13:06:05 #startmeeting Development IRC meeting 25 January 2017 13:06:05 Meeting started Wed Jan 25 13:06:05 2017 UTC. The chair is tcohen. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:06:05 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:06:05 The meeting name has been set to 'development_irc_meeting_25_january_2017' 13:06:30 #info Marcel de Rooy, Rijksmuseum, Netherlands 13:06:30 hi everyone, please introduce yourselves with #info 13:06:36 #info Magnus Enger, Libriotech, Norway 13:06:43 #info Tomás Cohen Arazi, Theke Solutions 13:06:47 #info Katrin Fischer, BSZ, Germany 13:06:50 #info Julian Maurice, BibLibre, France 13:06:51 #info Claire Gravely, BSZ, Germany 13:06:58 #info Rodrigo Santellan, ORT, Uruguay 13:06:59 #info Owen Leonard, Athens County Public Libraries, USA 13:07:02 #info Nick Clemens, ByWater Solutions 13:07:28 #info Josef Moravec, Municipal Library Usti nad ORlici 13:08:04 #info Jonathan Druart, France 13:08:36 #koha, please introduce yourselves 13:08:54 #topic Announcements 13:09:00 Any announcements? 13:09:49 druthb: announcements? 13:09:50 announcements are often automatic 13:10:18 the rdf mailing list, perhaps` 13:10:32 magnuse: go ahead 13:10:39 [off] I will move to Argentina in about 1 month, hopefully 13:10:47 [off] but keep it off :p 13:10:50 [off] woohoo 13:10:54 [off] !!! 13:10:57 #info Martin Renvoize, PTFS-Europe 13:11:00 [off] :O 13:11:03 [off] working on Koha? 13:11:06 bywater has created a mailing list for people who are interested in adding support for rdf/linked data/semantic web techniques to koha 13:11:07 [off] That's quite an announcement Joubu 13:11:16 [off] I'm getting married on March 4. :) 13:11:29 [off] congratulations :) 13:11:32 [off] :) 13:11:35 want to link that magnuse? 13:11:40 [off] :) 13:11:50 [off] just following my girlfriend, she got her funding for a 2 years position research 13:11:53 [off] congatulations to all ;) 13:11:54 [off] congrats druthb 13:12:04 [off] enjoy Joubu 13:12:06 don't know if there is a public page for signing up to the list, but get in touch with bag or khall or others at bywater and they can propbably add you 13:12:09 [off] nothing will change for my Koha-related job 13:12:18 [off] good 13:12:18 [off] congrats to K. for getting the funding :) 13:12:38 [off] hooray for druthb and Joubu! 13:12:42 [off] apart from your online hours Joubu ;) 13:13:01 [off] hehe yes indeed :) 13:13:02 #info ByWater has created an mailing list for people interested in adding support for rdf/linked data/semantic web techniques to koha 13:13:03 [off] lots of [off] and not much [on] 13:13:16 [off]congrats to all! 13:13:17 [on] tcohen will can move on ;) 13:13:38 Do we have an agenda link? 13:13:49 #info Lots of off-the-record announcements, for more information ask Joubu and druthb 13:13:49 https://wiki.koha-community.org/wiki/Development_IRC_meeting_25_January_2017 13:13:57 #link https://wiki.koha-community.org/wiki/Development_IRC_meeting_25_January_2017 agenda 13:14:08 ok, moving on 13:14:31 #topic Update from the Release manager 13:14:37 (the list is very low traffic, so far) 13:14:41 khall? bag? 13:15:03 * ashimema wonders if we missed the POD discussion 13:15:18 see agenda 13:15:21 ashimema: i preferred to have the announcements dirst 13:15:26 first 13:15:48 it is too early for bag 13:15:55 or khall 13:16:20 and khall seems offline 13:16:22 so, moving on 13:16:26 lazy people ;-) 13:16:29 #topic Updates from the Release Maintainers 13:16:44 #info 16.11.2 was released on 22nd 13:17:05 cait++ 13:17:07 the plan is to keep with the schedule - i hope it will always work out :) 13:17:12 cait++ 13:17:15 cait++ 13:17:18 packages have been made available the next day this time - drojf++! 