IRC log for #koha, 2020-11-10

All times shown according to UTC.

Time S Nick Message
00:00 khall joined #koha
00:30 lisettelatah joined #koha
00:31 inlibro joined #koha
00:33 koha-jenkins Project Koha_20.05_D11 build #58: STILL UNSTABLE in 1 hr 31 min: https://jenkins.koha-community[…]oha_20.05_D11/58/
01:27 hayleymapley_ joined #koha
01:31 inlibro joined #koha
01:45 khall joined #koha
02:31 inlibro joined #koha
02:35 hayleymapley__ joined #koha
03:14 deb-CSPL joined #koha
03:32 inlibro joined #koha
04:13 hayleymapley_ joined #koha
04:15 khall joined #koha
04:32 inlibro joined #koha
05:32 inlibro joined #koha
05:40 josef_moravec joined #koha
05:56 fridolin joined #koha
05:59 enkidu joined #koha
06:01 did joined #koha
06:04 chriss joined #koha
06:16 fridolin1 joined #koha
06:24 Oak joined #koha
06:32 inlibro joined #koha
06:55 alexbuckley joined #koha
06:56 davidnind joined #koha
07:06 cait joined #koha
07:10 severine_q joined #koha
07:21 alexbuckley joined #koha
07:31 lmstrand joined #koha
07:32 inlibro joined #koha
07:43 alex_a joined #koha
07:44 alex_a Bonjour
07:58 cait joined #koha
07:59 lds joined #koha
07:59 cait1 joined #koha
08:01 ashimema[m] Mornin' #koha
08:06 cait1 good morning all
08:06 cait1 and ashimema[m] :)
08:06 cait1 hm no magnuse
08:06 cait1 @seen magnuse
08:06 huginn cait1: magnuse was last seen in #koha 3 days, 21 hours, 4 minutes, and 44 seconds ago: <magnuse> yeah, more examples would be awesome!
08:08 davidnind hi cait! and ashimemea[m]!
08:08 ashimema[m] Grr.. why do I have [m] in my name again
08:08 alreadygone joined #koha
08:09 davidnind m for magnificant?
08:10 davidnind erh.. magnificent
08:15 dcook joined #koha
08:17 cait1 ashimema[m]: how would we know? :)
08:20 Oak__ joined #koha
08:26 ashimema[m] Lol
08:26 ashimema[m] tuxayo around yet?
08:33 inlibro joined #koha
08:36 fridolin joined #koha
08:50 tuxayo ashimema: yes :-)
08:50 ashimema[m] still using matrix?
08:50 ashimema[m] ooh.. yes.. you get 'tuxayo is typing' notifications
08:50 tuxayo Ho, the [m]
08:50 tuxayo Let me find where to change it
08:58 kohaputti joined #koha
08:58 minski joined #koha
08:58 ashimema did it actually drop to [m] now
09:00 kohaputti anybody know contact info for Bernardo González Kriegel / translate.koha-community.org admin? The site is down and I would like to inform about it.
09:01 davidnind ashimema: yes!
09:01 ashimema joined #koha
09:01 sophie_m joined #koha
09:02 tuxayo kohaputti: I'm PM-ing their email
09:02 kohaputti thanks
09:04 enkidu joined #koha
09:06 ashimema should be back now kohaputti
09:06 tuxayo it is!
09:06 * ashimema also has the keys to that server.. but knows little about how pootle actually works
09:06 kohaputti ah yes it is
09:06 ashimema most of the time a apache restart does the trick
09:07 kohaputti it appeared to have been down on many "mornings" UTC+3 time.
09:07 ashimema weird
09:08 kohaputti we will see if it goes down tomorrow again
09:08 tuxayo ashimema: ok if add you to the Website_Administration wikipage? If yes, do you also have administrator privileges on Pootle?
09:08 ashimema sure
09:09 ashimema I only recently got access and forgot to update the page
09:09 cait1 it might be a heavy job taking it down?
09:09 cait1 processing updates or so
09:09 ashimema I do indeed have full access.. though, as I said.. I'm not pootle guru.. I do the bare minimum to get it going again and defer to bernardo if it's not simple
09:10 ashimema indeed
09:10 cait1 kohaputti: a good way to find email addresses is bugzilla, you can type in a name in CC for example and it pulls up the candidates
09:11 Joubu tuxayo: on your way, please add me as well
09:13 tuxayo Joubu: ok, administrator privileges on Pootle also?
