20:08:46 #startmeeting Development IRC meeting 27 December 2017 20:08:46 Meeting started Wed Dec 27 20:08:46 2017 UTC. The chair is Joubu. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:08:46 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:08:46 The meeting name has been set to 'development_irc_meeting_27_december_2017' 20:08:51 #topic Introductions 20:08:52 #info wahanui, a bot that has become sentient 20:08:57 #info Jonathan Druart 20:09:00 #info Katrin Fischer, BSZ, Germany 20:09:11 #info Tomas Cohen Arazi, Theke Solutions 20:10:00 let's change the versioning again :) 20:10:04 so, quick? 20:10:09 #topic Announcements 20:10:18 #link https://wiki.koha-community.org/wiki/Development_IRC_meeting_27_December_2017 20:10:31 Anyone have something? 20:10:37 me 20:10:40 nope 20:10:53 hm nope, that's for later 20:11:12 #topic Update from the Release Manager (18.05) 20:11:43 As I do not think people read the minutes I have nothing to say 20:11:52 #topic Updates from the Release Maintainers 20:11:53 * cait always read the minutes 20:12:04 rangi: ? 20:12:17 me too, but I'm here hehe 20:12:18 im all ready to release, just waiting on the others 20:12:30 (who are waiting on packages) 20:12:41 same as frido and kidclamp 20:12:48 nothing from Mirko? 20:12:57 we need to auto generate the packages 20:13:00 to avoid such situations 20:13:09 because it's a nightmare to wait for people 20:13:14 there are nightly builds already 20:13:43 master is built, that's correct 20:13:47 but doesn't look like they have been updated yet 20:13:50 that and the manual, we need to automatize as much as we can 20:13:59 https://apt.abunchofthings.net/koha-nightly/ 20:14:27 ah 20:14:44 * rangi has to duck out for a meeting, bbiab 20:15:27 ok, so waiting for packaging manager 20:15:32 no need to add more I think, we (rmaints+me) are waiting for Mirko's answer 20:15:33 #topic Updates from the QA team 20:15:50 sec 20:16:08 #info the QA team had a QA sprint 20:16:10 #link https://wiki.koha-community.org/wiki/2017-12-21_QA_Sprint_Day 20:16:11 do we have Mirko's phone number? :-D 20:16:32 #info We got the SO queue down from 154 to about 108 currently 20:17:01 still lots to do 20:17:22 i have added a new entry page to the wiki 20:17:44 #link https://wiki.koha-community.org/wiki/How_to_QA 20:18:24 please let me know if something is missing 20:18:31 done 20:19:26 #topic General development discussion (trends, ideas, ...) 20:19:41 so we are supposed to vote on dozen of topics, and we are 3 20:20:02 Do we postpone or maintain the votes anyway? 20:20:14 we already postponed api votes to this meeting once 20:20:15 https://orig00.deviantart.net/835a/f/2014/199/0/a/nobody_cares__by_rejectedbandgeek-d7r7hs9.jpg 20:20:39 the meting was announced and all... i feel like we should go ahead - unelss we hit something really controversial 20:21:06 I'm comfortable if Katrin has added to the RFCs, it is mostly naming conventions (for what's REST-related) 20:21:33 #topic REST api - vote on cities, holds, libraries and patrons endpoint RFC 20:21:46 the problem was the change of the version I think 20:21:49 there is a notion from thd that I tihnk we could discuss first 20:21:59 3 of the endpoints have zip or zipcode 20:22:05 #chair cait 20:22:05 Current chairs: Joubu cait 20:22:06 could we use postal_code? 20:22:11 should 20:22:24 zip_code++ 20:22:34 or whatever rangi says 20:22:43 he's the only native-speaker attendant 20:23:12 it was thd and it was post_code :) 20:23:15 he has ducked out for a meeting 20:23:20 oh thx 20:23:20 :) 20:23:49 do we all agree with post_code? 20:23:54 yes 20:24:06 unanimous 20:24:20 my dictionary says postal code... 20:24:25 maybe clear that up before programming it 20:24:33 but otherwise i am ok with it :) 20:25:06 agreed on everything then apart the zipcode? 