13:17:35 #info 3.22.15 was released on 23rd 13:17:36 drojf++ 13:17:39 and bgkriegel++ for sending me the po files on short notice too 13:17:49 awesome 13:17:56 jajm++ 13:17:59 In some cases I asked for feedback or people spoke up that they'd like to have something included 13:18:12 I will try to review those the next week latest and see what I can do for the next release 13:18:20 have some patience, I have a list for those :) 13:18:28 great 13:18:38 Something about 3.22.x 13:18:39 cait: is there a list of outstanding bugs for the stability of 16.11? 13:18:40 jajm++ 13:18:48 it seems that it's buggy since 3.22.14 13:18:54 see bug 17943 13:18:54 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17943 critical, P2, ---, gmcharlt, NEW , "make test" failed when upgrading from koha-3.22.10 to 3-22.14 , also fresh installation of koha-3.22.14 made the same "make test" failing 13:18:59 right now I think my first list is done with - the one that Joubu had sent 13:19:00 mught be a critical one 13:19:14 Joubu, yep, I just saw that, i will take a look asap 13:19:19 i think there might still be some bugs in holds - I haven't reviewed them yet 13:19:22 jajm++ 13:19:35 cait: thanks 13:20:07 #info Katrin says there can be some minor bugs on holds, but 16.11 is stable enough 13:20:14 mtj: around? 13:20:30 didn't say minor 13:20:33 might be havier ones 13:21:13 #info holds bugs in 16.11 can be totally not-minor 13:21:26 :D 13:21:43 heh 13:22:13 cait: bug 17940 might be worth a look 13:22:13 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17940 major, P5 - low, ---, koha-bugs, NEW , Holds not going to waiting state after having been transferred 13:22:14 cait: I'm sure you will make us note those bugs 13:22:58 ok, moving on 13:23:15 #topic Review of coding guidelines / New POD guidelines 13:23:29 kidclamp: your turn 13:23:53 okay 13:24:28 * oleonard has a coding guideline item after kidclamp 13:24:32 I sent out an email - got two responses, I think we can go ahead with enforcing the current rule about POD and adding that it should be inline 13:25:04 kidclamp: no objections so far? 13:25:05 and you already got responses on last meeting 13:25:18 heh, there was more discussion at the meetings :-) 13:25:19 ++ for inline POD 13:25:23 i go for required, inline 13:25:30 ++ for inline POD 13:25:49 ++ inline 13:26:06 any objections? if there are, we should vote 13:26:17 so I say add to guideline: When updating a module lacking POD please add coverage for existing subroutines. All POD should be inline with subroutines 13:26:20 already voted on ml - inline and required :) 13:26:21 lets vote 13:26:23 heh 13:26:37 do you have to do all? or just the ones you use? 13:26:41 use/change? 13:26:51 only what you change 13:26:57 yeah 13:27:00 I think all is encouraged, what you change is required 13:27:03 yeah, only what you change is required 13:27:10 yeah 13:28:11 you probably wont like rebasing them when we change all 13:28:18 When updating a module you are required to add POD for existing subroutines, it is encouraged to add POD to all subroutines as possible 13:28:32 heh - true, but full coverage would be nice 13:28:59 hm existing and all seems the same 13:29:03 for changed/added? 13:29:12 full coverage would be lovely.. but enforcing someone to go back and write pod for each routine when they've done a 1 character change to one routine is a bit rough ;) 13:29:32 yup 13:29:37 existing should be modified 13:29:40 When updating a module you are required to add POD for added and changed subroutines, it is encouraged to add POD to all subroutines as possible 13:29:42 ashimema: the QA team is good when negotiating 13:29:49 I think QA judgement calls are okay 13:29:52 QA++ 13:29:55 I'd say.. ensure POD is accurate and correct for the routines you are altering as the rule 13:30:03 true 13:30:20 I'd love to see some pod only bugs coming in