09:13 ashimema +1 :)
09:14 ashimema it's good to have such access spread accross timezones
09:14 cait1 I got an admin account
09:14 ashimema we're a big community not
09:14 cait1 if you want to list that too
09:14 ashimema s/t/w/
09:14 cait1 but no servr access
09:15 * ashimema has web admin level access to bugzilla too.. but we don't appear to list that
09:15 ashimema as does Joubu I think?
09:16 tuxayo Let's update all! :)
09:16 Joubu yep
09:20 tuxayo diff of changes done so far:
09:20 tuxayo https://wiki.koha-community.or[…]26844&oldid=25749
09:21 ashimema looks good to me tuxayo, thanks :)
09:24 tuxayo yw
09:24 tuxayo Maybe on additional person is needed to have shell access to the bugzilla server. To relaunch splitter ^^"
09:24 tuxayo http://splitter.koha-community.org/?
09:25 ashimema that's a difficult one
09:25 tuxayo oleonard would +1 a lot
09:25 tuxayo oh :o
09:26 ashimema I believe it's a personal server of Chris's
09:26 ashimema and it hosts a whole bunch of things.. so understandable he's been a little reluctant to add too many users
09:26 ashimema I'll ask him
09:29 tuxayo thanks!
09:33 inlibro joined #koha
09:36 enkidu1 joined #koha
09:38 alexbuckley joined #koha
10:21 JCM joined #koha
10:24 TimothyAlexis joined #koha
10:26 TimothyAlexis On the serials/serials-collection page, the ones with status 'Expected' and 'Late' used to be automatically checked in earlier versions. When we upgraded do 20.05.03, this functionality changed. Does anyone here know which patch changed this?
10:27 TimothyAlexis I added bug 26846 after cait1 also mentioned noticing it.
10:27 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=26846 normal, P5 - low, ---, koha-bugs, NEW , serial-collections page no longer check the expected and late serials automatically
10:33 inlibro joined #koha
10:34 joe1 joined #koha
10:39 Joubu Can I get a quick SO+QA on bug 26986 please?
10:39 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=26986 normal, P5 - low, ---, jonathan.druart, Needs Signoff , Second try to prevent Selenium's StaleElementReferenceException
10:39 tuxayo Joubu: yes!
10:41 Joubu tuxayo: I managed to make it fail locally after ~10-15 runs
10:42 Joubu easier to test if you remove the other tests and keep only the "branches" part
10:42 Joubu s/to test/to recreate
10:43 tuxayo ok!
11:12 fridolin joined #koha
11:15 tuxayo Joubu: I got a failure due to a certificate expiration!?
11:15 tuxayo https://copycat.drycat.fr/?a39[…]UTbwZQWXHjjtUZJRq
11:15 tuxayo After 5 run, again
11:15 khall joined #koha
11:16 tuxayo That not the expected one right?
11:18 Joubu the expected error is An element could not be located on the page using the given search parameters: //*[@id="branchname"],xpath at /kohadevbox/koha/t/lib/Selenium.pm line 126.
11:18 Joubu you see the SSL cert because of something else, ignore it
11:19 Joubu but if you see the SSL cert error, it means that you got the other error
11:19 tuxayo Indeed! ^^
11:19 Joubu (error is uploaded, and the upload fails)
11:19 Joubu That's something else I need to deal with..
11:21 tuxayo InddeWait, that mean I still got the expected error? And then the post error hook fails.
11:21 tuxayo *Indeed, I got the error.
11:21 tuxayo Thanks
11:26 fridolin joined #koha
11:26 khall joined #koha
11:31 khall_ joined #koha
11:33 inlibro joined #koha
11:41 khall joined #koha
11:47 davidnind Wooho! Signed off bug 19532 - I hope that is the last time I have to go through the test plan (meant in a good way!), and happy QAing to the QA Team for 21.05!
11:47 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=19532 new feature, P1 - high, ---, aleisha, Signed Off , Recalls for Koha
11:47 davidnind Is 666 comments a record (or near record) for a bug?
11:49 AndrewFH joined #koha
11:51 TimothyAlexis Joubu++
12:04 davidnind left #koha
12:09 ashimema Wow.. we hit the devil number
12:09 ashimema I almost think that should require a new bug opened to start again.. clearly the devil has something to say... Haha
12:18 magnuse joined #koha
12:28 tuxayo davidnind++
12:29 tuxayo The 154 steps test plan!!!!!
