Time  Nick        Message
23:43 andreashm   time for bed. bye #koha
23:32 cait        wizzyrea++
23:26 cait        i will suggest that then
23:26 cait        ok - sorry late here :)
23:25 wizzyrea    it's not important.
23:25 wizzyrea    it's ok
23:25 cait        sorry, don't understand
23:24 wizzyrea    should be the rule.
23:24 wizzyrea    verb or don't verb
23:24 wizzyrea    ^ would be more consistent I think
23:24 cait        we could also say  [Send|Don't send blind copy (BCC) to logged in user when sending serial or acquisitions claims notices
23:23 wizzyrea    (or something like)
23:23 wizzyrea    BCC logged in user when sending serial or acquisitions claims notices
23:22 cait        maybe we could say claims notices for serial issues or acquisition orders?
23:22 wizzyrea    or BCC, if you feel that's clear
23:22 cait        :)
23:22 cait        hm yeah :9
23:22 wizzyrea    Blind copy logged in user when sending claims notices?
23:21 cait        the claims are done manually
23:21 wizzyrea    right
23:21 cait        because it's for the emails send out automatically
23:21 wizzyrea    right k
23:21 cait        ah yeah there you specify the email address
23:21 wizzyrea    no I meant the overduenoticebcc
23:21 wizzyrea    see that's confusing.
23:21 cait        it will use the one of the user doing the claiming
23:21 cait        nope
23:21 wizzyrea    yeah but the email addy is specified?
23:21 cait        hm not really helpful
23:21 cait        OverdueNoticeBcc (that we should rename and remove Overdue..) reads: Send all notices as a BCC to this email address
23:20 wizzyrea    Right
23:20 cait        i think ClaimIssues is the internal name of it - - and not sure about how clear active user is
23:19 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10076 enhancement, P5 - low, ---, charles.farmer, Signed Off , Add Bcc syspref for claimacquisition and clamissues
23:19 cait        this is bug 10076
23:19 cait        dcook: yeah - there are direct trains form there to here :)
23:19 wizzyrea    hm.
23:19 wizzyrea    oh
23:19 dcook       cait: I recall seeing Konstanz on the train maps in Zurich :)
23:19 wizzyrea    yes?
23:19 * cait      summons wizzyrea
23:18 cait        Send all claim notices in serials and acquisitions as a BCC to the user sending them ?
23:18 andreashm   cait: yeah. but we want to see a little of the country when we go down now.
23:17 cait        Send a copy in bcc of the ClaimAcquisition and ClaimIssues letters to the active user
23:17 cait        hm can someone help rephrase this a bit?
23:15 cait        quite a trip :)
23:15 cait        :)
23:14 andreashm   We'll be fling down to Zurich and then taking the train to Sion. Then another train to fly back from Geneva.
23:14 cait        you got very close to here then :)
23:14 dcook       I think I've only been to Geneva, Zurich, and Chur
23:13 dcook       Ah, I've never been out that way
23:13 andreashm   guess I am tired.
23:13 andreashm   don't know where that question mark came from
23:12 andreashm   sion, east of lausanne
23:12 dcook       hehe
23:12 andreashm   dcook: lots of fun thing to look forward too? but the lack of sleep is not one of them!
23:12 dcook       Oh? Whereabouts?
23:12 * andreashm is going on vacation on wednesday. down to switzerland to check things out.
23:12 dcook       Only a few months more until my spawn are ready to be born
23:12 dcook       andreashm: I'll probably be doing the same thing next year
23:11 andreashm   =)
23:11 andreashm   dcook: pfff
23:11 * dcook     thought that's where a person didn't do work
23:11 dcook       Maybe "vacation" is one of those words that gets lost in translation :p
23:11 andreashm   cait: vacation, smacation
23:11 * cait      has the excuse of being on vacation :)
23:11 andreashm   I'm awake with my six month daugther from time to time at night. I have answered some of your e-mails around 4 am or so. might explain some of my strange writing, haha
23:10 dcook       Only 10am here
23:10 * dcook     looks cait's way
23:10 dcook       Must be a European thing
23:10 dcook       I sometimes think you never sleep ;)
23:10 andreashm   00.10
23:10 dcook       Indeed. I imagine it must be rather late for you
23:09 andreashm   nice to be around here at the same time! usually only see your activity in the logs. =)
23:09 dcook       heya :)
23:09 andreashm   hi dcook btw
23:09 dcook       Ahh
23:09 andreashm   dcook: the ones to Devel
23:08 andreashm   but now I'm done reading... debating whether or not I should go to bed.
