IRC log for #koha, 2017-02-20

All times shown according to UTC.

Time S Nick Message
00:28 dcook joined #koha
01:34 Francesca_ joined #koha
02:04 z0k_ joined #koha
02:35 stozza joined #koha
03:00 * dcook realizes he might have a message waiting...
03:00 dcook Guess not
03:08 z0k_ joined #koha
03:11 stozza joined #koha
03:46 stozza joined #koha
04:03 BobB joined #koha
04:21 Francesca joined #koha
04:36 stozza joined #koha
04:42 edveal1 joined #koha
05:12 stozza joined #koha
05:41 jamesb_ joined #koha
05:53 Francesca joined #koha
06:03 Francesca joined #koha
06:06 stozza joined #koha
06:11 cait joined #koha
06:37 stozza joined #koha
06:39 fridolin joined #koha
07:12 reiveune joined #koha
07:12 reiveune hello
07:17 sophie_m joined #koha
07:21 fridolin hie there
07:22 Francesca joined #koha
07:26 * magnuse waves
07:32 Francesca joined #koha
07:32 * cait waves
07:35 magnuse guten morgen cait
07:45 josef_moravec joined #koha
07:46 marcelr joined #koha
07:47 laurence joined #koha
07:47 marcelr hi #koha
07:47 josef_moravec morning #koha;
07:47 josef_moravec hi marcelr
07:48 marcelr morning josef_moravec
07:49 oha joined #koha
07:56 LibraryClaire joined #koha
08:02 LibraryClaire morning #koha
08:08 baptiste joined #koha
08:11 marcelr hi LibraryClaire
08:12 LibraryClaire hey marcelr
08:21 cait joined #koha
08:26 AndrewIsh joined #koha
08:27 paul_p joined #koha
08:30 Francesca joined #koha
08:54 cait Joubu++
08:54 cait thx for the QA
08:56 Joubu cait: One I cannot QA is bug 18111 and I think you should have it for the next 16.11 release
08:56 huginn Bug[…]_bug.cgi?id=18111 major, P5 - low, ---, jonathan.druart, Signed Off , Import default framework is broken
08:58 cait no string change, so that's good
08:58 cait maybe we can ask tcohen or kidclamp to take a look?
08:58 cait bbiab
08:59 AndrewIsh joined #koha
09:18 biblibre_ joined #koha
09:18 biblibre__ joined #koha
09:37 eythian @wunder ams
09:37 huginn eythian: The current temperature in Amsterdam, Netherlands is 9.0°C (10:25 AM CET on February 20, 2017). Conditions: Light Drizzle. Humidity: 93%. Dew Point: 8.0°C. Pressure: 29.98 in 1015 hPa (Steady).
09:44 eythian[…]nship-handwriting
09:51 stozza joined #koha
09:52 irma joined #koha
09:53 deb-CSPL joined #koha
09:59 irma joined #koha
10:01 Joubu baptiste: do not forget to update the status of the patch when you fix a fail QA
10:25 Francesca joined #koha
10:26 stozza joined #koha
11:08 Scott-CSPL joined #koha
11:19 z0k_ joined #koha
11:34 oha regarding bug 17427, I was wondering: how many people uses plack as a backend, and how is not using it?
11:34 huginn Bug[…]_bug.cgi?id=17427 normal, P5 - low, ---, jonathan.druart, Failed QA , Replace CGI::Session with Data::Session
11:35 marcelr oha: you should use plack in the newer koha versions; recommended
11:35 benjamin joined #koha
11:36 oha yes of course
11:36 oha what i meant was, is there anyone NOT using plack out there?
11:36 marcelr yeah there will
11:36 oha will?
11:36 marcelr be
11:37 oha would be so easier to handle sessions within plack, is there any way to push them to jump in the 21th?
11:38 marcelr yeah why still use /tmp for session storage
11:38 marcelr i was thinking that Koha::Session calling Koha::Cache might also be an easy solution
11:38 oha marcelr: i'm not sure i follow
11:39 marcelr oha: you can still use tmp for sessions
11:39 oha what i'm suggesting is to handle all the session/auth at once in plack.