12:29 tuxayo congrats
12:30 alex_a_ joined #koha
12:30 tuxayo Joubu: the fix works great :) , the other tests also pass
12:33 inlibro joined #koha
12:33 fridolin joined #koha
12:34 fridolin joined #koha
12:36 magnuse "No property notforloan" for Koha::Serial::Item" anyone else seen that after upgrade to 20.05.x?
12:36 magnuse specifically 20.05.05
12:39 magnuse on a 20.05.04 i see "No property itype for Koha::Serial::Item" - both when receiving an issue of a periodical
12:44 tcohen morning
12:44 tuxayo o/
12:44 tcohen \o
12:47 Joubu magnuse: do you have a line number?
12:52 magnuse Joubu: No property itype for Koha::Serial::ItemUse of uninitialized value in concatenation (.) or string at /usr/share/koha/opac/cgi-b​in/opac/opac-MARCdetail.pl line 311
12:53 magnuse the one for Koha::Serial::Item is at /usr/share/perl5/Exception/Class/Base.pm line 88, which does not say a lot. i guess
12:55 Joubu I don't recreate on master, trying 20.05.05
12:56 Joubu can you detail how you proceed? Now subscription, then "received", mark it as "arrived"?
12:56 Joubu are items created?
12:57 Joubu magnuse: ?
13:04 magnuse sorry, it's when you go to mark an item as received, get the form for creating an item and then save
13:05 Joubu got it, you need makePreviousSerialAvailable
13:05 Joubu recreated on master, can you open a bug report please?
13:06 magnuse looks like the item does not show on the record, but the issue i received is not suggested for being received anymore
13:06 magnuse will do
13:09 magnuse Joubu: bug 26987
13:09 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=26987 major, P5 - low, ---, koha-bugs, NEW , No property notforloan for Koha::Serial::Item
13:10 magnuse i do have makePreviousSerialAvailable = Make, bth
13:10 magnuse s/bth/btw/
13:13 AndrewFH joined #koha
13:15 tuxayo Joubu: did you ever reproduced bug25514 (REST/Plugin/Objects.t) locally?
13:15 Joubu nope
13:16 Joubu but tcohen wrote that the patch won't fix the problem
13:17 magnuse Joubu: testing now
13:24 tuxayo Joubu: thanks, I misunderstood the meaning.
13:33 inlibro joined #koha
13:45 Dyrcona joined #koha
14:01 huginn News from kohagit: Bug 23916: (follow-up) Adapt all the things to the new column and accessor names <http://git.koha-community.org/[…]2454e306cbbc20bba>
14:01 huginn News from kohagit: Bug 23916: DBRev 20.06.00.065 <http://git.koha-community.org/[…]34e8c98ff4135d527>
14:01 huginn News from kohagit: Bug 23916: Allow embedding the issuer object <http://git.koha-community.org/[…]963427a39a735b1f3>
14:01 huginn News from kohagit: Bug 23916: (QA follow-up) Adapt all the things to the new column and accessor names <http://git.koha-community.org/[…]823449e265d6add26>
14:01 huginn News from kohagit: Bug 23916: Add new attribute to API <http://git.koha-community.org/[…]8c5b532e1c4dead94>
14:01 huginn News from kohagit: Bug 23916: (QA follow-up) Rename issuer => issuer_id [DB] <http://git.koha-community.org/[…]dc3ff03f17ba1610a>
14:01 huginn News from kohagit: Bug 23916: (follow-up) Add unit tests <http://git.koha-community.org/[…]8ab89be8e16461ee4>
14:01 huginn News from kohagit: Bug 26594: (QA follow-up) Make _UpdateBorrowerDebarmentFlags a public method <http://git.koha-community.org/[…]98a6877166026b8bb>
14:01 huginn News from kohagit: Bug 23916: (QA follow-up) Remove stray changes and fix typo <http://git.koha-community.org/[…]99574107f9af7477a>
14:01 huginn News from kohagit: Bug 23916: (follow-up) Fix unit test <http://git.koha-community.org/[…]278fcd71fbf765c04>
14:01 magnuse wheee
14:01 huginn News from kohagit: Bug 23916: (follow-up) Rebase unit tests <http://git.koha-community.org/[…]c2f9b0a18c9dfdbae>
14:01 huginn News from kohagit: Bug 23916: (follow-up) Fix terminology, use patron-title and make column consistent... <http://git.koha-community.org/[…]7aae2e24de410925a>
14:01 huginn News from kohagit: Bug 26594: add Unit Test <http://git.koha-community.org/[…]2cbee5531b91c3f41>