23:08 dcook       I was just going to pen another one I think ;)
23:08 dcook       Do you?
23:07 andreashm   =)
23:07 andreashm   but I saw that I had unread e-mails from dcook about the metadata-table so I figured I could stick around for a while.
23:07 cait        yeah, you go tno good excuse this time
23:07 andreashm   cait: sleeping actually. as should I to be honest.
23:03 cait        ruby? :)
23:03 * cait      waves
23:02 * andreashm waves
22:37 Francesca   sup
22:37 Francesca   hey cait :)
22:37 cait        hi Francesca
22:36 Francesca   hello
22:10 dcook       Ah, only cait now
22:10 dcook       hey cait and cait1 :)
22:10 cait        my evil clone ;)
22:10 cait        heh
22:09 bag         cait++
22:09 bag         phew I was a bit worried about that cait1 character - glad cait is back
22:09 * cait1     nick cait
22:09 cait1       ah the disconnect
22:07 bag         wizzyrea++
22:07 bag         heya cait1
22:07 cait1       hi dcook :)
22:07 cait1       hi bag :)
22:06 wahanui     hi, bag
22:06 bag         hello
22:04 dcook       hehe
21:56 wizzyrea    lol what did I do
21:56 wizzyrea    STOPIT
21:55 dcook       wizzyrea++
21:55 pianohacker wizzyrea++
21:55 wizzyrea    wot
21:55 dac         ^
21:55 cait        wizzyrea++ just because :)
21:29 cait        but nothing the user can see
21:28 cait        so they are easier to translate and update
21:28 nengard     crossing that one off :)
21:28 cait        nengard: it only simplifies the sql files for the web installer
21:28 nengard     cool
21:28 cait        nengard: nope
21:27 huginn      04Bug 10963: enhancement, P5 - low, ---, bgkriegel, Pushed to Master , Simplified creation of MARC21 sample frameworks
21:27 nengard     hi all - does this patch http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10963 change the way you work with frameworks - like do I need to update the manual at all?
21:15 cait        morning rangi
21:12 rangi       morning
19:30 tcohen      bye
17:26 pianohacker cait: yar, gracias
17:26 cait        does that make sense?
17:26 cait        but we usually don't remove the lines i think
17:26 cait        if it's a complicated one, sometimes it can be better to ask for another
17:25 cait        it usually depends on the nature of the rebase
17:25 pianohacker cait: yar. Do we drop to Needs signoff after a rebase?
17:24 cait        pianohacker: have you taken a look at the hisotry?
17:24 cait        hm i think they rebaed and reset
17:19 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=8753 enhancement, P1 - high, ---, charles.farmer, Needs Signoff , Add forgot password link to OPAC
17:19 pianohacker hmm. Should bug 8753 be at "Signed off" instead of "Needs signoff" from mveron's signoffs, or am I reading that wrong
17:16 huginn      cait: The operation succeeded.
17:16 cait        @later tell Joubu - could we add checks for readding C4::Dates magic to the QA script for a while?
17:12 drojf       i won't be the first person to get lost in Search.pm and die in there. i have seen skeletons in there. and i won't be the last one either.
17:11 drojf       i'm trying to draw a picture of where the marc data goes after retrieval und how often it gets converted into what. i don't think that is even possible
17:03 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11096 major, P5 - low, ---, gmcharlt, CLOSED FIXED, Koha cannot retrieve big records from Zebra
17:03 barton      Right... bug 11096 was pushed to master in 3.16; this is occurring on either 3.18 or 3.20. I don't think that we're running into the 99 999 byte marc limit. Everything is MARCXML. We have plenty of records larger than 99 999 bytes which we can search correctly... but once they get up around 1MB, we start having issues.
17:03 tcohen      so, GRS-1 + usmarc => GOOD
17:03 tcohen      barton: if I remember correctly, the problem we found was: DOM + usmarc format for exchanging => break
17:00 tcohen      i don't think so, we found that situation during kohacon13, and 11096 solution had to do with that
17:00 barton      tcohen: was there a bug for that?
16:59 tcohen      ^^^ that would mean exporting big records from Koha shouldn't be a problem
16:59 tcohen      i recall the problem was that the DOM indexing/retrieval code had trouble with big records
16:58 cait        my bet would be some failing conversion somewhere
16:58 tcohen      MARC::Record doesn't care about big records
16:58 barton      oh, no they're way too big to export as usmarc.
