for sure
06:47 dcook I just avoid it
06:47 dcook Which is too bad
06:47 ashimema indeed
06:48 dcook That said... I am hoping to resurrect some OAI stuff and I'd really like to use the message queue to handle it..
06:48 ashimema oh well.. onwards and upwards.. we've made our decisions so I am stuck continueing the work to try and get what I feel we need.
06:48 dcook But it might mean locally bolting something against it
06:48 ashimema cool
06:48 dcook Yeah, pros and cons to group projects for sure
06:49 ashimema how's your ES knowledge?
06:49 dcook Pretty limited unfortunately although we've started using it more for other projects
06:49 dcook I'm planning at some point to start looking at it for Koha
06:50 dcook Especially for our larger Koha instances
06:50 * ashimema has been pondering our use of the ES api's for record management.. seems there's bulk routines and also quick, small change, routines..
06:50 ashimema but we're using the builk one's for everything
06:50 dcook Hmm that sounds interesting
06:50 dcook Overall, I figure bulk changes tend to be more efficient, but I suppose it depends on what it is
06:50 ashimema things like this index queue stuff made me look and think.. aren't we just using the wrong api's entirely for half of this stuff
06:51 dcook Ohh interesting.
06:51 dcook I wouldn't be surprised but alas I don't know enough to comment meaningfully
06:51 ashimema we do one heck of a lot of 'change one field in one record' stuff in Koha.. like for a checkin
06:51 ashimema it feels a bit crazy to call a bulk api for that sort of change
06:52 dcook Hmm probably an artifact of the original compatibility work
06:52 dcook There were a number of choices made back then that I think were unfortunate..
06:52 ashimema I agree with you.. doing stuff in bulk generally is very optimized.. but for such a small change I'd put money on the added setup and take down cost of a bulk job to be more expensive than the small single record change
06:52 ashimema I was hoping someone might have some knowledge to put it to the test.
06:52 dcook At some point ;)
06:52 ashimema support I'll roll up my sleaves and try it.. haha
06:53 dcook That's the spirit haha
06:53 ashimema indeed
06:53 dcook That's actually what I was just thinking with this Zebra faceting. I'm out of my current depth, but the only solution besides waiting for someone else is to dive in myself
06:54 ashimema I can see us slitting back up what we do.. with Joubu's recent work everything now goes through the queue.. my gut tells me we need to be picky about what gets queued and what happens with a small fast change in real time
06:55 ashimema rather than taking a one size fits all approach.. in either direction
06:55 ashimema kinda wouldn't surprise me if Zebra has a similar flaw in our use.. but I've never dug into it as zebra's never really caused me pain on that front.
06:55 ashimema reading the history it once did.. seems that's whay we went the queue and daemon approach in the first place
06:56 ashimema I remember spending lots of time converting that to a daemon instead of a cronjob..
06:56 cait joined #koha
06:56 ashimema hehe.. back in the days when was sat in the cron
06:56 dcook Yeah, zebraidx just has that CLI utility and you have to point it at a directory :/
06:56 dcook Back when I was using openSUSE instead of Debian/Ubuntu...
06:56 dcook Yeah far out... you're making me have flashbacks
06:57 dcook Ahh how is it already this late..
06:57 ashimema Haha, sorry.. I distracted you
06:59 dcook Ah it's ok. I probably didn't have enough time anyway heh
07:01 dcook Debating whether to head to the gym tonight or telling myself I'm too tired.. :|
07:01 dcook While quickly trying to compile Zebra from source before I leave the office heh
07:07 ashimema Lol
07:07 ashimema Sadist ;)
08:27 cait dcook: for your facets question - we had to turn it off for bigger isntallations - it really slows things down too much
08:27 cait dcook: you are aware of the bug with the discussion about this? tcohen had looked into fixing it back then
08:28 cait bug 13665
08:28 huginn Bug https://bugs.koha-community.or[…]_bug.cgi?id=13665 normal, P5 - low, ---, gmcharlt, NEW , Retrieve facets from zebra is slow
08:33 cait maybe it has some clues
08:40 cait ... and good morning?
08:45 magnuse @later tell gmcharlt could you make some more people ops, please?
08:45 huginn magnuse: The operation succeeded.
08:47 * cait waves
08:48 ashimema who needs to be an op?
08:48 * ashimema has never opped on irc.. not sure why we need more?
08:49 ashimema didn't think you were around today cait?