11:39 marcelr yes iam too
11:40 marcelr just speaking another language :)
11:41 oha so, is there a way to make everyone use plack? :)
11:42 marcelr write a great patch
11:42 marcelr that no one can resist
11:44 oha sounds like a good plan!
11:51 cait it's not so long ago that we started recommending it
11:51 cait so people might need some more pushing
11:51 cait to get word around
11:56 cait jajm++ :)
11:59 oha cait: i see. well, i'm looking at the session handling, i might be able to find some nice fixes on the plack side
11:59 cait oha: there are also remaining issues that will not allow you to use plack
11:59 cait oha: for example atm shibboleth doesn't work
12:00 * ashimema is on the case for that as soon as I've got an OAuth patch finished
12:00 cait and - I think - that setting sys prefs in apache doesn't - which is is needed to build seaparate opacs in some consortia settings
12:00 cait so there are some blockers
12:00 oha cait: i have an idea, but i need some time for some testing, which might make plack way easier to handle, and still have apache working, but a bit uglier
12:00 ashimema that second one is solved I believe cait
12:01 cait i can tell you what's on bugzilla - for technical details i will have to refer to the 'more dev' people :)
12:01 marcelr more dev?
12:01 cait ashimema: yeah not entirely sure about that one, but i think someone on the mailing list said recently it still didn't work
12:01 cait more dev than me :)
12:01 oha basically, instead of having old school cgi and plack "adapt" to use them. we could have the other way around. proper web-app used directly by plack, and cgi-wrappers for apache
12:11 cait oha: sorry, that's the kind of discussion you will need to have with someone else :)
12:13 oha ehehhe, i'm more thinking out loud here
12:19 Joubu oha: sounds like you forgot to update the status for bug 17941
12:19 huginn Bug[…]_bug.cgi?id=17941 major, P5 - low, ---, oha, In Discussion , CanBookBeRenewed is very inefficient/slow
12:19 Joubu switch it to NSO if it's ready
12:20 Joubu Needs Signoff
12:20 wahanui rumour has it Needs Signoff is ridiculously high too
12:21 oha Joubu: ohhhh i'm so dumb. switched it
12:21 oha switched it now*
12:23 Joubu oha: Any ideas how big it would be to switch to a proper web app using PSGI?
12:23 francharb joined #koha
12:24 francharb Good morning #koha
12:25 oha Joubu: the psgi part is already there. what i find a bit clumsy is to handle auth and session in each script (and note that i found at least 2 different implementation for handling authentication, e.g. svc/checkout is differen than reserver/request)
12:25 oha what i think i will try is to move some of those scripts into functions, then see how it goes.
12:29 Joubu oha: Most of the time auth check is done by get_template_and_user, which does much more than just auth checking
12:29 Joubu it's used when we need templating basically
12:29 oha yup
12:29 Joubu script in svc are for AJAX script, which returns json usually
12:29 oha but it honors REMOTE_USER also, which isn't happening in svc scripts
12:29 Joubu oha: code from C4::Authis terrible, lot of code is dup
12:29 oha exactly
12:30 Joubu and it's hard to find someone who knows all the loggin possibilities
12:30 oha but at the same time, i'd like to have a common session handling and auth in plack
12:30 oha so my thinking was to move that to plack, BEFORE the actual code is run
12:30 oha maybe simplify it there
12:31 oha and keep the rest as legacy in the cgi scripts
12:31 Joubu I think ashimema is your guy if you want to talk about auth methods, like shib, oauth or ldap
12:31 ashimema in a call at the minute
12:31 ashimema but happy to talk after
12:31 oha great
12:32 Joubu oha: why did you take a look at bug 17427?
