Time  Nick          Message
00:25 rangi         oh hey another idea
00:25 rangi         mtj: it's all good with xslt eh?
00:25 rangi         (i dont think it would touch it at all, but just in case)
00:52 mtj           rangi: it looks ok with xslt enabled
00:53 mtj           afaik, the TT caching is just 'compiling?' .tt files into .ttc files
00:53 mtj           .. so, doesnt touch any xslt stuff :)
01:15 rangi         yeah, it shouldn't
01:29 mtj           i spotted paul.p/joubu's discussion about moving the bi.marc/marcxml fields - the numbers are impressive
01:37 rangi         yeah it should make a difference
01:37 rangi         i tell you the single fastest way to speed up the OPAC
01:37 rangi         turn off local cover images
01:38 rangi         otherwise you get 20(or whatever number of results per page) hits on that script
01:38 rangi         from 33secs to 6
01:39 rangi         just from turning that off
01:39 rangi         of course it would be better to fix it, so it isnt that slow, and we could still do local cover
01:39 rangi         s
01:39 rangi         but if a library isnt actually using them, making sure that is off is a big improvement
01:40 rangi         (opac-images.pl)
01:40 rangi         sorry opac-image.pl
01:40 rangi         i wonder if there is a bug for that
01:50 mtj           ah yes... ive seen that in the logs
01:51 mtj           ..i thought someone was DOSing the server, at 1st
01:51 rangi         exactly
01:54 mtj           hmm, i guess plack would help with opac-images.pl startup?
01:54 rangi         a little
01:54 rangi         but not enough
02:00 rangi         i think one trick is to get them to be fetched via ajax, so it doesnt slow down the whole page
02:00 rangi         and then also, set headers like 304
02:01 rangi         so that if youve seen that cover, you dont get it again
02:01 rangi         etc
02:02 rangi         and because i think it touches bibloitems, it would get faster with the changes i think
02:02 rangi         part of the problem is because it has to check if a localcover exists
02:02 rangi         so if you have 30 local covers
02:02 rangi         and 5 million not
02:03 rangi         you still have to check every record in the results
02:04 rangi         so if we got it so that the results knew
02:04 rangi         ie, index if there is a local image, so we know
02:04 rangi         then we dont have to check again
02:05 rangi         we get the records to display the results, if we know it doesnt have a local cover right then, then why dont need to check
02:05 rangi         does that make sense?
02:06 wizzyrea      do you mean mysql index or zebra index or some other context of index like "mark it"
02:07 wizzyrea      (I'm sorry if that's a silly question)
02:07 rangi         well i think we get the full record object back for the restuls page, ie we get stuff from zebra, then we grab the actual data from the db and hand that to be displayed
02:07 rangi         no reason we couldnt check local image exists as part of that
02:07 rangi         then it could be just a 1:0
02:08 rangi         and we only put the img src if its 1
02:08 rangi         it'd still be slow if you had tons of them
02:08 rangi         but i think thats gonna be the exception
02:08 rangi         my guess is in my result sets you have 0 or maybe 1 or 2
02:08 rangi         which is better than doing 20 hits on opac-image
02:10 rangi         just rubber ducking :)
02:11 wizzyrea      yeah fair enough ^.^
02:40 mtj           yep, makes sense on both points
04:18 mtj           i like your ajax idea for opac-images.pl rangi
04:18 mtj           ... seems like a good quick/dirty solution, until a better fix happens
04:18 rangi         it still is gonna dos you
04:19 rangi         if a few people search at once, perhaps even faster than before
04:19 rangi         so i kinda hate it as an idea now hehehe
04:19 rangi         i think not fetching the image unless you know you have one is better ;)
04:20 mtj           yeah
04:25 mtj           ..you are making me question the idea now too :)
04:27 wizzyrea      it seems sad to me that we can get covers faster from amazon and that it's our own that cause problems :(
04:28 rangi         or at least getting them from amazon puts no load on our system
04:28 wizzyrea      that frowny face looks frownier than usual today
04:28 rangi         the annoying thing with load, it spirals fast
04:29 rangi         http://www.visawoap.com/burger-wellington/
04:29 rangi         interesting
04:29 wahanui       i guess interesting is sometimes good and sometimes bad
04:29 wizzyrea      the GP is not on that list! Shocking!