08:49 cait the conference is next week
08:49 * ashimema has a meeting about 856's in consortia and how to handle them and wondered if you were the person to talk to
08:49 cait interesting
08:50 cait what#s the question?
09:29 paulderscheid[m] Hi Koha, can someone point me to resources for unit testing in perl?
09:31 cait there is somethign form oreilly I have used some in the past... but I am nto the biggest patch writer
09:31 paulderscheid[m] OK
09:32 paulderscheid[m] Perl Testing: A Developer's Notebook by Ian Langworth, Chromatic?
09:32 Joubu paulderscheid[m]: start with
09:32 paulderscheid[m] Thanks cait Joubu
09:32 cait paulderscheid[m]: was just seaching for it, I don't have it in the office it appears
09:33 Joubu we are also using Test::MockModule and Test::Warn
09:33 wahanui okay, Joubu.
09:33 cait the cover looks familiar, but the module docs are probably newer
09:33 paulderscheid[m] Joubu: Ok
09:33 paulderscheid[m] I'll look into those as well
09:35 paulderscheid[m] What's the difference between Test::More and Test2::Suite?
09:44 cait[…]S-FROM-Test::More maybe
09:45 cait the cpan page doesn't really explain the why...
09:45 paulderscheid[m] Thanks
09:46 ashimema Test2 is newer and apparently learnt from all the caveats of Test.. we're mostly using Test:: in Koha as there's no need to migrate
09:47 ashimema we do have the odd Test2 test now I think.. sure I saw one or two references to it.. but I might be wrong as I can't lay my hands on any now
09:50 paulderscheid[m] Thanks ashimema that's what I expected.
09:55 Joubu It's not my understanding
09:55 Joubu
09:55 Joubu STOP! If you're just getting started writing tests, have a look at Test2::Suite first.
09:55 Joubu This is a drop in replacement for Test::Simple which you can switch to once you get the hang of basic testing.
09:59 m23 joined #koha
09:59 paulderscheid[m] I also read that, but since Test::More is the current standard I'll use that, right?
10:06 Joubu yes
10:26 kidclamp joined #koha
10:47 cait ok, now I got confused, why the Test2::Suite first for learning?
10:47 cait oh, you quotes from the page actually
11:44 zb any guesses why exporting records using a file of biblionumbers fails for one user (windows with two different browsers) but not another (linux)? it works if you instead just give a range of biblionumbers
11:46 zb already tested that minimal user rights needed are catalogue and export_catalog. and it does not seem to be a (CR)LF file format issue
11:49 zb my only possible clue is that the export is rather slow, even when the file only contains a few biblionumbers -> some kind of timeout issue?
12:02 cait the timeout usually is more server side
12:03 cait so not sure what it woudl be different - maybe someone else hs some insight
12:04 ashimema file encoding or line endings would be my bet.. but I'm not really sure how I'd investigate that hunch
12:05 ashimema seems like you already considered that..
12:05 ashimema pass
12:27 cait Joubu: I believe 30831 is a duplicate to soemthing we had both worked on... do you remember?
12:40 liliputech joined #koha
12:55 Joubu yes 28152, you found it apparently :)
12:55 Joubu written 1 year ago..
13:05 ashimema cait stilla round?
13:06 tcohen hola
13:06 ashimema hi tochen 🙂
13:15 cait yes
13:16 cait ashimema:
13:29 ashimema I was looking to get bug 7021 moving again
13:29 huginn Bug https://bugs.koha-community.or[…]w_bug.cgi?id=7021 enhancement, P5 - low, ---, emmi.takkinen, Failed QA , Add patron category to the statistics table
13:30 ashimema you failed it
13:30 ashimema was it just on column name.. would you accept patron_categorycode ?
13:30 ashimema I agree that 'usercode' is weird and inconsistent.. but I also agree categorycode could be misconstrued easily enough too
13:32 ashimema having said that.. we don't re-use that term anywhere meaning anything else do we
13:33 cait sorry, got distracted again (translating still..)
13:33 Joubu pseudonymized_transactions has categorycode
13:33 ashimema so categorycode should be fine.. cool
13:33 cait i think usercode is not so good... if user it woudl be more of a groupß
13:33 cait ?
