Time Nick Message
23:59 cait need a qa person for your plans? :P
23:58 BobB or it could alternate between oz and nz
23:58 BobB there are guys up there who could help plan it, and papa is nearby too
23:57 eythian BobB: I wouldn't say no to an idea like that...
23:56 eythian dcook: yeah
23:56 BobB Gold Coast?
23:56 BobB oceana hackfest +1
23:48 dcook Anyway, "we" as in Catalyst, eythian?
23:46 dcook Of course now all I can do is read about it :p
23:44 dcook I've thought it best to avoid anything Mogwai might do or say aside from their music
23:44 dcook hehe
23:40 eythian http://thepopcop.co.uk/wp-content/uploads/2011/03/mogwai-blur-are-shite-t-shirt.jpg <-- dcook
23:38 eythian we've thought about it
23:38 eythian yeah
23:38 dcook I like the idea of an Oceania hackfest :)
23:36 dcook So a little dark?
23:36 dcook It's a bit overcast today...
22:08 eythian It's very much the opposite of dark here.
22:08 eythian I hear that happens there sometimes
22:05 cait europe is dark
22:04 cait just a bit slow tonight
22:04 cait ah sory
21:55 eythian still there?
21:55 eythian how's europe this morning?
21:54 cait hi eythian
21:51 eythian hi
21:33 cait more hackfests! :)
21:14 Viktor Ok so magnuse is out. But let it be known that I agree on the thought of a nordic hackfest.
21:05 huginn barton: The operation succeeded.
21:05 barton @later tell chris_n can I ping you on the status of Bug 12553 - copynumber, location not printing correctly on spine labels? I've got a second partner who just ran into the same issue.
21:02 barton hmm. he seems to be away from his desk...
21:01 huginn barton: chris_n was last seen in #koha 2 weeks, 4 days, 6 hours, 1 minute, and 24 seconds ago: <chris_n> even if its not intuitive behavior :-)
21:01 barton @seen chris_n
20:55 drojf 'how i wanted do look at doing signoffs and had to fix my mediawiki for 2 hours instead' :/
20:41 magnuse (on his laptop, that is)
20:41 * magnuse thinks so too, but is running out of battery
20:37 drojf scandinavia should definitely pay to fly rangi in :)
20:34 drojf magnuse: are you going?
20:34 rangi ill happily go anywhere to explain ncip .. maybe oslo, finland and sweden could combine to pay airfare? :-) (people can't say yes if you don't ask)
20:32 rangi heh
20:32 magnuse that is the first opportunity, i think :-)
20:30 drojf or we hack in joensuu ;)
20:30 magnuse :-)
20:28 drojf magnuse: you should start organizing one now :)
20:28 magnuse ...and corner rangi for half a day to talk about ncipserver
20:27 rangi :)
20:27 * magnuse would love to sneak in
20:27 magnuse yay
20:27 rangi i was thinking of running an oceania one at some point
20:26 magnuse (hopes that term is not seen as derogatory by the antipodeans)
20:25 magnuse an antipodean hackfest!
20:25 magnuse definitely!
20:24 rangi and invite nzers or something :)
20:24 magnuse kia ora paul_p_
20:24 magnuse we could do a nordic or scandinavian hackfest at some point...
20:23 magnuse i feel the same way even though i had 4 days ;-)
20:20 Viktor magnuse: I only wish I could have stayed a day or two more. I was just getting started when it was time to go home. Would have been nice to had a day gathering the nordics in a corner and working :)
20:15 magnuse Viktor: yeah, very nice to meet you too again!
20:10 Viktor And all of #koha - it was nice meeting some of you last week!
20:10 Viktor Kia ora Magnuse :)
20:07 magnuse kia ora Viktor
20:05 magnuse hiya barton
20:04 * barton waves back to magnuse.
20:03 * magnuse waves
20:02 barton eh. I created a new bug, but made copious notes. I'll see kyle tomorrow anyway.