20:25:25 i tihnk so 20:25:30 we will have a final check with a native speaker 20:25:38 I agree with my proposals, added cait's suggestions 20:25:51 #agreed change zip or zipcode to postal code or post code (whatever is the correct one) 20:26:06 #agreed unanimously the 4 RFC for cities, holds, libraries and patrons 20:26:07 heh :) 20:26:17 excellent 20:26:28 so it's just the zip thing and they are good to go 20:26:59 I keep my 'local manual page edition' topic for later then... 20:27:06 makes sense 20:27:25 kidclamp added " Build packages automatically for stable releases? ", postponed, will see what happen next month 20:27:31 I think we all agree on that 20:27:45 the only problem is gpg signing the packages 20:28:04 we could stage the automatically built packages somewhere 20:28:15 and then implement a way for rmaints to sign 20:28:18 We can ask more reactive people to do it 20:28:23 or have more people involved 20:28:30 ok, what you said 20:28:33 #topic Review of the coding guidelines 20:28:36 did you read them? 20:28:53 yes 20:28:57 #topic new coding guidelines J12 20:29:01 #link https://wiki.koha-community.org/wiki/Coding_Guidelines#JS12:_Include_javascript_at_the_end_of_template_JS12:_Include_javascript_at_the_end_of_template 20:29:08 link is broken 20:29:32 #linkhttps://wiki.koha-community.org/wiki/Coding_Guidelines#JS12:_Include_javascript_at_the_end_of_template_JS12:_Include_javascript_at_the_end_of_template 20:29:38 odd 20:29:40 anyway 20:30:03 look sgood to me 20:30:04 we already pushed lot of thing is that sense 20:30:08 an example is always helpful 20:30:40 #link https://wiki.koha-community.org/wiki/Coding_Guidelines#JS12:_Include_javascript_at_the_end_of_template 20:30:55 :) 20:31:04 i mean it sounds alright +1 form me 20:31:05 we need a way to catch that automatically, but will not be easy 20:31:29 #agreed coding guidelines J12 - include JavaScript code at the end of template 20:31:49 #topic New rule SQL12: Booleans (DRAFT) 20:32:01 #link https://wiki.koha-community.org/wiki/Coding_Guidelines#SQL12:_Booleans_.28DRAFT.29 20:32:15 #link https://wiki.koha-community.org/wiki/Coding_Guidelines#SQL12:_Booleans #for later 20:32:16 cait: are you aware of what is it about 20:32:16 ? 20:32:24 let me quickly look plz 20:32:55 tcohen: Is there a bug to update existing values? 20:33:02 so tinyint will be automatically converted to boolean when generating the schema? 20:33:07 or does it need to be changed? 20:33:18 there isn't, but I can do file it and implement it, JOubu 20:33:19 yes 20:33:34 cait: the idea is that we annotate boolean columns in the schema files 20:33:55 so the Koha::Object::TO_JSON method can translate them to proper booleans 20:33:59 so there is no way it will recognize them automatically? 20:34:09 there isn't 20:34:13 you should also explain the context in the same section 20:34:14 ok 20:34:23 yep i tihnk one or 2 mroe sentences would be good 20:34:37 Joubu: you mean mention TO_JSON? 20:34:40 yes 20:34:43 ydp 20:34:48 yep, or we will forget 20:34:51 and link to a bug 20:34:52 and may question it later 20:35:00 the codebase mostly uses tinyint(1) 20:35:05 but otherwise I am fine with it : ) 20:35:12 but I've seen a couple patches introducing other types 20:35:39 the problem is DBIx::Class calls them all integer (TINYINT is an alias for INTEGER) 20:36:12 if we agree, I can volunteer to expand the explanation of this decision 20:36:17 #action tcohen improve the description of coding guideline SQL12 and open a bug report 20:36:22 and file a bug for catching all booleans around 20:36:31 that 20:37:11 IMO it will not fair to vote on it as it, but I do not see why people would oppose 20:37:32 It is not fair* 20:37:33 i think go ahead and have the formal vote next meeting 20:37:43 i mean file the bug and adapt description 20:37:50 #info SQL12 vote is postponed, lack of information on the wiki page 20:37:54 yeah, no rush really 20:38:06 but you can consider it as ok I think 20:38:16 we are already doing it 20:38:23 #topic New rule "No DBIx::Class in atomic updates" 20:38:32 each endpoint we write, we track booleans and act on them 20:38:37 #link https://wiki.koha-community.org/wiki/Database_updates#Modify_the_data_or_the_database_structure 20:39:04 oh sorry 20:39:06 i broke that one 20:39:07 ha, been updated 20:39:12 #link https://wiki.koha-community.org/wiki/Database_updates#DBIx::Class_.28DRAFT.29 20:39:47 so we should use plain SQL? 20:40:17 it says what NOT to do, but not what to do :) 20:40:17 yes 20:40:28 and perl 20:40:45 but you have to select/update/delete/whatever using SQL queries 20:41:02 and do not call subroutines/methods 20:41:37 we can postpone this one too as it's an important one 20:41:56 i tihnk i would clearly state that 20:41:57 even if it will be hard to have good arguments against it 20:41:58 just to be sure 20:42:10 the QA team already fails otherwise, right? 