16:57 cait        can you still export those records as usmarc?
16:57 cait        i was thinking about 11096
16:56 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11096 major, P5 - low, ---, gmcharlt, CLOSED FIXED, Koha cannot retrieve big records from Zebra
16:56 barton      cait:  yeah, looks like bug 11096 is still addressing the 99 999 byte hard limit with USMARC. This is different. There seems to be a soft limit on the size of MARCXML records ... searches start failing up around 1000 attached items... that's specifically why I mentioned Ian Walls' comment -- I think that he touched on something that never got addressed: "Further research has shown that my issue is not with ISO 2709, but the implicit file siz
16:51 cait        i think
16:50 cait        grs-1 is not so much the problem, but usmarc vs xml
16:50 cait        well not really grs-1
16:49 barton      cait: yeah that one mostly addresses GRS-1 / MARC.
16:49 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=11096 major, P5 - low, ---, gmcharlt, CLOSED FIXED, Koha cannot retrieve big records from Zebra
16:49 tcohen      bug 11096
16:49 cait        i can try to find it in a bit
16:49 barton      Yeah, pianohacker thought there might be somethning recent.
16:48 cait        might know - but the one you are lookin git is ancient - probably not the one
16:48 cait        micht now
16:48 cait        tcohen w
16:48 cait        there was another bug
16:48 barton      (will look)
16:48 barton      3.18.x ... not sure which release.
16:47 wahanui     which version is recommended?
16:47 cait        barton: which version?
16:46 huginn      04Bug 2453: critical, P3, ---, gmcharlt, CLOSED FIXED, (very) large biblio/item handling
16:46 barton      hey, I'm trying to track down an issue with search failing on very large bib records... we're using DOM, and I've verified that we're using the -x option during re-indexing, so that doesn't seem to be the problem. I suspect that I'm seeing something like what Ian Walls reported here: http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=2453#c4
16:37 barton      morning #koha!
16:17 huginn      New commit(s) kohagit: DBRev 3.23.00.000: 2016 Year of the Monkey <http://git.koha-community.org/gitweb/?p=koha.git;a=commitdiff;h=3dfb6252faf6abec64cfc50a090aa0ce7d272115>
16:11 reiveune    bye
15:43 druthb      brr
15:43 huginn      Dyrcona: The current temperature in Shawsheen Area, Andover, Massachusetts is 0.5°C (10:35 AM EST on November 30, 2015). Conditions: Clear. Humidity: 66%. Dew Point: -5.0°C. Windchill: 1.0°C. Pressure: 30.56 in 1035 hPa (Falling).
15:43 Dyrcona     @wunder 01845
15:42 huginn      druthb: The current temperature in Blueberry Farm, Hockley, Texas is 10.7°C (9:42 AM CST on November 30, 2015). Conditions: Overcast. Humidity: 95%. Dew Point: 10.0°C. Pressure: 30.05 in 1018 hPa (Rising).
15:42 druthb      @wunder 77447
15:19 huginn      cait: The current temperature in Konstanz, Germany is 10.0°C (4:00 PM CET on November 30, 2015). Conditions: Partly Cloudy. Humidity: 61%. Dew Point: 5.0°C. Pressure: 30.32 in 1027 hPa (Falling).
15:19 cait        @wunder Konstanz
15:17 andreashm   mmm fun fun
15:13 huginn      andreashm: The current temperature in Stockholm Bromma, Sweden is 2.0°C (3:50 PM CET on November 30, 2015). Conditions: Clear. Humidity: 81%. Dew Point: -1.0°C. Windchill: -3.0°C. Pressure: 29.06 in 984 hPa (Steady).
15:13 andreashm   @weather stockholm, sweden
15:12 bag         :)
15:12 * cait      waves at bag
15:11 huginn      liw: The current temperature in Acequion, Torrevieja, Spain is 16.7°C (4:11 PM CET on November 30, 2015). Conditions: Clear. Humidity: 72%. Dew Point: 12.0°C. Pressure: 30.51 in 1033 hPa (Falling).
15:11 liw         @weather torrevieja, spain
15:11 huginn      liw: The current temperature in Viherlaakso, Espoo, Finland is 4.9°C (5:08 PM EET on November 30, 2015). Conditions: Mostly Cloudy. Humidity: 99%. Dew Point: 5.0°C. Windchill: 5.0°C. Pressure: 28.88 in 978 hPa (Steady).