12:32 huginn Bug[…]_bug.cgi?id=17427 normal, P5 - low, ---, jonathan.druart, Failed QA , Replace CGI::Session with Data::Session
12:32 Joubu just curious :)
12:32 oha benjamin told me about it
12:33 oha for the record, i've started working here at deichman with benjamin and petter
12:33 Joubu oha: I will be happy to help you if you need any help in this area :)
12:34 oha that's great
12:34 Joubu but you need to know that you are attacking Koha to one of the hardest side
12:34 oha i surely will need help with all the possibile combinations
12:34 Joubu the other one is C4::Search
12:44 fridolin joined #koha
12:48 eythian I always wanted to rebuild C4::Auth into a nice clean pluggable system.
12:49 eythian but it'd be a hyoooge job.
12:52 marcelr thx Joubu # 18037 (the horrible part left out )
13:16 tcohen joined #koha
13:16 tcohen morning
13:19 LibraryClaire hi tcohen
13:19 magnuse kia ora tcohen
13:38 Dyrcona joined #koha
13:41 ashimema a few of us have wanted to rebuild C4::Auth
13:41 * tcohen reads
13:42 ashimema but as eythian says.. it's a huge job so none of us have ;)
13:42 tcohen kivilahtio did it ;)
13:42 kivilahtio ashimema: yup
13:43 kivilahtio tcohen: I am not sure should I port it to the newest master or let it rot
13:43 tcohen you should communicate with ashimema
13:43 kivilahtio tcohen: I guess I will port it
13:43 tcohen because the problem is no one dares to QA it
13:43 kivilahtio tcohen: we need to do some auth extensions so it is much easier to work from my rewrite
13:43 ashimema it would need a very staggered approach to get in I think... kivi's did it all of it at once.. undoubtedly the easiest to write.. but the hardest to test ;)
13:44 kivilahtio ashimema: heh. There are PageObject tests to test it :)
13:44 tcohen exactly, what ashimema says
13:44 kivilahtio search_history and login logout
13:44 kivilahtio that's another monster
13:44 andreashm joined #koha
13:44 ashimema has it's fingers everywhere..
13:44 kivilahtio ashimema: I carefulyl deprecated it
13:44 marcelr nice typo
13:44 kivilahtio ashimema: my rewite simply deprecates it, not breaks it completely
13:45 kivilahtio so you can enjoy your unbelievably unreadable auth code
13:45 kivilahtio in places where my rewrite doesn'ät replace it
13:45 ashimema anywho.. it's still on my list to fully test.. but I think it's going to take a few weeks to fully comprehend it.. so hasn't happened yet :(
13:45 kivilahtio ashimema: nice to hear that
13:46 kivilahtio ashimema: I will port it to master then
13:46 * ashimema hates the unreadable auth code.. but he doesn't mostly understand it ;)
13:46 ashimema beware I'm not all that likely to get time to look at it all that soon..
13:46 ashimema but yeah.. it's still on the list.. it's a long list though :(
13:46 kivilahtio ashimema: np
13:46 marcelr ashimema++
13:46 kivilahtio ashimema: I donät mind rebaseing it since the C4::Auth never changes
13:46 ashimema :)
13:47 ashimema haha.. yeah.. C4::Auth scares everyone.. so it changes infrequently now ;)
13:47 kivilahtio ashimema: be aware that the permission system is rewritten for us as well
13:47 kivilahtio so it is not just the auth
13:47 marcelr there you go
13:47 ashimema I would personally submit a set of patches that depend upon each other..
13:48 ashimema one for authentication, one for authorization and permissions handling at the very least
13:48 francharb joined #koha
13:48 kivilahtio already separated it like that
13:48 ashimema but yeah.. no time right now :(
13:48 kivilahtio there is a Koha::Auth::PermissionManager which is ridiculously slow
13:48 kivilahtio but has a neat API to CRUS permissions and modules and user_permissions
13:48 ashimema CRUS?
13:48 kivilahtio easy to add a REST enpoint on top if it
13:48 kivilahtio ashimema: CRUD
13:48 ashimema k
13:48 kivilahtio sorry
13:50 * andreashm waves
13:51 kmlussier joined #koha
13:51 nengard joined #koha
13:54 tcohen joined #koha
13:55 tcohen hi
13:55 marcelr hi tcohen
13:56 rsantellan joined #koha
13:56 rsantellan good morning #koha
13:57 tcohen hi rsantellan
13:57 andreashm hi tcohen, rsantellan
13:58 andreashm speaking of auth... or developers noticed that they can do some stuff with the rest api without the cookie (holds, for example). Any one else experience this?
