15:03:04 #startmeeting Koha Developer IRC meeting, 2 September 2014 - part 1 15:03:04 Meeting started Tue Sep 2 15:03:04 2014 UTC. The chair is cait. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:03:04 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:03:04 The meeting name has been set to 'koha_developer_irc_meeting__2_september_2014___part_1' 15:03:07 there are a bunch of 'extra' topic on the trello.. but I suggest we stick to the wiki agenda to start with. 15:03:08 welcome :) 15:03:13 #topic Introductions 15:03:13 #info wahanui, a bot that has become sentient 15:03:23 please introduce yourself like wahanui with #info 15:03:30 #info Katrin Fischer, BSZ, Germany, QAM 15:03:40 #info Martin Renvoize, PTFS-Europe, QA Team 15:03:41 #info Zeno Tajoli, CINECA (Italy) 15:03:44 #info Tomas Cohen Arazi, Universidad Nacional de Cordoba 15:03:50 #info Mark Tompsett, Canada 15:03:57 #info Colin Campbell, PTFS Europe, UK 15:04:09 #info Jonathan Druart, BibLibre, France 15:04:41 #topic RM 3.18 comments 15:04:53 #info D Ruth Bavousett, in and out, no one of any importance 15:04:54 handing over to tcohen :) 15:05:02 hi 15:05:02 bonjour, tcohen 15:05:12 hello 15:05:13 hello, atheia 15:05:28 i've started pushing new features that were sitting on the PQA queue 15:05:30 #info Owen Leonard, Athens County Public Libraries 15:05:35 notably SRU 15:06:03 #info Alex Sassmannshausen, PTFS Europe, UK 15:06:09 tcohen++ 15:06:10 i'd also like to mention that i've been working on retrieving facets from zebra 15:06:29 tcohen++ 15:06:31 which is already functional, but i'm missing unit tests for it 15:06:38 yay! 15:06:39 * ashimema has tested tcohens zebra faceting.. so far so good.. 15:06:46 ashimema: is it speedy? 15:06:57 what i've got so far is here: https://github.com/tomascohen/koha/tree/bug_11232_facets_from_zebra 15:07:16 (tcohen: arf, too bad, you didn't win beers :p) 15:07:17 once i have unit tests for it, i'll attach to bz 15:07:26 It 'feels' quicker.. but I've not yet done any proper measurements.. 15:07:31 Joubu: i have time, until thursday 15:07:37 :-D 15:07:44 why thursday? 15:07:52 It also feels like it's giving more relevant facets to me.. but i'm no librarian.. was going to ask a customer to take a look 15:07:53 * jcamins waves. 15:07:56 its a super secret bet i put 15:08:05 #info Jared Camins-Esakov, C & P Bibliography Services 15:08:22 ashimema++ too 15:09:07 i expect people related to Search code to step in 15:09:28 I will definitely test 11232 once you post it all up and good. :) 15:09:39 other thing that i'm worried about is the UTF-8 work, but we have space to discuss later 15:09:42 Because it will affect how we hide things. :) 15:09:45 bye 15:10:33 hm bit late but 15:10:39 #info agenda http://wiki.koha-community.org/wiki/Development_IRC_meeting,_2_September_2014 15:11:00 #info tcohen has working code for retrieving facets from zebra, will write unit tests until Thursday 15:11:21 hi jcamins 15:11:31 move on? 15:11:34 ok, moving on 15:11:37 ^^^^^ that 15:11:44 #topic Cleanup UTF-8 15:11:57 Joubu 15:12:06 hum? 15:12:23 what's the status? :) 15:12:24 I don't have something new to tell 15:12:25 what is the current status? 15:12:27 i guess the current status is "Failed QA" 15:12:34 ok 15:12:49 I can say that Paola Rossi and I we have finish to test it 15:12:49 tcohen: the status is needs signoff 15:12:58 #info Bug 11944 Cleanup Koha UTF-8 15:13:00 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11944 major, P5 - low, ---, jonathan.druart, Needs Signoff , Cleanup Koha UTF-8 15:13:01 #link bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11944 15:13:06 the result is on the wiki 15:13:17 #info wiki page: http://wiki.koha-community.org/wiki/Handling_UTF8_in_development 15:13:19 (still signoff by dpavlin and ztajoli) 15:13:34 http://wiki.koha-community.org/wiki/Handling_UTF8_in_development 15:13:45 ok, so RM comments on this 15:13:49 ztajoli: can you sign off or are there specific problems to be solved still? 15:14:32 We do the test using biblibre specific git branch. 15:14:42 morning 15:14:42 morning is a state of cat 15:15:03 the patches are working 15:15:08 We can sign-off with master for last test if is all OK 15:15:09 ztajoli: you said finished testing - is it signed off or failed qa? :) 15:15:14 involved people have extensibly tested the results 15:15:43 In my opinion is 'signed off' as result 15:15:46 can we have it signed off? 15:15:57 but it a wide change 15:16:08 gmcharlt: around? 