19:02:56 #startmeeting Development IRC meeting 12 April 2017 19:02:56 Meeting started Wed Apr 12 19:02:56 2017 UTC. The chair is kidclamp. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:02:56 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:02:56 The meeting name has been set to 'development_irc_meeting_12_april_2017' 19:03:08 #topic Introductions 19:03:08 #info wahanui, a bot that has become sentient 19:03:17 #info Nick Clemens, ByWater Solutions 19:03:28 Hi # koha 19:03:37 meeting mveron? 19:03:42 #info Marc Véron, Koha Support Schweiz, www.koha-support.ch 19:03:51 #info Barton Chittenden, BWS, Louisville KY, USA 19:04:45 #info Adam Brooks, ByWater Solutions 19:04:56 meeting ashimema Joubu cait? 19:05:08 #info Christopher Brannon, Cooperative Information Network/Coeur d'Alene Public Library, Coeur d'Alene, Idaho 19:05:31 #info bag brendan gallagher bywater 19:06:33 looks to be a light meeting :-) 19:06:46 #topic Announcements 19:06:55 anyone? 19:06:56 Anyone? Anyone? Bueller? Bueller? 19:08:31 well - khall is gone for the day, cait mtj juliam are not here so release updates are lacking 19:08:42 sorry on phone 19:08:44 everything will be on time 19:08:55 string freeze on 15th, release on 22nd 19:08:58 #topic Update from Release maintainers 19:09:12 info 16.11.x string freeze on 15th, release on 22nd 19:09:40 (sorry not around) 19:09:45 #topic Updates from the QA team 19:09:54 no worries Joubu 19:10:17 We have a good amoutn of patches in the queue and NSO queue is growing 19:10:48 Everyone should read Joubu's email and encourage more involvement 19:11:51 #info joubu sent What's on in koha-devel #11, please read and encourage involvement 19:12:10 any one else have bugs to highlight? 19:12:23 #topic General development discussion (trends, ideas, ...) 19:12:33 I just had two things to mention 19:12:45 #topic Elasticsearch status 19:13:03 There are a number of open bugs, I am testing what I can and submitting patches 19:13:24 We need more testers to help us identify what needs to be improved and to help keep patches moving 19:13:40 Kohadevbox installs ES and makes it very easy to test it out 19:13:53 I will send an email with more specific bug numbers asap 19:14:12 #info ES needs testers and patchers, talk to kidclamp if you have questions 19:14:42 comments? 19:14:43 comments are turned on 19:15:16 #topic Noisy logs 19:15:35 When you send out that e-mail, can you include specifics to ES use in kohadevbox? 19:15:58 CrispyBran, sure, kohadevbox instructions should include it too but wil highlight 19:16:23 We have notices lots of noisy warns in production on value comparison with uninitialized values 19:16:28 bug 18400 is an example 19:16:28 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=18400 normal, P5 - low, ---, oleonard, Needs Signoff , Noisy warns in opac-search.pl during itemtype sorting 19:16:53 these may be data related but are causing huge numbers of errors in logs 19:17:10 will bring this up at next meeting or in lists 19:17:22 Anyone else have announcements? 19:17:49 kidclamp: about ES - is there any current documentaion/overview right now? 19:17:52 i think it would help 19:17:59 sorry, just reading back 19:18:20 docs on how to use? or what is to be expected? 19:18:28 what works/is exepcted 19:18:31 to wor 19:18:32 k 19:18:35 :-) 19:19:11 basically most things should work - otherwise search bugzilla for 'elastic' and those thigns might now 19:19:29 hm ok 19:19:51 bug 18318 19:19:51 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=18318 enhancement, P5 - low, ---, gmcharlt, NEW , Unicode support for Elasticsearch 19:20:22 ES is different from zebra, so some things are learning 19:20:35 partial terms don't find things, you must add a '*' 19:20:42 bug 18374 19:20:42 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=18374 major, P3, ---, nick, Failed QA , Respect QueryAutoTruncate syspref in Elasticsearch 19:20:56 but we need to decide if we want to try and auto truncate, or train suers 19:20:59 users 19:21:22 but testers are needed to tell us what they expect and it doesn't do 19:22:26 unicode support is big and necessary 19:23:35 I am trying to get an organized chain of thought together to send out cait 19:23:43 instead of just me shouting things :-) 19:24:22 kidclamp: what's not working in respect to unicode in elastic? 19:24:40 currently all characters must be exactly searched for 19:25:04 there are no equivalencies between accented and unaccented characters 19:25:17 gotcha. 19:25:24 so pokemon <> pokémon 19:26:25 no coding guidelines on agenda, so I think we just have to set next meeting (and hope is a better time for all) 19:26:26 :-D 19:26:34 I'm somewhat sure that can be defined, but I've never looked at it. 19:27:03 I changed 18318 to NSO - there is a patch there - basically makes it use the unicode plugin and analyzers 19:28:04 Next meeting: April 26th 19:28:21 13 UTC 19:28:25 any objections? 19:29:20 real quick, should we talk about the database update that was hanging? 19:29:29 ^^ kidclamp. 19:29:53 the floor is yours barton 19:31:45 larryb was running upgrades from 3.22, there's a database update to add itemtypes to statistics where type='return'. 19:32:01 ... this ended up taking 12 hours on one of our sites. 19:33:09 bug 14598 19:33:09 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=14598 major, P5 - low, ---, tomascohen, Pushed to Master , itemtype is not set on statistics by C4::Circulation::AddReturn 19:33:46 the db update to 'guess' the type in statistics is panifully slow on large statistics tables 19:33:46 kidclamp: do you know if larryb filed a bug about the upgrade itself? 19:33:54 I do not 19:33:57 usually that doesn’t happen barbara 19:34:01 ha! barton 19:34:12 auto-tab fail :) 19:34:31 #info bug 14598 DB update can be very slow/problematic in production sites with large statistics tables 19:34:42 I would say file a bug and announce to lists barton 19:34:58 will do kidclamp. 19:35:11 #info Next Meeting: 26 April, 2017, 13 UTC 19:35:22 last chance anyone :-) 19:35:33 it caused enough consternation that I figured that is deserved to be more widely known ;-) 19:35:44 barton++ 19:35:54 #endmeeting