13:59 tcohen andreashm: i haven't
13:59 andreashm good, might just be our setup then and not a major problem
14:00 tcohen andreashm: I have experienced using kohadevbox
14:00 tcohen that if you use localhost:8080 and localhost:8081 to access it
14:00 tcohen the session is shared, so you think you don't have a cookie, but you do
14:01 Joubu baptiste: did you test and sign off on bug 8548? You changed the status but no signoff patches were attached
14:01 huginn Bug[…]w_bug.cgi?id=8548 normal, P5 - low, ---, aleishaamohia, Needs Signoff , Add callnumber sort option to overdue report
14:01 andreashm tcohen: ah, interesting. the developers are working against a different server, but it might be something similar.
14:01 tcohen if you use a FQDN
14:01 tcohen it doesn't happen
14:01 tcohen it is just that we match 'localhost' only
14:02 andreashm ok
14:03 Joubu andreashm: see issue 121 of kohadevbox
14:04 tcohen hi Joubu :-D
14:04 Joubu hi tcohen! :)
14:05 baptiste Joubu, I committed but didn't attach, I've just attached it and switch again to signed off ;)
14:06 Joubu baptiste: ok thanks
14:06 * cait waves
14:06 tcohen koha sucks
14:06 tcohen was kicked by Joubu: nope, only C4::Auth
14:07 tcohen joined #koha
14:07 Joubu ;)
14:07 tcohen :-D
14:07 tcohen so you can now kick me out
14:07 ashimema lol
14:07 tcohen Joubu got new superpowers
14:07 * Joubu was trying his new power
14:07 marcelr well
14:07 marcelr might not be so a good idea ;)
14:08 Joubu not at all, I tend to kick people who disagree with me
14:08 Joubu I did not ask for anything!
14:08 * LibraryClaire sends Joubu beer
14:08 * cait hides
14:09 * tcohen stocks belgium beer just in case needs to brive Joubu
14:09 marcelr ooo
14:10 * andreashm can offer mediocre swiss beer
14:16 Joubu QAers, one of you should QA bug 18111, cait needs it ASAP to get it for the next release
14:16 huginn Bug[…]_bug.cgi?id=18111 major, P5 - low, ---, jonathan.druart, Signed Off , Import default framework is broken
14:17 tcohen ahhh
14:17 tcohen I just hit that one :-D
14:17 Joubu I think bug 18026 is a good one to backport as well
14:17 huginn Bug[…]_bug.cgi?id=18026 normal, P4, ---, jonathan.druart, Needs Signoff , URL to database columns link in system preferences is incorrect
14:21 francharb joined #koha
14:23 NateC joined #koha
14:26 cait i already got thatone
14:26 cait i think?
14:27 Joubu cait: nope
14:27 cait hm
14:27 cait is had the hardcoded one
14:27 cait i might skip it - to avoid translators having to translate it again
14:28 tcohen cait: maybe it is better havnig it not translated
14:30 cait ?
14:31 Joubu Yes otherwise the link is broken
14:32 cait now confused
14:38 tcohen tradeof
14:38 tcohen not translatged vs. non functional?
14:42 magnuse where are we now with grs vs dom indexing? do we just ignore record.abs if we want to change some indexing?
14:42 tcohen yeap
14:44 magnuse grs is deprecated, waiting for someone to find the energy to remove it?
14:53 * druthb notes that magnuse looks particularly energetic today.
14:53 magnuse lulz
14:53 magnuse druthb must be looking at the wring guy
14:53 magnuse *wrong
14:54 druthb :)
14:57 tcohen magnuse: yes
14:58 tcohen if you ask on the mailing list you will get a response from bad paul stating he still uses it, please don't delete
14:58 agoforth joined #koha
14:58 tcohen so don't ask, again
14:58 tcohen :-P
14:59 magnuse well, it would hardly get backported to where he is...