15:11 liw         @weather espoo, finland
15:10 bag         oh 28 let’s trade tcohen
15:09 huginn      tcohen: The current temperature in Bo Altos de San Martin - NW, Cordoba city, Cordoba City, Argentina is 28.0°C (12:04 PM ART on November 30, 2015). Conditions: Scattered Clouds. Humidity: 31%. Dew Point: 9.0°C. Pressure: 30.12 in 1020 hPa (Falling).
15:09 tcohen      @weather cordoba, argentina
15:07 huginn      bag: The current temperature in Portland - West Hills, Portland, Oregon is -2.7°C (7:07 AM PST on November 30, 2015). Conditions: Clear. Humidity: 81%. Dew Point: -6.0°C. Windchill: -5.0°C. Pressure: 30.18 in 1022 hPa (Rising).
15:07 bag         @weather portland or
15:06 bag         :)
15:05 tcohen      bag: 您好
15:05 tcohen      gmcharlt++ # back from holidays and ready to work for the community
15:05 bag         morning
15:00 cait        maybe a quick mail to the mailing list? I think some people are waiting and will be happy to have a date
14:57 * cait      waves at gmcharlt
14:56 drojf       no worries. and congratulations to your new community job ;)
14:56 drojf       hi gmcharlt
14:55 gmcharlt_   apologies for the delay, but I want to make sure I get it right before I tell folks to go at them
14:55 gmcharlt_   back from the US holidays, and wanted to give an update on the 3.22 packages -- namely, that I'm planning to release them by Thursday
14:54 tcohen      hi gmcharlt_ !
14:54 gmcharlt_   howdy folks
14:43 drojf       heh
14:40 liw         drojf, that's a variant of the classic programmer's snafu of editing and running differnet instances of a program (which I do regularly enough that you should time your backup runs based on it)
14:37 drojf       *one
14:37 drojf       did you know its hard to debug stuff if you work in one instance but refresh a different on in your browser?
14:33 andreashm   tcohen: yeah. it's never easy! =)
14:33 andreashm   tcohen: you might be right. using an extra table is the old-school way of doing it (but one we know works). an ajax call might be suitable for loan status etc.., but wouldn't it be the same problem for matching? I.e. that new updates fast after the other might end up racing the Ajax call (or index using Zebra/ES). What we're thinking is matching against 001, possible 035 (others probably want other fields)
14:32 tcohen      we are suffering from performance issues in one hand, and proposing multiple more joins on the other :-D
14:29 tcohen      that's all I'm saying
14:29 tcohen      so, maybe we can be more creative isntead of duplicating work
14:29 tcohen      some data (loan status) could be retrieved asynchronously through an AJAX call instead of locking the UI
14:28 tcohen      instead of polling every x second
14:28 tcohen      maybe the solution is to make indexing asynchronously work differently
14:27 andreashm   tcohen: the problem with using Zebra, or ES for that matter, for matching is that it is to slow. Take OAI for instance, that updates every 2-3 seconds. That would be problematic if not having special fields available in a separate table.
14:06 tcohen      andreashm: I agree maybe having a way to define special fields to have available... but I don't like the argument "if zebra is not available I would like to..."
14:03 cait        ... but i sitll need to read ... and take care of the luandry, brb
14:03 cait        maybe configurable
14:02 cait        0xx fields
14:02 cait        hm i think identifiers would be enouhg
14:01 andreashm   just noted that David in his e-mail asked if the whole record should be broken down into such a structure, but that seems excessive to me.
14:01 andreashm   tcohen: I think that sounds like a great idea too.
13:59 tcohen      andreashm: I'd agree with a metadata_records table, storing the serialized data, and, well, all the fields Koha::MetadataRecord has
13:58 cait        i plan to read - jus ta bit later
13:58 cait        thx for the summary :)
13:57 andreashm   cait: in short, an added table to store some metadata identifiers. to pro would be that it would allow fast matching without the need of Zebra/ES, source of the record can be displayed there etc. the con is that it's an added extra table.
13:57 tcohen      so a MARC record could be split into, say 120 rows on a table?
13:57 tcohen      sorry, wrong window
13:56 andreashm   tcohen: huh? not sure what that means...
13:55 andreashm   cait: bah! =)
13:54 cait        andreashm: it loooked too long for right now :)
13:54 tcohen      andreashm: it was a 2 minutes readlas encargo asi
13:53 andreashm   cait: you should go read! interesting stuff!