20:42:18 well yes, good arguments would be "it's easier to write" 20:42:20 i was thinking the other day that we might want to clear up coding guidelines (at another meeting) about koha object, dbic and when and when not to use sql 20:42:29 we have a lot of guildelines about how to get the sql right 20:43:12 might also be good to take another look and write down current practice for the koha namespace vs C4 (adding to it) 20:44:09 the update DB entries should be the only places where SQL queries are written, unless exceptions (too hard/slow with DBIx::Class) 20:44:49 as i have started this... do you want to add an action for me to make a suggestion? 20:45:40 #info forbid the use of DBIx::Class schema from update DB vote is postpone 20:46:08 #action cait improve the coding guidelines related to the use of SQL queries 20:46:13 make sense? :-/ 20:46:32 yup 20:46:32 +1 20:46:39 will review and see what i can come up with 20:46:44 #topic New rule "Write good commit messages" 20:47:01 #link https://wiki.koha-community.org/w/index.php?title=Commit_messages&action=historysubmit&diff=19998&oldid=11190 20:47:12 this is the changes I made to the Commit_messages page 20:48:17 did you read that? 20:48:49 I'm finishing it, sorry 20:49:40 I agree with the changes 20:50:00 ok for me, but please be lenient with new committers 20:50:11 especially with the open source academy coming up in a few weeks 20:50:16 it's basically what I wrote on my email one month ago 20:50:26 cait: as always 20:50:27 :) 20:50:30 yes, I don't think that will be a blocker, but it is good to enforce such practices 20:50:37 (i.e. having it written) 20:51:03 I am updating the commit messages on push, and let a note on the bug report when I reword the message 20:51:13 but I'd like to avoid to do it by myself 20:51:18 first because it's time consuming 20:51:32 and, most of all, my English is bad and wrong 20:52:09 the most important part is to understand that the bug title must describe the bug, whereas the commit messages must describe what the patch does 20:52:29 when it's a bug fix, most of the time they are different 20:52:38 what bugs me a little about this is release notes 20:52:45 we list the 'broken' in the release notes 20:52:49 instead of the fix :) 20:53:08 but don't see a good way to fix this 20:53:10 we list the bug report titles 20:53:21 ey yep 20:53:31 so release notes say: Feature X is broken 20:53:31 so it will not change anything 20:53:36 instead of Fix feature X 20:53:37 yep 20:53:59 just noting 20:54:15 ok 20:54:30 my point is more about the commit messages, when pushed they cannot be modified 20:55:04 we can think about your point, and use the "notes to go to the changelog" field is needed 20:55:43 * tcohen was distracted tweaking the boolean guidelines 20:56:03 postponed then? 20:56:10 we'd have to change all the bug titles, not easy 20:56:16 but separate problem :) 20:56:32 we should add a separate field for the release notes 20:56:33 for me it's ok - you already announced the changes on the list 20:56:37 so people could have commented 20:56:40 I do not change the bug titles, they are already like that 20:57:06 Joubu: yep sorry for sidetracking 20:57:36 nope you are right to raise it, I will try and keep that in mind 20:58:15 #info Read https://wiki.koha-community.org/wiki/Commit_messages to know how to write valid commit messages 20:58:40 #info If you are a regular contributors and do not follow these rules, RM may FQA the patches 20:58:45 no need to vote, I decide :D 20:59:02 Joubu++ 20:59:05 #topic Set time of next meeting 20:59:26 10 January 2018, 14 UTC? 20:59:45 I'll be absent that week 21:00:04 i should be around 21:00:11 can any of you review my changes 21:00:11 https://wiki.koha-community.org/wiki/Coding_Guidelines#SQL12:_Booleans_.28DRAFT.29 21:00:14 i am not there the third week 21:00:37 #info Next meeting: 10 January 2017, 14 UTC 21:00:44 #endmeeting