19:58 rangi (sorry had to duck off to take kids to school)
19:58 rangi barton: in that case, yeah id change that back to assigned, note that, then rewrite it and add a new patch
19:33 barton well... I *do* think that 13442 is probably going to end up re-written. It might be something as changing the regex to include 'mc-loc', or there may be code that needs to be ripped out. I'm looking at this through a programmer-scope (aka the core of a roll of paper towels).
19:27 rangi (unless that's what you want to happen :))
19:27 rangi if you change the scope of 13442, then it has to go back through the process again
19:27 barton k. will do.
19:27 rangi barton: the former
19:26 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=13442 normal, P5 - low, ---, kyle, Passed QA , Facet links broken when collection code limit is used
19:26 barton cait: I have a bug protocol question: I reported bug 13442, which passed QA... that bug fixes buildQuery() for collection codes ... now I've found an exactly analagous bug triggered by 'mc-loc' rather than 'mc-itemtype' (i.e collection code). This suggests that there's code duplication that probably needs to be removed. Do I file a second bug and link back to 13442, or do I change the scope of bug 13442?
17:37 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=13691 enhancement, P5 - low, ---, jonathan.druart, In Discussion , Add some selenium scripts
17:37 akafred Entered my first bugzilla discussion on bug 13691
17:29 fridolin see U
17:29 fridolin see y
17:29 fridolin :D my hands are tired, its evening here
17:23 barton_lunch it could be argued that fuzzy matching does work very goof ( as in goofed up ;-)
17:19 fridolin querystemming is much more important
17:18 fridolin good
17:18 fridolin goof
17:18 fridolin fuzzy does not work very goof
17:18 barton_lunch thanks fridolin... I was able to bypass the question... what the partner thought was a bug was not a bug.
17:13 clrh https://www.flickr.com/search/?tags=hfk15 new photos uploaded from hackfest :)
17:11 amyk no problem clrh any time
17:10 fridolin barton_lunch: http://www.indexdata.com/zebra/doc/index.html
17:10 fridolin otherwise the only doc is indexdata :
17:10 clrh thanks amyk
17:10 clrh :)
17:10 amyk he should answer email then
17:10 amyk ah okay
17:09 clrh (little bed for child)
17:09 clrh amyk: I wanted to know where he put my bed and mattress saturday morning ;) (I sent an email)
17:08 amyk clrh: can you let me know what you need from brendan?
17:05 fridolin barton_lunch: the best doc is the source code i'd say
17:00 clrh JesseM_away: I am looking for someone from bywater to ping brendan, can you help me ?
16:56 gaetan_B bye!
16:49 geek_cl anyone involved with acquisitions/budget module ?
16:23 barton Is there any documentation on the methods that zebra used for fuzzy matching?
16:18 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=13810 normal, P5 - low, ---, gmcharlt, NEW , Prevent updatedatabase.pl from breaking because of case insensitive utf8 collation
16:18 nlegrand Ok bug 13810 filled :)
16:05 reiveune bye
15:56 geek_cl any idea ?
15:51 geek_cl Hi guys, We are looking for ways to hide the budget to other libraries in the acquisitions module, each administrator is needed, see the budget that corresponds to your own library and NOT see everything.
15:48 cait nlegrand: would be good to file a bug i think
15:07 akafred to set it to utf8_danish_ci to get our quirky characters in order).
15:07 akafred cait: I agree - setting collation to anything but "utf8_bin" for columns that are used as ids, keys, usernames etc sounds like a bad idea, and perhaps a security risk. Perhaps even collation should default to utf8_bin except in fields that explicitly are used to sort for end users. And even in those cases it should be configurable (like we will probably have
14:52 nlegrand Joubu: are you there?
14:52 nlegrand ho sorry for the scruffy sentence *guess *we're *inside
14:51 nlegrand I guesse were not the only ones having same letter with different case insinde same framework and tagfield :)
14:50 nlegrand Do you think I should open a bug ?
14:50 nlegrand I got no problem with utf8_bin.