13:53 andreashm   tcohen: wow. that sounds a bit scary! =)
13:53 tcohen      I actually read part of it while driving (semaphore)
13:53 andreashm   this table would be kind of small I think. but it would need to be for everything in the database, more or less.
13:52 * cait      sitll has to
13:52 tcohen      i just read his discussion about the metadata table
13:52 andreashm   tcohen: any of the disadvantages David mentioned, or some other reason?
13:48 andreashm   why is that?
13:48 tcohen      I mean the big table with metadata fields
13:47 tcohen      andreashm: I don't like it
13:45 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10662 new feature, P3, ---, dcook, In Discussion , Build OAI-PMH Harvesting Client
13:45 drojf       oh i just saw that the plack settings were reset. so maybe its a little faster actually
13:45 andreashm   any idea how this would tie in with the proposed added table for metadata as suggested on bug 10662/David's e-mail to the Devel list? Good or bad?
13:44 * andreashm has been reading back on the discussion on search speed, metadata and filters. interesting stuff!
13:44 tcohen      drojf: the conclusion won't be to ditch DOM, but "have usmarc be the exchange format"
13:43 drojf       heh i will try that later :D
13:43 * tcohen    bets GRS-1 will do 2.3 secs
13:42 drojf       yes its using dom
13:42 tcohen      go try GRS-1 ;-)
13:42 tcohen      drojf: is 3.16 using DOM?
13:40 cait        you got way more patience with stuff like that than me :)
13:32 drojf       3.16.15 with plack 10 seconds for 1000 hits. so yes, still twice as fast compared to 3.22 with tweaks
13:30 drojf       but i think its enough for this investigation run. time to have a little weekend. oh what it's monday already? XD
13:29 drojf       *it
13:29 drojf       no i think that is not true. but ut was faster on 3.16 still
13:28 drojf       so i am 1/3 faster now than regular 3.22. which is still probably only half as fast as 3.16 with plack
13:27 drojf       30+ is the bad one
13:27 drojf       i would have expected both performance gains cancelling each other out. but i won't complain if its better
13:27 cait        value
13:27 cait        what's the bad balue?
13:26 drojf       with 15263 and reverting 11842 i get < 20 seconds for 1000 results
13:20 tcohen      gmcharlt_: morning
13:19 magnuse     drojf++
13:19 magnuse     yeah, improvements around search and zebra and all that jazz should be very welcome
13:19 cait        also drojf++ for investigating performance issues :)
13:17 cait        i'd day - go for it too :)
13:16 huginn      magnuse: The current temperature in Bodo, Norway is 2.0°C (1:50 PM CET on November 30, 2015). Conditions: Scattered Clouds. Humidity: 93%. Dew Point: 1.0°C. Windchill: -2.0°C. Pressure: 28.88 in 978 hPa (Steady).
13:16 magnuse     @wunder boo
13:15 tcohen      heh
13:07 huginn      drojf: The current temperature in Berlin Tegel, Germany is 7.0°C (1:50 PM CET on November 30, 2015). Conditions: Mostly Cloudy. Humidity: 66%. Dew Point: 1.0°C. Windchill: 3.0°C. Pressure: 29.89 in 1012 hPa (Steady).
13:07 drojf       @wunder berlin, germany
13:04 drojf       tcohen: maybe tcohen1 can help
13:02 tcohen      already have some stuff i promised for this week
13:02 tcohen      i can start next week
13:02 magnuse     just do it
13:02 drojf       let's do it now! :D
12:58 tcohen      and is queued, etc
12:58 tcohen      because de indexing process is asynchronous
12:58 tcohen      it is cheaper to prepare the data so zebra has it, than calculate it on rendering
12:57 tcohen      or it should
12:57 tcohen      that information is already in zebra
12:57 tcohen      also, with the current indexer, we shouldn't re-calculate loan status, etc for rendering
12:56 tcohen      get MARC(XML) data => have filters for MARC(XML) data => period
12:56 drojf       that sounds very good to me
12:56 tcohen      get JSON data => have filters for JSON data => period
12:56 tcohen      and this might get worse with JSON data from ES
12:55 tcohen      if you think about it, we are serializing/deserializing things more than 3 times
12:55 tcohen      but move to Koha::Fitler ofr the mid/long-term
12:55 tcohen      so my opinion regarding the syspref retrieval for each record is that we should optimize that in a short term
12:54 tcohen      and for retrieval (which is OK, but seems really under performant in the current state of the way we do things)
12:54 tcohen      both for indexing (which is OK)
12:54 tcohen      regarding zebra, in 3.16 we made DOM setups exchange XML between Perl and Zebra
12:53 tcohen      and start abstracting things a bit for a future multi-metadata schema support
12:53 tcohen      that's from a perspective in which we re-organize the code
12:52 tcohen      and pass them through a pipe of Koha::Filter-s using the Koha::RecordProcessor thing
12:52 tcohen      so, put search results in Koha::MetadataRecord object
12:51 tcohen      drojf: my plan with https://github.com/tomascohen/koha/tree/xslt_filtering was to avoid useless filtering of the retrieved records
12:48 drojf       there is probably more redundant stuff in there, iam just starting
12:47 drojf       *think
12:46 drojf       regarding zebra
12:46 drojf       tcohen: what do you tink is the problem exactly?