04:29 rangi         yeah
04:30 rangi         most of the ones i heard buzz about being great arent
04:30 rangi         it's funny how that happens
04:30 wizzyrea      tbf, I forgot to rate a couple of mine :/
04:30 rangi         *nod*
04:30 wizzyrea      it could be our fault :'/
04:38 mtj           wizzyrea: whats the GP?
04:38 wahanui       the GP is not on that list! Shocking!
04:38 wizzyrea      General Practitioner, corner of manners and willis
04:38 mtj           ..also NZ pj harvey tickets on sale tmrw, doods
04:39 wizzyrea      ohh
04:39 wizzyrea      I wish I had funds for that >.<
04:40 rangi         ditto
04:41 mtj           http://www.undertheradar.co.nz/news/11568/PJ-Harvey-Announces-Two-New-Zealand-Shows.utr
04:51 * wizzyrea    cries a bit
05:07 teja          koha software
05:46 * magnuse     waves
06:38 reiveune      hello
07:08 alex_a        Bonjour
07:09 morgane       hi #koha
07:09 fridolin      hie
07:38 magnuse       @wunder boo
07:38 huginn        magnuse: The current temperature in Bodo, Norway is 10.0°C (9:20 AM CEST on August 31, 2016). Conditions: Light Rain. Humidity: 94%. Dew Point: 9.0°C. Pressure: 29.62 in 1003 hPa (Steady).
09:10 LibraryClaire morning #koha
09:10 cait          morning LibraryClaire
09:11 LibraryClaire hi cait :)
09:11 cait          :)
09:13 eythian       hi LibraryClaire & cait
09:13 LibraryClaire hey eythian :)
09:23 janPasi       is it ok to change koha to marc mappings for publishercode in biblioitems table or will that break something?
09:24 janPasi       we would like to use 028b here instead of the koha default 260b
09:24 cait          it won't break things - but it won't apply immediately to records already in the system
09:24 Joubu         should be ok
09:24 cait          thhere is a job you can run to do that
09:24 janPasi       cait: yeah, i know, it needs running the script to make it happen for the old records
09:25 janPasi       i can't recall the name of the script, but i think i can find it again :D
09:26 janPasi       thanks, we'll try that :)
09:26 cait          REbuild...Bibllio sth
09:26 janPasi       yeah
09:26 cait          i think it's in misc or bin depending on your installation
09:26 cait          i was looking at it yesterday
09:28 janPasi       yeah, i think it's in misc somewere :D
09:28 Joubu         misc/batchRebuildBiblioTables.pl
09:28 janPasi       Joubu: cool, thanks :)
09:28 Joubu         and misc/batchRebuildItemsTables.pl I guess
11:33 marcelr       hi #koha
11:47 oleonard      Hi #koha
11:47 marcelr       hi oleopard
11:56 magnuse       hi o'leopard
12:03 eythian       olé, pard
12:04 eythian       @wunder ams
12:04 huginn        eythian: The current temperature in Schiphol, Badhoevedorp, Netherlands is 25.0°C (1:52 PM CEST on August 31, 2016). Conditions: Partly Cloudy. Humidity: 49%. Dew Point: 13.0°C. Pressure: 30.09 in 1019 hPa (Steady).
12:05 magnuse       @wunder boo
12:05 huginn        magnuse: The current temperature in Bodo, Norway is 13.0°C (1:50 PM CEST on August 31, 2016). Conditions: Light Rain. Humidity: 94%. Dew Point: 12.0°C. Pressure: 29.56 in 1001 hPa (Steady).
12:05 magnuse       oh, it's possible to delete the AnonymousPatron?
12:05 magnuse       shouldn't that be kind of easy to block?
12:05 eythian       yeah, but you can't delete the HeadlessHorseman
12:06 oleonard      It would be nice to be able to block deletion of any selected patron
12:06 magnuse       ooh, there is a syspref for that?