14:01 huginn News from kohagit: Bug 26594: Fix patrons merge problem with restrictions <http://git.koha-community.org/[…]1efcafb48b6d230d7>
14:01 huginn News from kohagit: Bug 23916: (follow-up) Don't anonymise issuer and don't update action_logs on upgrade <http://git.koha-community.org/[…]19af4909abb42f5c2>
14:01 huginn News from kohagit: Bug 23916: Record and display item issuer <http://git.koha-community.org/[…]bee8fb4a71e6d875e>
14:01 huginn News from kohagit: Bug 23916: (follow-up) Fix column name collision <http://git.koha-community.org/[…]086d050ea8f42f7a2>
14:01 huginn News from kohagit: Bug 23916: Add unit tests <http://git.koha-community.org/[…]ad4a883b5ac286f1a>
14:01 huginn News from kohagit: Bug 23916: Anonymise 'issuer' when required <http://git.koha-community.org/[…]16bb835e9edfdc1a3>
14:01 huginn News from kohagit: Bug 23916: Add "RecordIssuer" syspref <http://git.koha-community.org/[…]5548193da736b05bd>
14:03 magnuse Joubu++ for a quick solution
14:22 cait1 bug  26987
14:22 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=26987 major, P5 - low, ---, jonathan.druart, Signed Off , No property notforloan for Koha::Serial::Item
14:22 cait1 duh, why did i not find that yesterday? ok :)
14:22 cait1 magnuse++
14:22 cait1 and why is that not pqa yet?
14:26 ashimema caroline around?
14:26 caroline hello!
14:26 ashimema any thoughts on merge request 416 (the cash management stuff) for the manual?
14:26 ashimema is it awaiting further work from me?
14:26 caroline I haven't looked at it yet -_- sorry
14:27 ashimema no worries
14:27 caroline I think we only have to find where to put it right?
14:27 ashimema I added it as a top level ;)
14:28 ashimema the only thing I needed guidance on was adding images.. but I can do that if that's what is outstanding
14:29 caroline oh ok, images are easy
14:29 ashimema also
14:29 caroline first add then somewhere in the images folder
14:29 caroline then create an alias in images.rst
14:29 ashimema did you add the descriptions of account actions I added in a comment to the cancel debits bug
14:29 ashimema I can submit that too
14:29 caroline and then add the alias in your chapter
14:30 ashimema but don't those aliases in images.rst clash awefully often causing a need to rebase?
14:30 caroline no I havent
14:30 * ashimema wondered why we use an incremental number as a filename
14:30 caroline only if many people sumbit at the same time
14:30 ashimema haha..
14:30 ashimema ok..
14:31 ashimema I'll just follow that path than
14:31 ashimema clearly you don't get as many submissions as I thought ;)
14:31 caroline We'd like a different method, but it's just so much work on top of what we already have to do
14:31 caroline it's not high priority
14:31 ashimema yeah.. I understand
14:32 lmstrand_ joined #koha
14:33 ashimema OK.. I'll fix up that image and then this merge request should be ready for you
14:33 caroline thanks!
14:34 inlibro joined #koha
14:34 ashimema also: did you add the descriptions of account actions I added in a comment to the cancel debits bug
14:34 caroline no I haven't
14:35 ashimema OK, I'll submit something for that too
14:35 caroline thanks so much!
14:36 ashimema haha.. it's not entirely altruistic.. it's a good place for me to note it down so I remember how it fits together myself ;)
14:36 caroline still, that you do the work, it helps everyone :)
14:50 koha-jenkins Project Koha_Master_D9 build #1491: SUCCESS in 46 min: https://jenkins.koha-community[…]a_Master_D9/1491/
14:54 koha-jenkins Project Koha_Master_U20 build #139: SUCCESS in 50 min: https://jenkins.koha-community[…]a_Master_U20/139/
14:54 ashimema caroline.. do we ever actually run 'make spelling' on the manual
14:54 caroline I don't know
14:54 ashimema I just did to double check I'd not introduced some new issues
14:55 koha-jenkins Project Koha_Master_D9_MDB_Latest build #437: SUCCESS in 51 min: https://jenkins.koha-community[…]9_MDB_Latest/437/
14:55 caroline I turn it off locally because I have problems with it
14:55 ashimema and it spits out 2748 mispelled words!