too ;) 13:30:41 barton volunteered to do some work there 13:30:47 barton 13:30:47 awesome 13:30:50 barton++ 13:30:55 barton++ 13:31:08 #info Barton volunteered to write patches fixing POD 13:31:38 #info Nick volunteers to update the coding guidelines text to reflect this change 13:31:49 When updating a module you are required to add POD for added and changed subroutines or ensure existing POD is accurate, it is encouraged to add POD to all subroutines as possible 13:32:14 we need a template 13:32:34 what is usually done : 13:32:35 For two examples, see the POD comments in the Search.pm module (perldoc Search.pm) and the Biblio.pm module. 13:32:40 =head3 method 13:32:51 head2 reads nicer for subroutines 13:32:54 my $return_value = $obj->method ( $params ); 13:32:58 describe the params 13:33:03 what does the method 13:33:05 =cut 13:33:24 Joubu: do you think you can find a good example? 13:33:26 But as I already said 13:33:40 $patron->get_overdues 13:33:49 this method returns overdues for a patron 13:33:56 fantastically useful... 13:33:58 heh 13:34:12 https://wiki.koha-community.org/wiki/Coding_Guidelines#PERL3:_Fix_warnings <- Do you think the current example needs added verbosity? 13:34:17 well, the object type is not useless information 13:34:26 https://wiki.koha-community.org/wiki/Coding_Guidelines#PERL13:_POD 13:34:28 wrong link 13:34:31 tcohen: nope, I don't think we have any 13:34:50 functions or methods? 13:35:13 any good examples 13:35:25 [off] khall noticed the meeting and ducked out 13:35:26 the return type might be useful (hash, list, Koha::*) 13:36:13 distinction between class methods and instance methods ? 13:36:16 return: Koha::Overdues object 13:36:24 yes we could, but still think it's too verbose 13:36:30 it's implicit in the name of the method 13:37:53 def return type 13:37:54 please 13:38:02 #actions Tomas will look for a well formed and useful set of examples to vote on next meeting 13:38:03 I always find it helpful if the documentation is clear about these things 13:38:13 not too much duplicated info in the pod that will be outdated after a few years because it wasnt updated? 13:38:38 very often the params donot match 13:38:54 just three lines lower.. 13:39:03 now we are moving to Koha::Object(s) that are passed around, things will be simpler 13:40:36 #koha: I volunteered to look for good sample PODs to be used as templates so we vote on that next meeting. Right now we need to either vote or accept unanimously the changes Nick proposed 13:41:00 does anyone disagrees with the proposed changes? 13:41:06 nope 13:41:09 not me 13:41:12 vote now 13:41:51 or just do #agreed :) 13:42:02 #startvote Should we add "When updating a module you are required to add POD for added and changed subroutines or ensure existing POD is accurate, it is encouraged to add POD to all subroutines as possible" to the coding guidelines? (yes,no) 13:42:02 Begin voting on: Should we add "When updating a module you are required to add POD for added and changed subroutines or ensure existing POD is accurate, it is encouraged to add POD to all subroutines as possible" to the coding guidelines? Valid vote options are , yes, no, . 13:42:02 Vote using '#vote OPTION'. Only your last vote counts. 13:42:16 #vote yes 13:42:17 #vote yes 13:42:19 #vote yes 13:42:21 #vote yes 13:42:28 #vote yes 13:42:28 #vote yes 13:42:30 #vote no 13:42:38 would have prefered the pod at the end 13:42:39 :-P 13:42:44 #vote yes 13:42:51 Joubu++ 13:42:58 #vote yes 13:43:07 anyone else? 13:43:35 #endvote 13:43:35 Voted on " Should we add "When updating a module you are required to add POD for added and changed subroutines or ensure existing POD is accurate, it is encouraged to add POD to all subroutines as possible" to the coding guidelines?" Results are 13:43:35 yes (8): cait, oleonard, ashimema, jajm, marcelr, kidclamp, tcohen, rsantellan 13:43:35 no (1): Joubu 13:43:44 sorry, Joubu 13:43:51 'perl best practice' suggests after the __END__ block.. but I personally prefer inline ;) 13:43:51 khall: around? 