12:06 magnuse       ah donotdelete checkbox on each patron? sounds like a good idea, yeah
12:11 LibraryClaire @wunder LCY
12:11 huginn        LibraryClaire: The current temperature in London City, United Kingdom is 22.0°C (12:50 PM BST on August 31, 2016). Conditions: Partly Cloudy. Humidity: 50%. Dew Point: 11.0°C. Pressure: 30.12 in 1020 hPa (Steady).
12:25 magnuse       nengard!
12:25 wahanui       somebody said nengard was fast
12:25 nengard       morning magnuse
12:25 nengard       how are you?? :)
12:25 magnuse       good!
12:26 magnuse       and yerself? missing us yet? ;-)
12:30 tcohen        hola #koha
12:31 tcohen        @later tell rangi I got the point on the first read. I answered that only so it is in the logs too
12:31 huginn        tcohen: The operation succeeded.
12:31 nengard       magnuse, I'm here every day :) nothing to miss :) hehe
12:31 nengard       but I am loving my new job :)
12:32 LibraryClaire hi tcohen :)
12:32 tcohen        hi LibraryClaire
12:33 magnuse       nengard: that is good to hear :-)
12:34 eythian       wahanui: australia is <reply>http://www.sbs.com.au/comedy/article/2016/08/31/australia-beautiful-see-hawk-literally-throw-snake-innocent-family
12:34 wahanui       ...but australia is pretty great in a lot of ways, but they are woefully deficient in the pierogi department...|https://twitter.com/BR3NDA/status/410647513633284096/photo/1...
12:34 eythian       wahanui: australia is also <reply>http://www.sbs.com.au/comedy/article/2016/08/31/australia-beautiful-see-hawk-literally-throw-snake-innocent-family
12:34 wahanui       okay, eythian.
12:38 oleonard      As opposed to the time a hawk threw a snake at a family who totally deserved it.
12:38 oleonard      (which is so common in Australia as to not make the news)
12:39 jcamins       oleonard: good point.
12:44 LibraryClaire terrifying
12:51 cait          @later tell drojf ping!
12:51 huginn        cait: The operation succeeded.
12:55 tcohen        I don't think blou's comment about documentation is correct
12:59 Joubu         the entry points are hard to find
12:59 tcohen        probably, but we are all around to answer about those
13:00 Joubu         yep yep
13:00 tcohen        I accept we should think about improving documentation
13:00 * tcohen      is not on denial
13:00 Joubu         and patches are waiting for signoff for a while (before the ground has been pushed)
13:01 Joubu         have to go, I will be afk for the next of the week. I had to get some rest or submit a very big patchset, I referred to postpone the submission :D
13:01 Joubu         prefered
13:02 tcohen        hehe
13:02 tcohen        have a nice whatever you're gonna do
13:02 mtompset      Greetings, #koha.
13:03 mtompset      Sorry, tcohen, I didn't get to it yet.
13:03 mtompset      I got distracted by a bigger problem!
13:03 mtompset      check in
13:03 wahanui       i heard check in was different though.
13:03 mtompset      check out
13:03 mtompset      restart mysql server
13:03 mtompset      Oops... got that backwards...
13:03 magnuse       have fun Joubu!
13:04 mtompset      check out, check in, restart mysql server, check out, check in... old_issues bug!
13:04 mtompset      InnoDb forgets its highest auto_increment after server restart!
13:05 mtompset      Any ideas on the best way to solve it?
13:07 marcelr       Joubu: still looking at bug 15758
13:07 huginn        04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=15758 enhancement, P5 - low, ---, jonathan.druart, Signed Off , Move the C4::Branch related code to Koha::Libraries - part 4
13:07 marcelr       almost done
13:16 oleonard      marcelr++
13:16 marcelr       oleonard++ for testing it
13:33 barton        good morning #koha!
13:33 oleonard      Hi barton
13:33 nengard       morning barton
13:33 marcelr       hi barton
13:34 barton        I'm looking at bug 16581... I think there's still a number of un-answered questions there...