15:16:09 tcohen: I am not it is relevant to add the SO status on all the patches 15:16:32 I maitain the branch, it's up-to-date. It's easier to maintain that the patches on the bug report 15:16:33 ... support call :( 15:17:47 The patches should be pushed to master quickly. If we wait 1month before the 3.18 release, it will be too late and too dangerous (too late to fix bugs). 15:17:55 So, I don't know what to do now 15:19:02 ztajoli: you can switch the status to signoff, don't update the patches. 15:19:11 has anyone thought on how to have regression tests for this? 15:19:22 that is my main concern 15:19:26 I was about to ask that tcohen.. 15:19:28 But to do QA ? 15:19:33 no, I don't know 15:19:42 i volunteered to do something about it, but didn't manage to have the time 15:19:55 do we need to introduce any further processes to qa.. or the coding guidelines to ensure we don't hit this issue again. 15:20:25 ashimema: we should add use cgi -utf8 15:20:36 easy to add to the qa script 15:20:38 OK 15:20:41 I think C4::Templates should be tested somehow 15:21:18 as far as testing is concerned ... what we need is testing via LWP::UserAgent or the equivalent 15:21:29 Joubu.. that was exactly the sort of answer i was hoping for ;) 15:21:44 that automates passing and receiving text and verifies that it comes back with appropriate coding 15:22:36 If adding a use cgi -utf8 be clear on the cases where you definitely do not want to use that 15:23:03 gmcharlt: that needs apache running, right? 15:23:13 yeah, given that we have several places where binary input needs to be uploaded 15:23:34 tcohen: yes 15:23:47 * mtompset steps out for 1.5 hours. 15:23:48 that could be managed either with a self-contained config 15:23:56 gmcharlt: should those tests be prepared to be run on jenkins? 15:24:24 or, for dev machines, perhaps by taking advantage of multiple-Apache-instance support on Deb/Ubuntu 15:24:26 tcohen: IMO, yes 15:25:02 so we need to test plain templates, templates using template plugins 15:25:13 pages retireving binary data 15:25:16 what else? 15:25:59 certain function tests 15:26:17 - searching patrons with/without diacritics 15:26:24 - search patrons with both NFC and NFD forms 15:26:38 - collation in SQL query results 15:27:29 Unit tests are requested for 2 months and nobody wrote them. So I am not sure we will find someone to write them. 15:27:40 silly question.. if we're doing the big db shift to utf8_unicode_ci.. should we not make the jump to utf8mb8_unicode_ci instead? 15:27:55 to inlcude the entire character set.. not mysql's abridged one? 15:28:52 ashimema: utf8mb8? 15:29:03 ashimema: not a silly question; at first glance, doing that seems sensible 15:29:08 Joubu: http://dev.mysql.com/doc/refman/5.5/en/charset-unicode-utf8mb4.html 15:29:09 Joubu: https://mathiasbynens.be/notes/mysql-utf8mb4 15:29:40 do we need to test through apache our cgi scripts? can't we just use a sensible IO layer ? 15:29:43 in short.. when mysql implimented utf8 first time around.. they got it worng and only included like half the character set.. 15:29:48 mb4 is the solution to that. 15:30:03 Joubu: I will write tests; I realize I have been saying that all summer, but I /do/ want to get the improvements into 3.18 and will work to make that happen 15:30:10 tcohen: we need full-stack testing, IMO 15:30:21 ok 15:30:36 * tcohen volunteers to help gmcharlt 15:30:54 gmcharlt: Are you sure we need UT before pushing the patches? 15:31:09 if we need UT to catch regression, we can wait for UT 15:31:32 if the changes are pushed, all developers will catch errors 15:33:48 #info gmcharlt & tcohen are going to write some tests for utf8 regressions 15:34:19 caits on the phone so isn't moving the meeting forward at the minute.. 15:34:28 sorry had a call to answer - do i have to do something? reading back as fast as I can 15:34:35 is there anything to agree on regarding the above then.. or shall we move on. 15:34:38 i'll resign my beers jcamins, to have some time for the utf8 unit tests 15:34:39 cait your back.. 15:34:45 * ashimema hans back the reighns 15:35:21 So I put the bug 11944 as 'Signed OFF' 15:35:23 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11944 major, P5 - low, ---, jonathan.druart, Needs Signoff , Cleanup Koha UTF-8 15:35:38 QA based on biblibre branch 15:35:51 Joubu: tcohen and I are in a sidebar conversation about timing, one moment 15:36:03 :) 15:36:07 If it ok, insert in master to test 15:37:24 #action gmcharlt will write unit tests 15:37:33 we will set a deadline of 9/19 for this work 15:38:31 ok? 15:38:36 Joubu: ? 