13:43:56 but can see where Joubu is coming from 13:43:57 I can live with that :) 13:44:09 I am from Belgium man! :p 13:44:15 haha 13:44:33 ok, next topic then 13:45:52 #topic Plans for bug 17908 13:45:52 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17908 enhancement, P5 - low, ---, m.de.rooy, ASSIGNED , Authority merge omnibus 13:46:02 ok, have been working on some older authority merge bugs 13:46:16 reorganizing them in a omnibus bug 13:46:25 a merge test has been pushed as well as the sub merge fixes on bug 17913; we are in "stage 3" now 13:46:25 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17913 normal, P5 - low, ---, m.de.rooy, Pushed to Master , Merge three authority merge fixes 13:46:41 #info Kyle Hall, ByWater Solutions 13:46:42 but we need to some more, and would appreciate support on that 13:46:51 marcelr++ 13:46:54 some more ideas: 13:47:00 auth_header should be autoincrement / remove the max(authid) logic 13:47:09 move zebra stuff outside of sub merge, use/extend CountUsage, add a GetBiblionumbers sub/method 13:47:18 restructure need_merge_authorities (perhaps rename it) : include old authno, old authtype and contents of old reporting field 13:47:31 distinction between migration tool and cron job 13:47:43 the table on 14769 should facilitate adjusting indicators when remerging authorities (via a script) 13:47:52 it is quite a lot actually 13:48:29 would be nice to keep some pace in that regard 13:48:32 sounds easy 13:48:33 :D 13:48:56 well thats easy qa too ;) 13:49:05 #info Barton Chittenden, BWS, Louisville KY USA. 13:49:27 I saw that I got volunteered ;-) ;-) 13:49:36 Joubu: solve one bug, find two new ones 13:49:48 #info Marcel has been working on resurrecting old authority merge patches, needs help and feedback 13:49:51 I think we need very good tests coverage for that change 13:50:18 i started with writing merge tests 13:50:25 and that helped i guess 13:50:39 marcelr++ 13:50:53 marcelr: have you seen jcamins work on the authority de-duplication tool? 13:51:00 yeah 13:51:01 could we reuse bib merging code? 13:51:22 in a later stage probably yes 13:51:26 marcelr: I think there was good stuff in there 13:51:45 will come there one day.. 13:53:19 marcelr++ # great work on old under-tested code 13:53:21 #topic Updates on bug 17196 13:53:21 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17196 enhancement, P5 - low, ---, jonathan.druart, RESOLVED FIXED, Move marcxml out of the biblioitems table 13:54:09 Joubu: anything to say about it? 13:54:32 nope 13:54:41 I did not add it 13:54:56 I know, but anything would be your fault :-P 13:55:03 i did not too, but have a tiny remark 13:55:16 There is the patch to automgically convert the SQL 13:55:19 that will need testing 13:55:24 I just had a thought about changes to the wiki the marcxml queries... 13:55:29 had some weird problems with indexing until i restarted koha-indexer after upgrading 13:55:42 Joubu: #? 13:55:42 i think # is for root 13:56:09 bug 17898 13:56:09 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17898 enhancement, P5 - low, ---, jonathan.druart, Needs Signoff , Add a way to automatically convert SQL reports 13:56:10 marcelr: it makes sense :-D 13:56:25 the wiki itself is internally version controlled ... we could put a note at the top of the wiki pointing to the old version. 13:56:30 marcelr: that reminds me that we need to make the packages restart koha-common on upgrade 13:57:00 ok 13:57:25 marcelr: that should be fixed by 16733, right? 