13:34 huginn        04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=16581 enhancement, P5 - low, ---, gmcharlt, RESOLVED WONTFIX, ICU tokenization bug in idzebra-2.0 2.0.59-1
13:34 marcelr       mtompset: nasty problem, any chance of more issues cause of that?
13:35 mtompset      anything with an old_!
13:35 marcelr       barton: i did not understand why it was closed either
13:35 mtompset      because if we are moving from one to the other, we'll hit overlap and lose history.
13:36 marcelr       well an argument to stop with those deleted tables
13:36 mtompset      A strong one.
13:36 barton        so here are the things that I'm wondering about: 1) is it possible to require the use of the indexdata repo? 2) what version of idzebra-2.0 *should* we be using?
13:37 mtompset      1) why not? It's a good idea, 2) the current one -- I think something 62?
13:52 oleonard      khall around?
13:55 nengard       oh good he's ignoring you too :) hehehe
14:00 oleonard      @later tell khall Bug 16903 says it has been pushed to master but I don't think it really has.
14:00 huginn        oleonard: The operation succeeded.
14:01 mtompset      Oh, and sorry for not doing it sooner. Greetings, oleonard nengard marcelr barton magnuse mario JesseM amyk :)
14:02 mtompset      Apologies. Greetings, JoshB too. :)
14:02 marcelr       bye
14:02 mtompset      Bye. :)
14:02 JesseM        Hello
14:02 JoshB         hello
14:02 amyk          greatings all! :-)
14:02 amyk          oops
14:02 JesseM        oleonard: khall is on vacation
14:02 amyk          greetings! not awake yet ;-)
14:03 tcohen        hi amyk JesseM
14:03 amyk          hi nengard!
14:03 amyk          hi tcohen
14:03 JesseM        Hi tcohen
14:03 nengard       yeah!! it's morning love fest :) hehe
14:03 nengard       hi all!
14:03 nengard       I miss this
14:03 amyk          :-)
14:03 nengard       at the new job we're all on so many calls that we hardly talk in IRC
14:03 oleonard      calls--
14:04 barton        oleonard, nengard khall is out; he'll be back Friday.
14:04 nengard       oh, okay :)
14:04 nengard       and yes oleonard
14:07 mtompset      OH! I'm not awake either. nengard!
14:08 mtompset      How is the new job going?
14:13 nengard       a-ok
14:13 nengard       lovign my team and the work
14:16 oleonard      And the hat, nengard? Don't foget the hat!
14:16 * oleonard    assumes nengard wears it 9-5 every day now
14:17 * oleonard    spends the rest of his day feverishly working on designs for a Koha hat
14:18 nengard       ha
14:18 nengard       that as kind of a lie ... it wasn't my hat - mine is still on the way
14:19 nengard       was not as
14:19 eythian       barton: requiring the use of another repository is not trivial, and not something I would expect drojf to be comfortable automating.
14:19 eythian       I suspect it is wontfix because it's not feasible to fix from Koha's POV.
14:22 barton        eythian: I see... so the proper avenue would probably be to file bugs with debian and indexdata to get idzebra-2.0 upgraded in jessie?
14:24 eythian       https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=777515 <-- like this :)
14:24 eythian       which unfortunately got no response.
14:24 eythian       it's not something indexdata has control over
14:25 eythian       My suggestion would be to have a warning in the staff client about page if a particular zebra version is in use that links to a wiki page with some resolutions, for example maybe a debian backport could be arranged.
14:26 eythian       Or maybe just prodding the maintainer to see what he suggests
14:40 tcohen        anyone to talk about the REST api and permissions?
14:42 * kidclamp    waves
14:50 oleonard      kidclamp: Are you sick of fielding questions from me about Coce yet?
14:53 kidclamp      not yet
14:54 kidclamp      what's the question oleonard
14:54 oleonard      I'm still getting mixed http/https results from Amazon: https://search.myacpl.org/cgi-bin/koha/opac-search.pl?idx=kw&q=dogs&offset=40
14:54 oleonard      Some covers from https://images-na.ssl-images-amazon.com, some from http://ecx.images-amazon.com
14:55 oleonard      Do you know anything about that?