13:33 ashimema I think we need to move or drop the script that patchset includes too
13:33 cait categorycode I'd like better because people recognize it
13:33 cait if we use something different on the api also ok with that
13:33 ashimema it's in a weird place
13:34 Dyrcona joined #koha
13:34 cait we have category_id in the api
13:35 cait a script for populating the column in older databases?
13:35 cait ashimema: basically i'd just like to avoid a third name for the same thing I guess
13:36 Joubu I also meant that people could enable the pseudonymization feature if they want to retain the info :)
13:36 Kathir joined #koha
13:36 Kathir Hi
13:36 wahanui what's up, Kathir
13:36 cait possibly true, but maybe a little more complicated
13:36 ashimema yeah, I realise
13:36 cait well no
13:36 ashimema but not everyone is enabling that yet 😜
13:36 Joubu it's just a syspref ;)
13:36 cait i think the issue might be a little different
13:37 cait we log the patron category at the time of the transaction
13:37 ashimema yes cait, there's a script
13:37 cait so it might change
13:37 Kathir
13:37 cait when kids get to be adults etc... we had that request before for statistics
13:37 Kathir This our student teacher login
13:37 cait what's your question Kathir?
13:37 Joubu I don't understand cait
13:38 Joubu pseudonymized_transactions will contain the info from when the transaction takes place
13:38 cait maybe I misunderstand you, I was just saying it's the right place to log - because it's something that might change during the life of a patron, so the lookup on borrowers might not be accurate
13:38 Kathir this IP to We need point our AWS as sub domain
13:38 cait Joubu: ah ok, where does it draw from?
13:40 Kathir How to point this IP to
13:41 Kathir Are you there?
13:41 Joubu cait: ? pseudonymized_transactions.categorycode = borrowers.categorycode
13:43 Kathir left #koha
13:49 ashimema I find it really confusing have two tables that do nearly the same stuff here..
13:49 ashimema well.. 3 if you also count action_logs
13:50 Joubu action_logs is different
13:51 Joubu if you want real statistics, on the long term, with anonymization of the data, you will need to enable the pseudonymization feature
13:52 ashimema OK, new question then
13:53 ashimema is there a mgiration path from statistics to pseudonymization feature and is there a plan to drop statistics entirely for the replacement
14:00 cait Joubu: I should have phrased it differently I guess - the transaction itself, where does it get that from
14:01 cait I think not all libraries want to anonymize
14:01 cait pseudononymize
14:01 cait so a migration makes nos sense, it woudl break someo people's workflow
14:02 cait and then there is always a need for at lest having more information for a certain amount of time and what is considered safe for pseudonymization
14:03 cait like... running a report about which age group uses which material the most
14:03 Joubu ashimema: no, but that should be easy to migrate.
14:03 ashimema indeed
14:03 ashimema might be nice to have an off the shelf script for that
14:04 Joubu cait: the pseudo* tables are filled in with the transaction's info at the moment the transaction is happening
14:04 Joubu (right after it happened to be exact)
14:17 cait hm so we store a lot of similar but not quite the same data
14:40 cait I got to run from the incoming storm/rain :(
14:40 cait bye #koha
14:40 cait left #koha
15:10 reiveune bye
15:10 reiveune left #koha
15:26 cait joined #koha
15:35 alex_ joined #koha
15:39 alex__ joined #koha
17:03 khall joined #koha
17:43 khall_ joined #koha
18:06 fridolin joined #koha
18:07 fridolin hi tehre
18:13 tcohen hi fridolin !
18:13 fridolin tcohen: yey
18:13 * fridolin coffeeeeeeee
18:13 tcohen same here, need another doze
18:13 fridolin had Théo birthday this weekend, 5 years old now
18:14 fridolin he has a Gaming device now, Nintendo 2DS
18:14 fridolin (daddy also injoys)
18:19 tcohen :-D :-D
18:19 tcohen yay
18:20 tcohen do you have a soundtrack for the release?
18:48 m23 joined #koha
18:51 tcohen #koha can schema relationships include filters on the resultset?
18:54 huginn News from kohagit: Bug 30747: (follup-up) Add missing otherholdings_datelastborrowed column in settings <[…]cdc2e25aa0c374acd>
18:54 huginn News from kohagit: Bug 30730: (follow-up) Simplify query <[…]8c7760ffd11903b69>
18:54 huginn News from kohagit: Bug 30747: Make column settings work for otherholdings table using forEach loop <[…]881090d7337a41fea>
18:54 huginn News from kohagit: Bug 30730: Unit tests <[…]86502a4139b492e4c>
18:54 huginn News from kohagit: Bug 30730: Holds to Pull should not list items with a notforloan status <[…]4225ebb675b0ca09d>
18:56 fridolin tcohen: where can i find task to do for the big release please ?