13:57:31 bug 1673 13:57:31 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=1673 blocker, P3, ---, gmcharlt, CLOSED FIXED, Error Fields Need Highlighted 13:57:35 bug 16733 13:57:35 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=16733 enhancement, P5 - low, ---, m.de.rooy, RESOLVED FIXED, More flexible paths in debian scripts (for dev installs) 13:57:36 no Joubu 13:57:52 not related to that bug 13:58:02 #info Now that removing marcxml from 'biblioitems' is pushed, it is urgent that people test bug 17898 so it can be pushed and tested A LOT before the release 13:58:42 koha-indexer launches rebuild_zebra.pl -d 13:59:30 from a fixed location 13:59:46 anything else? 13:59:47 somebody said anything else was just being crap 14:00:04 #topic REST api / Bug 17932 14:00:04 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17932 enhancement, P5 - low, ---, tomascohen, Needs Signoff , Koha::Object should provide a TO_JSON method 14:00:45 I wanted to mention that I'm working on a couple new endpoints and I found that our API data types where incorrect 14:01:38 they are fixed by bug 17927 (which resurrects a porting of bug 17008's patches) 14:01:38 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17927 normal, P5 - low, ---, tomascohen, Passed QA , REST API: /holds and /patrons attributes have wrong types 14:01:39 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17008 enhancement, P5 - low, ---, lari.taskula, RESOLVED WONTFIX, REST API: Correct data types in Swagger 14:02:31 so I put together some ideas an wrote 17932 14:02:52 which basically adds a TO_JSON method that handles data types correctly 14:02:58 sorry, have to leave early 14:03:08 me too 14:03:32 the main problem where integers and DBD::mysql which is explained on the first commit message 14:03:38 but also Boolean values 14:03:48 we have a situation with booleans 14:04:05 we store them on the DB as TINYINT(1) 14:04:27 Yes that's a known issue, that can contain only 2 values 14:04:28 but we also store other tinyint values which are not booleans 14:04:47 TINYINT(4) 14:04:59 you just grep -i tinyint on kohastructure.sql and you will notice 14:06:54 so, we cannot deduce from the code if something is a boolean 14:06:54 (i.e. lokking up on the schema) 14:07:09 because (1) is just a mask to tell the DB how do we want to have the 8-bit value sent 14:09:15 DBIC doesn't know about it 14:09:15 so, the subject I wanted to raise 14:09:15 is on the 'Fix /patrons endpoint' patch 14:11:08 * ashimema thinks we shouldn't be dbicdumping any more.. and booleans should be marked as booleans in dbic result classes 14:13:30 noone else still around? 14:13:51 I added a syntax for explicitly stating on the schema files that a field is a boolean value 14:13:51 basically, we need to add __PACKAGES__->add_columns( '+column_name' => { is_boolean => 1 } ); 14:13:52 * oleonard is but is sitting this one out 14:13:53 I'm not opposed to the idea 14:13:54 @quote random 14:13:54 gmcharlt_: Quote #264: " Grokking someone else's code is still trying to reverse-engineer someone else's brain." (added by wizzyrea at 11:50 PM, June 26, 2013) 14:13:55 i already had it that way, huginn. 14:13:55 a proper solution could involve moving all boolean values into (say) BIT column type 14:13:57 which is supported by Postgres as well 14:14:40 hmm 14:14:49 * druthb perks up, smiles, and returns to her hacking. 