14:55 oleonard      Or perhaps fredericd does?
14:55 kidclamp      my guess, without looking into code, is that image links previously cached are http
14:55 kidclamp      and new ones are pulled using https
14:55 kidclamp      so as the links expire the issue will resolve itself
14:56 kidclamp      or I can try to flush the amazon storage, but I don't know off top of my head
14:57 oleonard      Out of curiosity, do you know what the lifetime of links is?
14:57 mtompset      oleonard: ecx doesn't exist in git. Is it in a system preference?
14:57 oleonard      mtompset: https://github.com/fredericd/coce
14:58 kidclamp      86400000 seconds
14:58 mtompset      Okay... external git... :)
14:58 kidclamp      1000 days
14:59 oleonard      Wow that's a long time
14:59 kidclamp      I think it was to counter the fact that initial fetch doesn't load images
14:59 kidclamp      when you do a new search coce gets the linkss, but doesn't display them until the next access
15:00 kidclamp      so we keep them a long time to increase results displayed
15:04 kidclamp      looking into how I can clear the amazon links oleonard, but not overfly familiar with redis
15:05 oleonard      It's not a high priority kidclamp, but I would like to clear up those warnings.
15:06 kidclamp      can you open a ticket oleonard - that will keep it on my radart
15:06 oleonard      Sure, thanks.
15:23 morgane       bye #koha
16:09 reiveune      bye
16:52 kidclamp      around oleonard?
16:52 oleonard      Yes
16:53 kidclamp      bug 14060 - I don't think we should warn on closing datepicker if no date is selected
16:53 huginn        04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=14060 normal, P5 - low, ---, jonathan.druart, Signed Off , Remove readonly on date inputs
16:56 kidclamp      just wanted your thought
16:57 oleonard      Agreed. I didn't notice that when I tested.
16:58 kidclamp      Okay, FQA for now, but works well elsewise
18:54 CrispyBran    Question for the room: How does one determine how Koha and when Koha calculates overdue and long overdue statuses?  Is it on the fly, or is it done by a cron job?
18:59 eythian       Pretty sure it's a cron job
19:00 CrispyBran    Sorry, app issues.  :/
19:00 CrispyBran    If anyone responded to my question, I missed it.  :(
19:04 mtompset      CrispyBran: It's a cronjob.
19:05 mtompset      It's in /etc/cron*/koha-common I put *, because I don't know which one has it. :)
19:05 mtompset      Have a great day (24 hour period), #koha.
19:45 * andreashm   waves
19:51 * cait        waves back
20:35 Mauricio_BR   Hola a todos buenas tardes. alguien habla español?
20:41 Mauricio_BR   please i need to know where is stored the 856 marc-tag in the koha database schema...
20:43 eythian       @marc 856
20:43 huginn        eythian: The information needed to locate and access an electronic resource. The field may be used in a bibliographic record for a resource when that resource or a subset of it is available electronically. In addition, it may be used to locate and access an electronic version of a non-electronic resource described in the bibliographic record or a related electronic resource. (Repeatable) (1 more message)
20:43 eythian       @more
20:43 huginn        eythian: [a,b,c,d,f,h,i,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x,y,z,2,3,6,8]
20:45 kidclamp      Mauricio_BR 856$u is likely mapped to / stored in biblioitems.uri in the DB
20:45 kidclamp      make that url
20:46 Mauricio_BR   thank you kodclamp, the thing is that I stored the url-cover in 856a
20:46 Mauricio_BR   but i do not know where the 856a marc-tag is stored in the database schema
20:47 kidclamp      if not mapped to a koha field it is stored in the full record /  marcxml field in biblioitems
20:50 Mauricio_BR   and can i map the 856a as a koha field after import a .mrc file?
20:57 kidclamp      you can alter the mapping, but then you will need to run a script to rebuild the db from the biblios
20:58 Mauricio_BR   mmm ok i see,  thank you
21:06 tcohen        hi kiwis
21:08 wizzyrea      hi tcohen :)
21:14 tcohen        hey wizzyrea
21:14 wahanui       i think wizzyrea is so dumb at rewrite rules
21:14 tcohen        wahanui: i don't belive you
21:14 wahanui       tcohen: what?