15:38:59 tcohen: You are the RM :) 15:39:03 9/19? 15:39:04 0.473684210526316 15:39:07 september 19 15:39:10 19/09/2014 15:39:21 sounds good to me tcohen.. 15:39:27 the deadline is for unit tests? 15:39:30 info it cait.. then they can't forget ;) 15:39:32 cait: yes 15:39:40 #info deadline for utf-8 unit tests is 19 September 15:39:43 In same time anyone do generic QA on the patch ? 15:40:23 I don't have the time currently - probably not much time the next 2 weeks 15:40:26 or a bit longer 15:40:33 I'll take a look at qa on it.. to start with anyways.. 15:40:53 I'm quietly coersing cait to collaborate on qa for it in another window ;) 15:41:08 #action ashimema will start QA on it 15:41:12 just me then.. 15:41:31 * cait doesn't see another window 15:41:38 :P 15:41:49 #action cait will try to help, but won't have much time the next 2 weeks or a bit longer 15:41:55 the misses won't like me much.. think it's gonna have to reside as an evenings job.. but certainly important one to get in. 15:42:09 * cait thinks we should send the misses cookies 15:42:09 shall we move on 15:42:31 do we have something for the coding guidelines? 15:42:43 #topic Additions to coding guidelines 15:42:55 2 spaces for tt files :-P 15:43:02 * cait hasn't seen that 15:43:24 if noone has a proposition, can we move on to the next topic? ;) 15:43:43 move on 15:43:57 #topic Bugs 15:44:17 ashimema: problems can be on search results (facets, resort, etc.) 15:44:38 anyone wants to talk about his/her favourite bug? :) 15:44:45 Cheers Joubu.. i'll pay special attention to those 15:44:49 you can also do so in the next topic :) 15:44:52 i'd like to mention bug 12788, which solves a regression on 3.16 15:44:53 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12788 normal, P5 - low, ---, tomascohen, Signed Off , Refine search displays 4xx fields from authority records 15:44:54 cait: certainly. 15:44:54 certainly is confusing.. thanks for bringing ashimema's attention to it. 15:45:13 We should fix the issue about marc mod template 15:45:26 but I am asking for 10 months 15:45:39 * Joubu stops to complain 15:45:40 #info bug 12788 SO - Refine search displays 4xx fields from authority records 15:45:50 which bug is that Joubu 15:45:51 i think marcelr started but then got stuck somewhere 15:45:54 what's the current status? 15:45:55 well, the current status is "Failed QA" 15:46:01 cait: NQA 15:46:44 cait: you wanted to promote bug 10212 15:46:44 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10212 new feature, P5 - low, ---, jonathan.druart, Pushed to Master , Columns configuration for tables 15:46:57 heh yes 15:47:23 i.e. lets get developers using the column configuration stuff more widely 15:47:26 I sent patches for bugs in architecture/plumbing. The need signoff. 15:47:26 Joubu: coudl you send me a reminder about the bug numbers? or leave a comment on the first one? 15:47:27 not sure how to achieve this 15:48:03 cait: 11395 and others 15:48:06 i think letting people knwo it's there is a first step 15:48:46 ashimema: I will try to integrate the columns conf for tables in the acq module on rewriting the tax stuff. 15:48:47 Joubu: i made a note 15:48:57 hmm.. maybe add reference to it on the datatables wiki page. 15:49:07 :) yeay Joubu 15:49:10 ashimema: that a good idea 15:49:17 I'll do that cait. 15:49:18 a note to the devle list maybe too? 15:49:35 i think it will only be useful when we start using it for other tables besides currencies :) 15:50:00 acq will be great 15:50:01 http://wiki.koha-community.org/wiki/DataTables_HowTo is the page I'm going to add to 15:50:33 any dbic experts care to weight in on bug 11518 15:50:34 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11518 enhancement, P5 - low, ---, kyle, Needs Signoff , Add new method to Koha::Schema::Result::Item that will always return the correct itemtype 15:50:36 #action ashimema is going to add a note about Columns configuration (bug 10212) to the wiki (http://wiki.koha-community.org/wiki/DataTables_HowTo) 15:50:47 I need feedbacks on bug 12826 15:50:48 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12826 enhancement, P5 - low, ---, jonathan.druart, ASSIGNED , GST / VAT rewrite - plumbing 15:50:49 the last followup patches.. 