14:55 caroline but it's good to know that it works!
14:55 ashimema I think at least some could be added to the custom words list
14:55 caroline Does it give you the list of misspelled words?
14:55 ashimema like 'subfield'
14:56 ashimema we've settled on subfield not sub-field haven't we
14:56 ashimema so subfield should be OK
14:56 ashimema it gives a list of mispelled words and what line or what file they appear on
14:56 ashimema deffo some are issues..
14:56 ashimema but many just need to be added to our custom vocal file
14:56 ashimema like 'opac'
14:56 caroline can you create an issue or a bug report with it?
14:56 ashimema 'subfield'
14:56 ashimema deffo
14:57 caroline ashimema++
15:06 koha-jenkins Project Koha_Master_D9_My8 build #460: STILL UNSTABLE in 1 hr 4 min: https://jenkins.koha-community[…]aster_D9_My8/460/
15:08 did joined #koha
15:20 oleonard joined #koha
15:34 inlibro joined #koha
15:34 koha-jenkins Project Koha_Master_U18 build #941: SUCCESS in 44 min: https://jenkins.koha-community[…]a_Master_U18/941/
15:42 koha-jenkins Yippee, build fixed!
15:42 wahanui Congratulations!
15:42 koha-jenkins Project Koha_Master_D10 build #401: FIXED in 47 min: https://jenkins.koha-community[…]a_Master_D10/401/
15:46 koha-jenkins Project Koha_Master_D10_Deps build #131: SUCCESS in 52 min: https://jenkins.koha-community[…]ter_D10_Deps/131/
16:02 ashimema done, done and done caroline :)
16:02 caroline thanks so much!
16:02 caroline ashimema++
16:02 caroline I'll merge as soon as I have a bit of time
16:03 caroline (or if I need a break from my tasks)
16:03 ashimema no worries :)
16:03 koha-jenkins Project Koha_Master_U16 build #123: SUCCESS in 57 min: https://jenkins.koha-community[…]a_Master_U16/123/
16:07 fridolin left #koha
16:17 koha-jenkins Project Koha_Master_D11 build #137: STILL UNSTABLE in 42 min: https://jenkins.koha-community[…]a_Master_D11/137/
16:27 margaret joined #koha
16:29 koha-jenkins Project Koha_Master_D10 build #402: SUCCESS in 46 min: https://jenkins.koha-community[…]a_Master_D10/402/
16:29 enkidu joined #koha
16:34 inlibro joined #koha
16:35 koha-jenkins Project Koha_Master_U2010 build #45: SUCCESS in 49 min: https://jenkins.koha-community[…]_Master_U2010/45/
16:46 JesseM joined #koha
17:20 enkidu joined #koha
17:24 cait1 left #koha
17:25 cait joined #koha
17:34 inlibro joined #koha
17:40 josef_moravec joined #koha
17:47 cait left #koha
18:01 cait1 joined #koha
18:02 cait1 left #koha
18:34 inlibro joined #koha
19:34 inlibro joined #koha
20:04 did joined #koha
20:09 rangi joined #koha
20:10 hayley joined #koha
20:11 alexbuckley joined #koha
20:12 cait joined #koha
20:14 aleisha joined #koha
20:15 aleisha hi all
20:32 aleisha excited that bug 19532 is signed off! thanks davidnind!
20:32 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=19532 new feature, P1 - high, ---, aleisha, Signed Off , Recalls for Koha
20:32 rangi yay!!!
20:35 inlibro joined #koha
20:45 tuxayo Hi o/
20:47 aleisha hi tuxayo
20:47 tuxayo About Recalls, I ran all the test suite and no issues :)
20:47 tuxayo Also I had to rebase and fix several conflicts. Is that ok for me to attach the rebased patches?
20:47 tcohen hola
20:49 tuxayo tcohen: congrats on managing to get rid of one of your too many community hats ^^
20:49 tuxayo (context: Andrew Fuerste-Henry from ByWater will RMaint oldstable)
20:50 tcohen thanks :-D
20:50 tcohen it is always refreshing to see new ones take on those roles!
21:04 aleisha tuxayo: yes probably fine to attach the rebased patches, or i can do it if you like. it has to get rebased literally all the time!