14:50 nlegrand Records: 12880 Duplicates: 0 Warnings: 0
14:50 nlegrand mysql> ALTER TABLE marc_subfield_structure CONVERT TO CHARACTER SET utf8 COLLATE utf8_bin;Query OK, 12880 rows affected (0.15 sec)
14:49 nlegrand Someone says UTF8_GENERAL_CI is accent insensitive and UTF8_BIN not.
14:49 nlegrand I found some info here: <http://stackoverflow.com/questions/4070642/converting-mysql-tables-from-latin1-to-utf8>
14:48 cait that woudl be quite bad
14:48 cait because ci = case insensitive
14:47 cait it looks like it doesn't like uppercase subfield codes maybe? :(
14:47 cait sorry, not so far, but it looks troublesome
14:42 nlegrand is this something someone else had trouble with?
14:42 nlegrand It seems like 'ACQ-995-B' is the same key as 'ACQ-995-b' for utf8_unicode_ci
14:42 nlegrand ERROR 1062 (23000): Duplicate entry 'ACQ-995-b' for key 'PRIMARY'
14:42 nlegrand It gives me this : mysql> ALTER TABLE marc_subfield_structure CONVERT TO CHARACTER SET utf8 COLLATE utf8_unicode_ci;
14:41 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11944 major, P5 - low, ---, jonathan.druart, Pushed to Master , Cleanup Koha UTF-8
14:41 nlegrand I'm running into something with 3.19 database update because of Bug 11944
14:40 nlegrand Re #koha.
14:12 barton no problem. ping me in channel when you get to it, if you don't mind.
14:10 barton *loss -- although losss is more descriptive of my feelings ;-)
14:09 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=6499 enhancement, P5 - low, ---, barton, Failed QA , MARC21 035 -- Other-control-number -- Indexing & Matching
14:09 barton would you mind taking a peek at bug 6499? I'm *very* interested in knowing why its working for me, but not for you. I've seen similar issues between our test and production environments, but I'm at a total losss.
14:08 cait probably won't before thursday
14:08 cait sorry i had no time yet to look at the index problem again
14:08 cait morning barton
14:07 barton afternoon, cait! {morning for me}
13:52 drojf :/
13:52 drojf but have nothing to add, really.
13:52 drojf sorry, had to prepare pizza :)
13:49 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12072 enhancement, P5 - low, ---, veron, ASSIGNED , New dateformat dd.mm.yyyy
13:49 mveron Joubu: What would you advise regarding Bug 12072 ? What would be the best approach?
13:47 cait mveron: not sure about the technical aspects :( maybe Joubu culd suggest something
13:47 akafred squitch looked interesting magnuse, https://metacpan.org/pod/distribution/App-Sqitch/lib/sqitch.pod
13:46 mveron There are includes with C4::Data. Touching them affects a lot of pages.
13:46 mveron cait: step by step wold be fine, but...
13:43 cait like we did with some of the TT plugins?
13:43 cait add the new date format to the new modules and then switch over page by page?
13:43 cait hm could we do a step by step approach?
13:39 mveron Every call to a C4::Date function results in an error message with new format added.
13:38 mveron Wrong word, Fixing, patching, about 50+ files
13:37 drojf mveron: what do you mean by much tweaking anyway?
13:36 drojf i'd consider it broken for german instead of an enhancement, but joubu did not seem to agree ;)
13:35 mveron The problem is that you have to do a lot of tweaking anyway. To much technical debt. It is better to resolve it at the root.
13:33 drojf in case it is very important to a library, roll your own packages with release+patch until it is resolved?
13:32 mveron :-(
13:32 mveron drojf: I'm afraid that there is no easy workaround to add dd.mm.yyyy date format. Difficult to explain to potential users.
13:27 huginn drojf: The current temperature in Berolinastr., Berlin-Mitte, Berlin, Germany is 14.9°C (2:25 PM CET on March 09, 2015). Conditions: Partly Cloudy. Humidity: 63%. Dew Point: 8.0°C. Pressure: 30.30 in 1026 hPa (Steady).
13:27 drojf @wunder berlin, germany
13:26 mveron :-)
13:26 mveron drojf: Oh no...