21:15 wizzyrea      still mostly true
21:15 wizzyrea      ^.^
21:15 tcohen        wahanui: what you heard
21:15 wahanui       tcohen: wish i knew
21:15 tcohen        no one understands rewrite rules
21:15 tcohen        or at least everyone has to review the docs each time
21:15 tcohen        i'm sure
21:17 wizzyrea      hehe
21:17 wizzyrea      I"m still not sure I solved the problem that precipitated that original statement
21:18 wizzyrea      this thread on developer problems is highly interesting
21:19 tcohen        which one?
21:19 wahanui       which one is t?
21:20 wizzyrea      the one about koha-zebra-jessie
21:20 wizzyrea      and "why is this so haaaard"
21:20 tcohen        hahaha
21:20 wizzyrea      (I was having a moan about this yesterday, independent of this)
21:22 wizzyrea      it is reasonably hard to test things like elasticsearch
21:22 wizzyrea      and plack. And any dev that touches lots of things
21:23 wizzyrea      (the "advanced cataloguing interface" comes to mind)
21:24 tcohen        yes, they are hard
21:25 tcohen        and it is not the rest of the devs' fault
21:25 tcohen        only for making it easy to test plack i re-did the kohadevbox
21:26 tcohen        I struggle more with the fact that devs are done in environments that don't match production, than with difficult bugs
21:27 wizzyrea      *nod* the lack of a really good production scale dataset is a serious problem
21:27 tcohen        dataset, library versions, previous tries cruft, etc
21:27 wizzyrea      thousands of borrowers, tens of thousands of records
21:28 wizzyrea      hundreds of thousands of issues/old_issues
21:28 wizzyrea      hundreds of reserves
21:28 wizzyrea      nonlinear upgrades
21:28 wizzyrea      and on and on
21:29 tcohen        if we had a complete test suite
21:29 tcohen        that should happen on jenkins
21:30 tcohen        and is completely do-able
21:30 wizzyrea      a good place to start maybe, would be to take say the members tests and build it up to not do like, one issue and return, but hundreds
21:30 wizzyrea      or tens at the very least
21:31 tcohen        you need to first think 'what question would you like the tests to answer'
21:32 wizzyrea      I want a tool off the side that basically simulates a day at the library
21:33 wizzyrea      users added, users removed, biblios/items added/removed, reserves added/filled/waiting/deleted/suspended, issues and returns
21:34 wizzyrea      note I'm not asking you to write that.
21:34 wizzyrea      ^.^
21:34 tcohen        hehe
21:35 tcohen        you want t/db_dependent/selenium/basic_workflow.t
21:35 tcohen        at least extending it
21:37 wizzyrea      oh true, I didn't realise that was there.
21:37 wizzyrea      Oh and that's another problem
21:37 tcohen        ?
21:38 wizzyrea      keeping on top of every new thing. :)
21:38 tcohen        ah
21:38 tcohen        he
21:38 tcohen        it needs to be around, daily
21:38 tcohen        otherwise you loose track
21:38 tcohen        that's true
21:39 tcohen        that was what I tried to reply to blou
21:39 wizzyrea      even I just simply don't have enough spoons for all the things that I should be on top of.
21:40 * tcohen      knows about that
22:30 rangi         hey hey
22:30 wahanui       niihau, rangi
22:31 rangi         guess who moved into number 2 pushing liblime down in the companies list today
22:32 rangi         http://git.koha-community.org/stats/koha-master/authors.html#commits_by_domains
22:33 tcohen        :-D
22:33 tcohen        catalyst++
22:33 tcohen        google is close too :-P
22:33 rangi         hehe yeah, if you combine ? and gmail .. the unknowns are 2nd :)
22:34 rangi         how cool is it that a library myacpl is 4th though
22:34 tcohen        oleonard is awesome :-D
22:35 rangi         but also, 2 directors now, who have let him do that
22:35 rangi         thats pretty amazing
22:38 tcohen        true
23:47 tcohen        bye #koha
23:48 rangi         cya tcohen