21:14 ashimema Andrew++
21:14 tcohen AndrewFH++
21:14 tcohen AndrewFH++
21:15 ashimema AndrewFH++
21:35 inlibro joined #koha
21:46 tuxayo aleisha: Thanks, I'll attach now, I just finished doublechecking :)
21:48 aleisha hey tuxayo youll need to re-attach all of the other patches following the rebased ones
21:48 aleisha otherwise they are in the wrong order
21:50 tuxayo aleisha: yes, I'm in the middle of fixing that.
21:50 tuxayo A syspref was in the wrong alphabetical order and I somehow duplicated it's commit with just the order change. So I interrupted the attachment.
21:50 aleisha cool :)
21:59 cait joined #koha
22:00 tuxayo aleisha: done ^^"
22:01 tuxayo aaah, I know the cause, I had to retry applying the patches 3 times due to git-bz hanging
22:02 tuxayo So the last time I forgot to cleanup before retrying.
22:35 inlibro joined #koha
22:52 AndrewFH joined #koha
23:07 tuxayo dcook around? I'm wondering how to test bug 26363. In KTD which patch should be tested? Source one or package one?
23:07 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=26363 enhancement, P5 - low, ---, dcook, Needs Signoff , Provide a systemd unit file for background_jobs_worker
23:07 dcook tuxayo: yup I'm around
23:07 tuxayo I don't know what does gittify do
23:07 dcook Let's see what I wrote..
23:08 tuxayo I would say package ^^
23:09 dcook Yeah I didn't do a great job writing my test plans there I guess
23:09 dcook So the first patch... that's just testing that the service gets rewritten correctly. I think that test plan should be OK
23:09 dcook As for the packaging one... I suppose a way to test it in KTD would be to...
23:10 dcook cp debian/templates/koha-worker@.service /etc/systemd/system/.
23:10 dcook (i'll post that on bugzilla)
23:10 dcook systemctl start koha-worker@kohadev.service
23:12 dcook I haven't tried that specifically so maybe I should..
23:12 koha-jenkins Project Koha_20.05_D9 build #114: SUCCESS in 40 min: https://jenkins.koha-community[…]oha_20.05_D9/114/
23:13 koha-jenkins Project Koha_20.05_D10 build #111: SUCCESS in 43 min: https://jenkins.koha-community[…]ha_20.05_D10/111/
23:13 dcook Ahh right.. it's a container
23:19 tuxayo thanks, I'm trying the 1st one now :)
23:19 dcook tuxayo: Yeah it wont' be testable in KTD - not without extensive work
23:20 tuxayo the 2nd?
23:20 wahanui the 2nd is the staff interface
23:20 dcook Yeah the 2nd
23:20 dcook The systemd one won't be testable that is
23:20 dcook Unless you have kohadevbox heh
23:20 tuxayo no wahanui
23:20 dcook Or some other linux machine
23:20 tuxayo Aren't they both systemd?
23:21 dcook Docker doesn't use an init system
23:21 dcook PID 1 is always whatever you run using Docker
23:22 dcook The container has a OS filesystem but all the different containers use the Docker host kernel
23:22 dcook And the Docker host is what coordinates the containers
23:22 dcook So they don't need an init system
23:22 dcook That's a terrible explanation heh
23:22 tuxayo it's ok, I get it
23:22 dcook The real init system is whatever is running on the Docker host
23:22 dcook The container's processes are just namespaced
23:23 dcook I think there are ways of getting systemd things to work in Docker, but I guess it might be considered an anti-pattern
23:23 dcook Since the ideal is to only have 1 process running per Docker container
23:23 dcook We cheat with KTD >_>
23:23 koha-jenkins Project Koha_20.05_D9_My8 build #111: SUCCESS in 51 min: https://jenkins.koha-community[…]20.05_D9_My8/111/
23:24 dcook In the real container world, the koha-worker would be running in its own container
23:24 tuxayo That means that the services don't use systemd ? (the ones when doing restart_all)
23:25 dcook If you do a "ps -efww" in KTD, you'll see that PID 1 is /bin/bash /kohadevbox/run.sh
23:25 dcook And things like apache2 are started up using the "service" command I think
23:25 tuxayo > We cheat with KTD >_>
23:25 tuxayo Indeed, even if it's 5 containers, it "should" be more
23:26 dcook Which reads Sys V init scripts
23:26 dcook runs*
23:26 dcook Yeah it should be significantly more
23:26 AndrewFH joined #koha
23:26 tuxayo ok so obviouly no Sys V in itself in docker either
23:26 dcook Correct
23:26 tuxayo we just run the scripts
23:26 dcook But since Sys V init scripts are just shell scripts really, it's trivial to run them still
23:26 dcook Yep
23:27 koha-jenkins Project Koha_20.05_U16 build #111: SUCCESS in 53 min: https://jenkins.koha-community[…]ha_20.05_U16/111/
23:28 dcook So if we look at /usr/sbin/service...