13:26 drojf mveron: we could have a german fork of koha :P
13:14 mveron cait: The problem is that adding a new date format affects 50+ files (because C4::Date issues error messages for new formats).
13:11 cait mveron: not yet, i think it' smore a step by step thing right now
13:08 JesseM akafred thank you Im looking at that now
13:07 akafred JesseM: I split the browser test stuff into a separate repo: https://github.com/akafred/browsertests
12:58 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=12072 enhancement, P5 - low, ---, veron, ASSIGNED , New dateformat dd.mm.yyyy
12:58 mveron ... it blocks Bug 12072
12:54 mveron The use of C4::Dates is depracated. Is there already a Bug to replace / remove it from Koha?
12:50 mveron Hi again to everybody :-)
12:11 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=13068 enhancement, P5 - low, ---, paul.poulain, Passed QA , New feature for DB update and sandbox
12:11 * magnuse hopes bug 13068 will be pushed soon
12:03 kivilahtio jcamins: And it is a basis for other people to refine the solution.
12:03 kivilahtio jcamins: maybe a bit too nice? If it makes you happy, I wouldn't push my patches either. But it is a convenient way of sharing the work we do, which is mostly good enough for our current state.
12:02 jcamins Tomas is a lot nicer than I am.
12:01 kivilahtio jcamins: He has pushed some already, so I cannot be that bad.
12:00 kivilahtio jcamins: I think this is the first time I said this out loud. Maybe things will change
12:00 jcamins Actually, I'm surprised Tomas will consider your patches.
11:59 kivilahtio jcamins: :)
11:58 jcamins kivilahtio: good thing you discovered this updatedatabase trick after I stopped being RM.
11:58 BobB cait about?
11:37 kivilahtio Joubu: Being a RM is serious business, I heard
11:37 kivilahtio Joubu: on the hindsight I saw some pictures about the place, I regret a bit I didn't go, but just felt so tired. Next year for sure!
11:37 Joubu our RM is not funny...
11:36 kivilahtio Joubu: yes
11:36 Joubu kivilahtio: Did Tomas abandon you?
11:35 kivilahtio Joubu: I didn't feel like going alone
11:35 Joubu kivilahtio: how was the club?
11:31 * magnuse crosses fingers
11:31 akafred ... probably appear in not too distant future, I guess...
11:30 akafred 13068 has passed QA and will
11:19 magnuse ah http://sqitch.org/
11:18 * magnuse can't think of the name of the one he is thinking of right now
11:18 magnuse but there are perl tools too
11:18 magnuse i don't think anything that requires java on every server will have a high chance of making it into koha
11:17 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=13068 enhancement, P5 - low, ---, paul.poulain, Passed QA , New feature for DB update and sandbox
11:17 magnuse akafred: bug 13068
11:17 magnuse i think some people looked at it in marseille
11:17 magnuse there has been some work on an improvment
11:13 kivilahtio with this change in my working routine, the things is working quite well.
11:13 kivilahtio Imagine you are a russian submarine on American waters, where would you hide?
11:12 kivilahtio so avoid the bottom like plague and put changes somewhere where no-one has touched in a looong time.
11:12 kivilahtio akafred: I used to put them as the last entry but htey always conflict with everything and that is driving me crazy. So I just "hide" the update subroutine somewhere amongst the v3.8 update routines and less. It is less likely for them to conflict there. The release manager can then sort the statements to the correct place, it is not a big effort.
11:11 kivilahtio akafred: I always put my updatedatabase.pl modifications somewhere up to the file
11:09 akafred a drawback with LiquiBase is that it requires java to run the migration, but it solves exactly this problem.
11:08 akafred conflicts and issues with updatedatabase.pl must be common, right? has http://www.liquibase.org/ been looked at? or some similar tool?
10:45 magnuse yup
10:44 ashimema_ correct
10:42 akafred I am looking at a bug X with multiple patches (first one I look at with more than one, actually, yes I am a noob) - so when I 'git bz apply X' there is a list. I answer (y)es and the first patch doesn't apply cleanly to master. This means the bug should be moved to "doesn't apply", right?