14:14:51 I hadn't considered that one 14:14:54 this is only useful for the TO_JSON function 14:14:54 and once we figure a better way, we just need to change the implementation on TO_JSON 14:15:00 I added bug 17992, to highlight the consequences of the change on our reference implementation 14:15:00 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17992 minor, P5 - low, ---, tomascohen, Needs Signoff , REST api: Cities controller should not use ->unblessed 14:15:01 I encourage everyone interested on the REST api implementation to comment and/or tests those bugs 14:15:01 and also contact me so we can coordinate and maybe solve issues together 14:15:29 BIT sounds like what we should pick 14:15:36 the title does not really reveal this topic 14:15:49 BIT can store more than true/false value 14:16:22 not BIT(1) 14:16:49 ashimema, I agree, but there's some work to be done so that becomes real 14:16:50 ok 14:17:08 #info Tomas requests comments on bug 17932 and the way it handles booleans, until we figure a DB structure change to ease it 14:17:08 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17932 enhancement, P5 - low, ---, tomascohen, Needs Signoff , Koha::Object should provide a TO_JSON method 14:18:01 #actions Tomas volunteers to write a patch to move boolean values into BIT 14:18:26 tcoehn++ 14:18:29 tcohen++ 14:18:40 tcoehn++ 14:18:50 restapi++ 14:20:30 ok, i'm waiting for the flooding emails and comments about the REST api 14:20:30 the other part of the topic 14:20:30 was 14:21:23 do we have an indentation coding guildeline for JSON? 14:21:23 I added it lat minute and don't have DNS resolution right now to check myself 14:22:46 not that i can see 14:23:16 I would ++ some though ;) 14:23:21 * ashimema likes tidy code 14:24:54 * Joubu votes for 2 tabs + 3 spaces 14:25:22 the best discussion go about tabs and spaces :) 14:25:35 am i online now? 14:25:43 Yes tcohen1 14:25:44 sure 14:25:50 we see you tcohen1 14:26:10 ok, wifi is back, edge connection was really bad 14:26:15 (nick tcohen 14:26:56 ok, i cannot chair as tcohen1 I think 14:27:44 is an op around to kick tcohen? 14:27:54 ? 14:27:56 try 14:28:17 I'm not an op, magnuse? 14:28:53 thanks magnuse! 14:28:54 like that? 14:28:55 like that is hard to vote on. :) 14:29:01 Let that be a lesson to you tcohen 14:29:06 lol 14:29:25 ok, we have JSON with 2 and 4 spaces 14:29:35 * tcohen unhides 14:29:54 any preference? 14:30:02 we should add it to our shipped .editorconfig file 14:30:44 :q 14:30:53 tcohen: couldn't hurt! 14:31:11 Joubu: i bet there's a vim plugin for editorconfig 14:31:33 tcohen, there is 14:31:36 https://github.com/editorconfig/editorconfig-vim 14:32:22 the idea is that you can have your preferred indentation config globally, but the editor picks the right one for the Koha source 14:32:24 anyway 14:32:32 moving on I guesss 14:32:49 #topic Bug 17981 14:32:49 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17981 enhancement, P5 - low, ---, jonathan.druart, Needs Signoff , Add the ability to preview generated notice templates 14:33:20 Nothing to do, just checkout the remote branch and test 14:33:38 it's on top of a quite big tree of patches, non-pushed yet 14:33:40 * magnuse wanders off 14:33:54 Everything is explained in the commit message 14:33:57 * oleonard is working on a template follow-up for it 14:34:04 oleonard: great 14:34:11 and yes, oleonard is going to make it awesome :) 14:34:20 #info Owen is working on a followup for 17981 14:34:36 oleonard++ 14:34:40 oleonard++ 14:35:11 #topic Critical bug for 3.22.x: bug 17943 14:35:11 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=17943 critical, P2, ---, gmcharlt, NEW , "make test" failed when upgrading from koha-3.22.10 to 3-22.14 , also fresh installation of koha-3.22.14 made the same "make test" failing 14:35:12 I could also ask on the ML for some people to send me non-standard templates 14:35:25 to see if the conversion works as expected for complex templates 14:35:29 (that's all) 14:35:52 about 17943, I talked about it earlier 14:36:08 i'll look at this bug asap 14:36:33 #info Julian is going to work on bug 17943 ASAP 14:36:53 #topic Security bugs to QA 14:37:07 there are some security bugs in the QA queue 14:37:16 will look at some on friday 14:37:27