12:45 huginn      04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=15263 enhancement, P5 - low, ---, mirko, NEW , XSLT display fetches sysprefs for every result processed
12:45 drojf       tcohen: bug 15263 appears to speed up opac search on the raspi by 1/5. from ~30 to ~24 seconds for 1000 results
12:41 drojf       hi tcohen
12:40 tcohen      drojf: the problem is XML retrieval/processing from Zebra
12:39 tcohen      morning
10:43 huginn      drojf: The operation succeeded.
10:43 drojf       @later tell wizzyrea it seems like pbuilder does not like the default distribution set by build-git-snapshot (squeeze-dev) either. unstable set via parameter should work. maybe relevant for you too
10:40 drojf       hi #koha
10:10 petter      http://debian.koha-community.org/koha/dists/
10:10 petter      Any news about debian packages for 3.22 ?
08:25 wahanui     what's up, gaetan_B
08:25 gaetan_B    hello
08:24 Joubu       Bonjour magnuse :)
08:23 magnuse     bono? sorry about that
08:23 magnuse     bonojur Joubu
08:21 Joubu       hi #koha
08:12 huginn      magnuse: The current temperature in Bodo, Norway is 3.0°C (8:50 AM CET on November 30, 2015). Conditions: Mostly Cloudy. Humidity: 87%. Dew Point: 1.0°C. Windchill: -2.0°C. Pressure: 28.82 in 976 hPa (Steady).
08:12 magnuse     @wunder boo
08:11 * magnuse   waves
07:35 reiveune    hello
07:22 fridolin    hie there
06:50 eythian     wizzyrea: making it a dependency would be an appropriate solution
05:32 dcook       Probably saved way more time by optimizing it now rather than trying to optimize it then
05:32 dcook       Spent some time doing profiling the code, analyzing the queries, and found an optimization in the end which speeds it up 4-10x.
05:32 dcook       I recently designed a script which comprehensively analyzes data, but... it was taking a really long time
05:31 dcook       Thinking about how he advised me ages ago to avoid premature optimization
05:31 dcook       Hmm surely eythian must be up now
02:09 wizzyrea    i18n
02:09 wizzyrea    with encoded characters
02:09 wizzyrea    for generating PDF's
02:09 ibeardslee  why is a particular font required?
02:09 wizzyrea    that may be a naive assessment.
02:08 wizzyrea    but if we made it a dependency, we wouldn't have that particular problem anymore
02:08 wizzyrea    you have to install it in pbuilder
02:08 wizzyrea    and it makes Koha tests fail
02:08 wizzyrea    is that when you go to build packages for Koha, pbuilder doesn't have it
02:07 ibeardslee  so can't assume that
02:07 wizzyrea    the reason I'm asking
02:07 ibeardslee  but not ttf-dejavu
02:06 ibeardslee  .. fonts-dejavu-core is installed on a couple of our internal debian servers
02:06 wahanui     hmmm... the package is 'perldoc'
02:06 wizzyrea    the package
02:05 ibeardslee  the font or the package?
01:58 huginn      Francesca: The current temperature in Wellington, New Zealand is 17.0°C (2:30 PM NZDT on November 30, 2015). Conditions: Clear. Humidity: 77%. Dew Point: 13.0°C. Pressure: 30.33 in 1027 hPa (Steady).
01:58 Francesca   @wunder wlg
01:53 wizzyrea    are we assuming that ttf-dejavu is installed on debian systems?
01:51 wizzyrea    soooo