10:06 magnuse mveron: that would be awesome!
10:04 kivilahtio morning
10:04 kivilahtio paul_p: I forgot my mouse there!
10:03 cait morning paul_p
10:01 paul_p mveron you'll be welcomed, of course
10:01 paul_p BibLibre office almost cleaned & re-arranged as usual.
10:01 mveron Hi paul_p. I hope I will make it to Marseille next year!
10:01 * paul_p dropped tomas at the airport tomorrow, after a last trip in the country. he should be in buenos aires in 1 hours, 4 hours before his last plane, for Cordoba
10:00 paul_p hello #koha. I hope everyone is back home safely after #hkf15 !
09:59 mveron hiya magnuse :-)
09:59 magnuse hiya mveron
09:53 mveron Hi #Koha
09:53 magnuse ouch!
09:52 cait got a bit of a cold tho
09:52 cait yep
09:52 magnuse bonjour nlegrand
09:47 nlegrand Hey #koha!
09:43 magnuse safely back in the office again?
09:42 magnuse kia ora cait
09:41 cait morning #koha
09:16 magnuse hiya codavid
09:11 codavid hi
09:04 fridolin HFK was just amazing
09:03 fridolin you miss me so much
09:03 fridolin bonjour tout le monde
08:40 magnuse bonjour gaetan_B fridolin
08:34 gaetan_B hello
08:27 sophie_m hi #koha
08:27 sophie_m hello magnuse
08:25 magnuse bonjour sophie_m
08:15 * magnuse adds a snapshot of the scoreboard to the hfk15 wiki page. 201 kittehs saved!
08:02 alex_a bonjour
07:55 * magnuse waves
07:55 matts hi #koha !
07:46 reiveune hello
06:58 drojf morning #koha
06:18 dcook laterz
06:16 * eythian &
06:16 eythian later all
06:16 eythian tomorrow I'll figure out how to just get the count out, rather than the whole pile of data...
06:14 eythian Probably the've done some cleaning since I copied it.
06:14 eythian hmm, though either the data in prod is different or zebra isn't doing it right - my (old) copy of the database gets more results.
06:12 eythian well, biblio search really
06:12 eythian hmm, looks like my auth search should work
06:10 dcook Sprogs will get you everytime
06:10 dcook Ah, right
06:09 eythian as in, they have to go fetch kids etc
06:09 * dcook wonders who came up with that word
06:08 dcook You don't have deadlines?
06:07 eythian yeah, the others here all have to work to deadlines usually
06:07 * dcook should really really go home
06:07 dcook Do it up!
06:07 eythian not going to be here too much longer though, just want to get this query going so biblios can be counted.
06:07 dcook Although when I do have a kidlet, I'll have to be pretty precise due to care
06:06 dcook Yeah, I like later starts/stops myself
06:06 eythian I tend to arrive late though
06:06 dcook And dcook rarely leaves on time
06:06 eythian normally it doesn't as we're salaried.
06:06 * dcook thinks eythian is always still working when dcook leaves these days
06:06 dcook eythian: How does overtime work for you folks?
06:05 * dcook looks at the time
06:05 dcook That could certainly be interesting
06:05 eythian (no real reason you couldn't do that with zebra too I suppose)
06:05 eythian it would be neat to have a data inspector interface that lets you put together queries/reports using the search directly.
06:04 eythian you can also have a web interface to talk to it, given it's all JSON.
06:04 dcook Yeah, I think it would be nicer just using HTTP
06:04 dcook hehe
06:04 eythian my command line is interactive :)
06:04 dcook I suppose I like that yaz-client is interactive
06:04 dcook hehe
06:04 eythian good luck with that :)
06:03 dcook If I don't get any emails tomorrow, I might actually get through the rest of my email backlog...
06:03 eythian I do like how you can inspect what ES is storing with curl, it's much easier than yaz I think.