23:28 dcook It's also just a shell script
23:32 tuxayo great!
23:32 dcook When bug 26741 and related bugs are pushed, we could move rabbitmq out of koha_koha_1 and into a koha_rabbitmq_1 container too
23:32 huginn Bug http://bugs.koha-community.org[…]_bug.cgi?id=26741 minor, P5 - low, ---, dcook, Signed Off , Make rabbitmq-server suggested rather than required
23:33 dcook The main tough thing with koha-worker is that it needs access to Koha Perl libraries and database access
23:33 dcook But I do have thoughts on that..
23:33 tuxayo > PostgreSQL support is highly experimental.
23:33 tuxayo I didn't knew that there was an embryo of support :o
23:33 dcook Just not enough time to do all the things I want hehe
23:33 dcook Oh?
23:33 wahanui Oh are there instructions?
23:33 dcook Years ago, BibLibre were behind a push to use PostgreSQL I think
23:34 dcook I think I actually prefer PostgreSQL to MySQL/MariaDB at this point, so it's too bad they didn't succeed
23:34 tuxayo > But I do have thoughts on that..
23:34 tuxayo For putting it in another container?
23:34 dcook Yeah
23:34 dcook Or even a different container
23:34 dcook Just making it more distributable
23:34 dcook (which is the part of the point of using a message queue after all)
23:34 tuxayo For dev envs, is there a benefit
23:34 tuxayo > Years ago, BibLibre were behind a push to use PostgreSQL I think
23:34 tuxayo oooh :o I never heard of it ^^"
23:35 tuxayo *benefit?
23:35 dcook Well, I'd say the benefit of the dev env is that it's easy for us to test things before supporting production envs
23:35 dcook How long were you at BibLibre, tuxayo?
23:35 inlibro joined #koha
23:35 tuxayo 2 years
23:36 dcook Ahh ok
23:36 dcook Yeah this was quite a few years ago iirc
23:36 dcook In January, I will have been working with Koha for 9 years O_O
23:36 dcook In March, I will have been contributing to Koha for 9 years heh
23:37 dcook Nearly half the life of Koha which is crazy to think about..
23:37 dcook Feel like I've got the less hectic half though..
23:38 tuxayo he he ^^
23:39 AndrewFH joined #koha
23:39 dcook We don't use Koha in containers (yet), but I think if we did maybe I'd be more actively building up that ecosystem..
23:40 dcook At the moment, my attention seems to be on Mojolicious..
23:40 dcook although that's still in my own time >_>
23:40 tuxayo >_<
23:41 dcook I suppose the key is trying to get work-based projects using it..
23:41 dcook I feel like I'm so close..
23:41 tuxayo :D
23:41 dcook I think that I've more or less solved the AuthN/AuthZ challenges... and now it's mostly just the View (ie templates, internationalization)
23:41 dcook I can make it work with PSGI or CGI..
23:42 dcook I feel like too once I carve a path, other people can jump on it
23:42 dcook So even if I have to use my personal time to solve the View challenge..
23:42 tuxayo Just to be sure, is that about running Koha as Mojolicious app and be fully functional?
23:43 dcook And then maybe one sample controller..
23:43 dcook Mmm not exactly
23:43 dcook Rather, I'm planning to write Mojolicious controllers
23:43 dcook So instead of shoehorning Koha into Mojolicious, it would be Mojolicious native so to speka
23:43 dcook speak*
23:44 tuxayo controllers!=router right?
23:44 dcook I'm planning to replace export.pl with a Mojolicious controller
23:44 dcook Mmm not exactly
23:44 dcook There's usually a root controller which containers a router
23:44 dcook But then it can pass requests to other controllers
23:44 tuxayo ok!
23:45 dcook I'd say the controller is just about handling input/output (request/response)
23:45 tuxayo So not strictly about controllers
23:45 dcook So I have a "Koha::Staff" module
23:45 dcook Which gets mounted to "/intranet/staff" in plack.psgi
23:45 dcook So web requests are sent to the PSGI application mounted there
23:46 dcook And the router in that PSGI application passes requests to different controllers based on the URL
23:46 dcook You could just use 1 controller if you wanted to but that would be crazy because it would be too big
23:47 dcook How to do the controllers is debateable..