15:51:14 the also affect bug 12603 I beleive 15:51:15 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12603 enhancement, P5 - low, ---, yohann.dufour, Signed Off , TestBuilder - Module to simplify the writing of tests 15:52:23 Something else, I am going to wake bug 11926 up 15:52:24 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11926 enhancement, P5 - low, ---, jonathan.druart, ASSIGNED , Render community koha statistic usages 15:52:53 ok, i am getting a bit lost about teh #info s - please add one yourself for those i missed 15:52:54 I would like to know if some companies/libraries is candidate for testing it 15:53:46 #info Feedback needed on bugs 11518, 12826 and 12603 15:54:55 I'm keen on 11926 Joubu.. but would need to discus with some customers/colleagues first regarding data collection policies 15:55:24 ashimema: It can be anonymous 15:55:56 I plan to test GST / VAT. 15:55:57 Indeed.. but I still need to check that it's ok to collect such data at all.. 15:56:05 ashimema: the data will be pref values, library name (or not), volumetry (patrons, records) 15:56:17 As I see the first to sign-off is 12827 15:56:20 ashimema: yes, of course :) 15:56:28 :) 15:57:08 Generaly speaking we can work on 'sign-off' 15:57:16 cait.. I think you also wnated to raise a question about bug 11625 adn bug 11747 15:57:18 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11625 enhancement, P5 - low, ---, kyle, In Discussion , Default to logged in library for circ rules and notices & slips 15:57:19 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11747 enhancement, P5 - low, ---, kyle, In Discussion , Default to logged in library for Overdue notice/status triggers 15:57:37 ashimema: maybe better for the next meeting :) 15:57:37 regarding 'default select branch' and how it should be handled? 15:58:25 ztajoli: does paola have a nickname? 15:58:30 ztajoli: Actually it's the most important bottleneck! 15:58:33 no 15:58:38 ztajoli++, paola++ :) 15:58:55 yep sign offs are quite important 15:59:00 please keep up the good work 15:59:01 ztajoli: Paole does a wonderful job on signoff 15:59:02 serverside handling of datatables.. 15:59:03 Paole doesn't like to be present on chat 15:59:15 ztajoli: please tell her we love her work :) 15:59:18 So I speak also for here 15:59:23 (Paola, sorry) 15:59:34 tcohen, Joubu.. do either of you have any thoughts on standardising where code should live? 15:59:34 No problem if you write her 16:00:03 ashimema: elaborate? 16:00:16 ashimema: yes please :) 16:00:36 currently but 9811 puts the backend code for the datatables ajax code in one place.. and bug 11703 puts it in a completely different place.. 16:00:38 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11703 enhancement, P5 - low, ---, kyle, Pushed to Master , Convert checkouts table to ajax datatable 16:00:55 it's hard to qa such a thing withough a rule to refer to in the guidelines somewhere. 16:01:16 C4/Utils/Datatables and /srv are the two places so far i've spotted this issue. 16:01:37 ashimema: C4::Utils::Datatables uses a script in svc 16:01:48 (we have similar problems for the api for instance).. but I thought I'd try and grab the datatables one before either plce become to prevatlent. 16:01:50 svc/members/search 16:01:51 all business logic should be packaged and testable 16:02:39 endpoint/routing code should be clean and unimportant in terms of complexity 16:04:11 I think maybe save that debate for the next meeting.. 16:04:23 * cait tends to agree 16:04:35 ashimema: does your question only about DT ? 16:04:41 There's a bunch of stuff still to discuss in the dev meeting trello board.. 16:04:48 in this case it was only about DT.. 16:05:00 ok, i have to leave 16:05:02 but there's the more general cases out there.. 16:05:07 i think we need to wrap up for tody 16:05:11 tcohen: when for the next meeting? 16:05:22 there's a lot to talk about in the pipe 16:05:28 next week 16:05:28 hmmm... next week is project week 16:05:37 forget next week 16:05:39 cait: I forgot next week 16:05:42 same time next week ;) 16:05:44 the next otherwise 16:05:45 ok... same day? 16:05:49 * tcohen agrees 16:05:52 same day/time 16:06:04 #topic Agree on time for next meeting 16:06:19 that would be september 9th, 15 + 22 16:06:21 UTC 16:06:32 +1 16:06:43 +1 16:07:01 #agreed Next meeting will be (given agreement from second meeting today) - September 9th, 15 + 22 UTC 16:07:05 #endmeeting