06:03 dcook :D
06:03 eythian well that's hopeful
06:03 eythian "an":[["227"],["228"],["229"],["230"],["177"],["39"]]
05:54 dcook All the brackets!
05:53 eythian { query => { filtered => { query => { match_all => {} }, filter => { term => { an => $authid } } } } }
05:53 eythian elasticsearch sure can get verbose sometimes:
05:30 dcook Totally
05:29 eythian if you were making big changes, it might look weird during the process, but that's better than downtime :)
05:27 dcook :D
05:26 eythian you could do a full reindex while everything was running and it'd be fine
05:26 dcook I think I follow ya
05:26 dcook Ahh, right
05:26 dcook online indexing?
05:24 eythian however, es has the ability do to online indexing, so in real life it wouldn't be a big deal.
05:23 eythian I think just cold caches in everything, it's sped up now
05:23 eythian actually, maybe it's something else.
05:23 dcook I guess so :/
05:23 eythian adding those mappings sure slows down the indexing
05:20 dcook :(
05:19 dcook Ahhh
05:19 eythian i.e. where the auth number is to be found
05:19 eythian the 4 digit numbers are 123$9
05:18 dcook I don't even know what I'm looking at :S
05:18 dcook :S
05:18 eythian http://paste.koha-community.org/2 <-- don't you love how clear MARC makes everything...
05:10 dcook But it looks like it is
05:10 dcook It probably shouldn't be a BIGINT
05:10 dcook At least in the DB I'm looking at
05:10 eythian maybe it's OK then
05:10 eythian huh
05:10 dcook According to auth_header, the authid is a BIGINT
05:09 dcook Hmm.. then again..
05:09 dcook They should always be numeric, but it's probably not safe to assume it
05:08 eythian actually, it's probably not safe to assume that
05:07 eythian will authids always be numeric?
04:59 dcook Ahh, that's neat
04:59 eythian that's where mappings live for elasticsearch
04:58 dcook SQL?
04:58 eythian lets see about doing some vim trickiness.
04:58 eythian I just have to extract them all for marc21, unimarc and nomarc, and turn that into sql :)
04:57 eythian yep
04:57 dcook There's a lot of stuff for authorities iirc
04:57 dcook Yeah, that defines what gets indexed, I believe
04:56 eythian erk, in marcdefs there's a proper brazillion of them
04:55 dcook Yeah ccl.properties is usually the place to go
04:54 dcook I thought it was satire, but nope... real.
04:54 dcook Sorry. Got distracted by political news.
04:54 eythian specifically
04:54 eythian Koha-Auth-Number Authority-Number
04:54 eythian an Authority-Number
04:54 eythian Authority-Number 1=Koha-Auth-Number
04:53 eythian which is actually probably correct
04:53 eythian apparently \W doesn't match the start of a line
04:53 eythian for some reason my grep didn't find it
04:53 eythian found it
04:53 eythian oh
04:51 eythian as in, when the biblio is indexed, where does it know what to put into 'an'
04:51 eythian where is 'an' defined?
04:51 eythian when counting the number of biblios that reference an authority, a search is done like 'an:<authid>'
04:50 dcook Yep. Finally getting around to having lunch. Sup?
04:50 eythian dcook: you around?
01:18 dcook AKA not much pushback :)
01:18 * dcook couldn't care less one way or another atm
01:18 dcook hehe
01:12 eythian plenty more where that came from!
01:12 mtj its good opinion tho, eythian :)
01:11 eythian right now there only seems to be my opinion on IRC :D
01:11 mtj i'll attempt to get some more opinion on irc
01:09 mtj ok ta
01:09 eythian I don't know if there'll be much pushback against that or not, probably not
01:09 mtj should i bother sending a patch with C4 code... that doesnt use dbix?
01:08 eythian probably recommended
01:08 eythian I don't know that there is one, but I think it'd be perfectly fine for C4 stuff to use it.
01:08 mtj is it required?
01:07 mtj hey #koha, whats the policy on new C4::* code using DBIX in Koha?
00:33 BobB ok, thank you mtj and gmcharlt