18:56 fridolin i have to update contributors, history ...
18:59 tcohen I don't know!
18:59 tcohen ashimema?
18:59 wahanui somebody said ashimema was back on security bugs today.. sorry, it's been a bit of a crazy week
19:02 fridolin no pb i'll mail all assistants ;)
19:09 tcohen :-D
19:09 tcohen fridolin++
19:09 tcohen koha++
19:16 alohabot 🎁 🦄 Koha 'master' packages pushed to 'koha-staging' repo  ðŸˆðŸ§ðŸ¥¯
19:36 koha-jenkins Project Koha_Master_D9 build #1977: SUCCESS in 42 min: https://jenkins.koha-community[…]a_Master_D9/1977/
19:39 koha-jenkins Project Koha_Master_U20 build #409: SUCCESS in 44 min: https://jenkins.koha-community[…]a_Master_U20/409/
20:01 koha-jenkins Project Koha_Master_D11_MDB_Latest build #926: SUCCESS in 1 hr 5 min: https://jenkins.koha-community[…]1_MDB_Latest/926/
20:13 koha-jenkins Project Koha_Master_D11_My8 build #843: SUCCESS in 36 min: https://jenkins.koha-community[…]ster_D11_My8/843/
20:26 koha-jenkins Project Koha_Master_U22 build #69: SUCCESS in 47 min: https://jenkins.koha-community[…]ha_Master_U22/69/
20:32 koha-jenkins Project Koha_Master build #2031: STILL UNSTABLE in 1 hr 8 min: https://jenkins.koha-community[…]Koha_Master/2031/
20:50 koha-jenkins Project Koha_Master_D11_My8 build #844: SUCCESS in 36 min: https://jenkins.koha-community[…]ster_D11_My8/844/
21:04 bag joined #koha
21:08 koha-jenkins Project Koha_Master_D10 build #655: SUCCESS in 55 min: https://jenkins.koha-community[…]a_Master_D10/655/
21:09 koha-jenkins Project Koha_Master_D10 build #656: SUCCESS in 42 min: https://jenkins.koha-community[…]a_Master_D10/656/
21:09 koha-jenkins Project Koha_Master_D12 build #144: SUCCESS in 1 hr 8 min: https://jenkins.koha-community[…]a_Master_D12/144/
21:45 bag joined #koha
22:38 koha-jenkins Project Koha_20.11_U20 build #228: FAILURE in 20 min: https://jenkins.koha-community[…]ha_20.11_U20/228/
22:50 alohabot 🎁 🎃 Koha '20.11.x' packages pushed to 'koha-staging' repo  ðŸ¬ðŸ¥¬ðŸ¥•
22:51 koha-jenkins Project Koha_20.11_U18 build #209: SUCCESS in 33 min: https://jenkins.koha-community[…]ha_20.11_U18/209/
23:08 koha-jenkins Project Koha_20.11_D10 build #217: SUCCESS in 49 min: https://jenkins.koha-community[…]ha_20.11_D10/217/
23:12 dcook @later tell cait Thanks for that facet bug but I was the last person to comment on it nearly a year ago hehe
23:12 huginn dcook: The operation succeeded.
23:13 dcook @later tell fridolin Oh cool. Where did you get a Nintendo 2DS? I feel like Nintendo switch lite is the only thing widely available now?
23:13 huginn dcook: The operation succeeded.
23:18 koha-jenkins Project Koha_20.11_U16 build #182: SUCCESS in 39 min: https://jenkins.koha-community[…]ha_20.11_U16/182/
23:23 koha-jenkins Yippee, build fixed!
23:23 koha-jenkins Project Koha_20.11_U20 build #229: FIXED in 31 min: https://jenkins.koha-community[…]ha_20.11_U20/229/
23:23 wahanui Congratulations!
23:31 koha-jenkins Project Koha_20.11_D11 build #245: SUCCESS in 56 min: https://jenkins.koha-community[…]ha_20.11_D11/245/
23:47 koha-jenkins Project Koha_20.11_D9 build #213: SUCCESS in 1 hr 28 min: https://jenkins.koha-community[…]oha_20.11_D9/213/

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