23:47 dcook At the moment, I'm creating something like... Koha::Staff::Controller::Tools::Export
23:47 dcook and then that controller has different functions which correspond to "actions"
23:48 tuxayo You take care about export.pl  because there is something wrong currently with background jobs right?
23:48 tuxayo *take care first
23:48 dcook Yeah, the PSGI app doesn't handle export.pl, because it tries to buffer the full response before sending it
23:48 dcook Which will never work for large responses
23:49 dcook And it does that because we use something like... Plack::App::CgiBin or something like that
23:49 tuxayo > correspond to "actions"
23:49 tuxayo Which are mapped to subroutines in a class i guess?
23:49 dcook Which tries to wrap the CGI with a PSGI friendly interface
23:49 dcook Yeah subroutines == functions
23:49 dcook And the "actions" are mapped to those
23:49 dcook And in this case controller == class
23:49 dcook Well controller == module I suppose
23:50 dcook It's been too long since I worked on this to remember all the little details heh
23:50 * dcook double checks
23:50 dcook Yeah it looks like class
23:50 dcook Inheriting from a base class of Mojolicious::Controller
23:51 dcook So replacing export.pl will be practical. It will mean we can use PSGI for the tools export functionality and won't have to use CGI for it anymore
23:51 dcook And then we can do that for all the other scripts that we're using CGI for
23:51 tuxayo It's all about moving toward MVC right? Or is it more complicated?
23:51 koha-jenkins Project Koha_20.05_U18 build #116: SUCCESS in 39 min: https://jenkins.koha-community[…]ha_20.05_U18/116/
23:51 dcook I'd say it's more moving towards PSGI
23:52 dcook I think a person could argue that Koha is already MVC in that we use the templates for the view, Perl scripts for the controller, and the Perl modules for the model
23:52 dcook Although... we embed way too much in the Perl scripts >_>
23:52 dcook And in the move to PSGI we're moving towards persistent processes
23:53 tuxayo > embed way too much in the Perl scripts
23:53 tuxayo No enough helper modules, it that the right way to say it?
23:53 dcook I wouldn't say helper modules..
23:53 dcook Rather, modules/libraries that contain the business logic or domain model
23:53 tuxayo buisiness modules?
23:53 tuxayo ok!
23:53 dcook We put too much business logic into the Perl scripts
23:54 dcook And we pass CGI objects all over the place which is *cringe*
23:54 tuxayo Right! The controller are the .pl scripts and the modules/libraries are the .pm
23:54 dcook That's how I think of them
23:54 dcook People could easily argue that's wrong
23:54 dcook But I think conceptually it works
23:55 dcook Since the Perl scripts are responsible for managing the STDIN and STDOUT/STDERR interfaces with the Apache webserver
23:55 dcook And here's the thing
23:55 dcook If we had all the business logic in the modules/libraries
23:55 dcook It would be trivial to implement PSGI controllers
23:56 dcook We'd just need to hook up the PSGI input/output to the modules and we'd be golden
23:56 dcook But because we have so much logic in the Perl scripts, we have to do a lot more coding
23:56 dcook You can think of the REST API as another set of controllers
23:56 dcook I mean we even talk about them that way
23:57 dcook Just take a look at Koha::REST::V1::Cities
23:57 dcook (Koha/REST/V1/Cities.pm)
23:57 tuxayo ok
23:58 dcook And Koha::Cities would be the model
23:58 dcook Koha::REST::V1::Cities just controls the input/output a particular way
23:58 koha-jenkins Project Koha_20.05_U20 build #150: SUCCESS in 45 min: https://jenkins.koha-community[…]ha_20.05_U20/150/
23:58 dcook In this case for a HTTP API
23:58 dcook But we could do Koha::Staff::Admin::Cities which would be a controller for HTML forms
23:59 dcook We could build a CLI tool for managing cities too
23:59 dcook And they would all use the same Koha::Cities model
23:59 dcook But they'd all handle inputs/outputs differently
23:59 tuxayo Shouldn't the logic (it's still reasonable) be in the model instead of the controller?
23:59 dcook Or *control* inputs/outputs differently
23:59 dcook Exactly

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