14:59 * magnuse wanders off to make dinner
14:59 tcohen hehe
15:04 Joubu tcohen:[…]ruary/041995.html
15:04 Joubu never got an answer
15:05 * tcohen should read his memory pill more often
15:05 Joubu I have a local branch called bug_14302 with a start, I can finish it and submit it if there is testers intested around needed
15:05 Joubu s/needed//
15:06 tcohen cool
15:12 stozza joined #koha
15:34 Joubu Does anybody use rewrite-config.PL?
15:36 stozza is it possible to use koha without barcodes? is there a way to do the checkout just looking for the book's title? in our school there will hardly be more than one copy per book, and the barcodes are going to be an issue
15:38 Joubu stozza: nope
15:40 eythian stozza: that's very likely to run into bad and confusing issues.
15:46 Joubu rewrite-config is used from makefile actually..
15:46 stozza ok I guess we'll just put a label and use a pen to write down a unique numeric code on each book
15:47 stozza that should work, does not have to be an actual barcode as far as I understand
15:48 eythian that's correct, however it might be easier to just use barcodes.
15:51 stozza I agree, probably overkill at this stage for us
15:53 eythian I think you can just buy sheets of barcode labels. I don't imagine they're expensive.
15:53 lds joined #koha
15:53 LibraryClaire left #koha
15:54 lds hello
15:54 wahanui bonjour, lds
16:09 edveal joined #koha
16:12 stozza_ joined #koha
16:20 fridolin left #koha
16:45 laurence left #koha
16:53 kmlussier joined #koha
17:13 stozza_ joined #koha
17:16 magnuse stozza stozza_:  have a look at[…]odeFallbackSearch
17:17 magnuse circ without barcodes
17:17 magnuse yay for bug 14302
17:17 huginn Bug[…]_bug.cgi?id=14302 enhancement, P5 - low, ---, jonathan.druart, Needs Signoff , Remove GRS1 indexing related code
17:19 reiveune bye
17:24 reiveune bye
17:24 reiveune left #koha
18:08 kmlussier joined #koha
18:09 kmlussier joined #koha
18:21 cait left #koha
18:27 stozza joined #koha
18:31 LibraryClaire joined #koha
18:48 cait joined #koha
19:22 mtj hey tcohen, about for a quicky?
19:22 tcohen hey
19:22 tcohen leaving
19:22 tcohen quick?
19:23 mtj hmm, yeah, i pm you
19:23 stozza joined #koha
19:25 tcohen bye!
19:34 kidclamp joined #koha
20:02 NateC joined #koha
20:19 cait bgkriegel++
20:19 cait bgkriegel++
20:19 cait bgkriegel++
20:20 druthb bgkriegel++
20:25 wizzycray hi
20:25 wizzyrea hi
20:25 wahanui kia ora, wizzyrea
20:26 wizzyrea it was kinda fun to be wizzycray
20:26 wizzyrea but I get messages on wizzyrea so :(
20:28 * druthb waves to wizzyrea.
20:28 wizzyrea heya
20:33 stozza joined #koha
21:27 stozza joined #koha
21:46 stozza joined #koha
22:07 irma joined #koha
22:17 agoforth joined #koha
22:21 Francesca joined #koha
22:27 Francesca joined #koha
22:40 stozza joined #koha
22:51 dcook @later tell wizzycray Maybe you get messages too!
22:51 huginn dcook: The operation succeeded.
22:51 wizzyrea keke
22:51 dcook Guest1218 is an op, eh? :p
22:52 wizzyrea that's old wizzyrea
22:52 dcook So many wizzys!
23:07 wizzyrea there, all is right in the world again
23:09 wizzyrea joined #koha
23:09 wizzyrea Yes. that will do.
23:21 * dcook celebrates
23:21 dcook Ooh I have chocolate too...
23:21 dcook somewhere..
23:34 stozza joined #koha
23:36 papa joined #koha
23:49 Francesca joined #koha
23:55 Francesca joined #koha

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