Time Nick Message
00:00 jcamins Useful might be over-selling it.
00:00 jcamins "Useful" implies "would be listened to."
00:13 rangi :)
00:47 wizzyrea The fox is <reply> http://www.youtube.com/watch?v=jofNR_WkoCE
00:48 eythian wahanui: stonehenge is <reply>http://youtu.be/mbyzgeee2mg
00:48 wahanui OK, eythian.
00:48 eythian I think that one's better anyway
00:49 wizzyrea the fox relevant because well. Kids.
00:50 wizzyrea that is completely silly
00:54 mtompset Greetings, #koha.
01:10 rambutan hi mtompset
01:26 jcamins Hey, does anyone have Koha + MySQL on a test server to do a test for me real quick?
01:26 wizzyrea sure, sup
01:26 jcamins I'm trying to confirm that deleting a biblio from the staff client updates the timestamp to the date deleted in deletedbiblio.
01:27 jcamins The test plan is as follows: create biblio, note biblio.timestamp from MySQL client. Delete biblio using "Delete record" button in staff client. Check deletedbiblio.timestamp for that biblionumber.
01:27 wizzyrea yep sec
01:29 * dcook thinks that works these days but didn't back in the day
01:29 * dcook has a bunch of badd deletedbiblio timestamps from 3.2
01:30 dcook I haven't tested though right now though, so I'd wait for liz on this one..
01:30 wizzyrea 2013-09-30 14:30:16 < record I just deleted timestamp from deletedbiblio
01:31 jcamins Woohoo!
01:31 jcamins Thanks.
01:31 wizzyrea that was edit -> delete items, then edit -> delete record
01:31 wizzyrea that is what you meant right?
01:32 jcamins Yup.
01:32 wizzyrea cool
01:32 jcamins It'd make me happy if I could run a report every day for all records touched in the last day.
01:33 rambutan jcamins: that would be useful information?
01:34 wizzyrea yep, I just did it again, timestamp looks just fine in deletedbiblio
01:34 jcamins Exceedingly useful.
01:34 jcamins It would allow for keeping a discovery layer synchronized without the trouble of setting up an actual push notification for changes.
01:37 dcook You could use OAI :p
01:37 rambutan well, hummm
01:37 dcook That might not fit your purposes though
01:39 jcamins dcook: I may be mistaken, but my understanding of OAI is that it doesn't really allow for incremental updating.
01:40 dcook It depends on how you implement it
01:40 dcook The spec has selective harvesting built-in
01:40 dcook It just depends on how often you run your harvests and if you update your "from" date with each harvest
01:40 dcook And whether your repository honours that "from" date properly
01:40 dcook At the moment, Koha doesn't
01:40 dcook brb
01:42 jcamins dcook: nor does Koha report deleted records, making the use of OAI with Koha exceedingly problematic at best.
01:47 dcook I think there's a bug for reporting deleted records at some point :p
01:49 dcook bug 3206
01:49 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=3206 enhancement, P5 - low, ---, gmcharlt, NEW , OAI repositry deleted record support
01:49 dcook I guess it's not an active one though :p
01:50 jcamins Doesn't seem to be.
01:50 dcook I might work on it someday, depending on how that union catalogue idea pans out
01:51 dcook bug 10824 is actually just awaiting sign off
01:51 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10824 major, P5 - low, ---, dcook, Needs Signoff , OAI-PMH repository/server not handling time in "from" argument
01:51 dcook I should actually revise that to say "from" and "until" parameters...
02:10 dcook I am having an issue where I get a software error after harvesting 34,000 records from one Koha instance...
02:10 dcook I imagine 680 HTTP requests might upset any server...
02:11 dcook Although there are a few seconds between each request, me thinks
02:39 mtj peeps - shall i log a bug to delete .the greybox/GreyBox_v5_5.zip file(s) in Koha?
02:40 mtj ...they are unneeded yes? (and duplicated per lang and theme?)
02:40 * dcook has no idea about Greybox
02:41 rangi i cant see why we would need a zip file
02:48 mtompset Hmmm....
02:56 * dcook ponders
02:56 dcook OAI-PMH repositories are supposed to use UTC timestamps
02:56 dcook Koha does not...
02:57 dcook Nor does Koha acting as a OAI server..
03:00 eythian bgkriegel: should bug 9611 have its status changed, or are you still working on it?
03:00 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=9611 enhancement, P5 - low, ---, srdjan, Needs Signoff , Changing the password hashing algorithm from MD5 to more secure Bcrypt
03:01 bgkriegel in a minit
03:01 bgkriegel i'm writing my last comment
03:01 wizzyrea your last comment EVAR?! NOOOOOO
03:01 bgkriegel hehe
03:02 bgkriegel :)
03:10 bgkriegel done ! :)
03:13 mtompset EVAR?
03:13 mtj a random pkg question… when installing a new koha package - where are the db updates logged too?
03:15 wizzyrea /var/log/apt/term.log
03:15 wizzyrea (at least it was this morning when I looked at it)
03:16 mtj hmm, thats blank for me
03:19 mtj perhaps that DB-upgrade info only gets sent to the screen, and not a file - i might be imagining it?
03:20 wizzyrea hm it was in the one I looked at this morning
03:20 wizzyrea maybe it's different per apt config
03:20 wizzyrea you installed the package using apt-get eh?
03:23 mtj aah, i just noticed my 'version' was higher than my package
03:23 mtj (prolly testing a master build before)
03:24 wizzyrea oh so did you install it with dpkg?
03:24 mtj yeah, sorry - dpkg -i ./foo.deb
03:24 wizzyrea ah okies :) then it won't be in the apt logs. :)
03:24 wizzyrea but it might be in /var/log/dpkg.log
03:25 mtj yeah, there was indeed some info in there
03:29 mtj cool, was my high version number, causing the problem :)
03:41 bgkriegel bye
04:04 mtj hmm, i've just spotted a subtle bug in updatedb.pl - for the version checking code…
04:04 mtj if ( C4::Context->preference("Version") < TransformToNum($DBversion) ) {
04:04 mtj seems to work better than...
04:05 mtj if ( CheckVersion($DBversion) ) {
04:05 mtj .
04:05 eythian I just found a situation my csvtomarc script can't handle. Embedded newlines in a title...
04:06 eythian mtj: in that case I guess CheckVersion should be fixed.
04:06 mtj wow, i cant blame you for not expecting that eythian
04:06 eythian liberty is not the most sane software out there.
04:13 mtj the mentioned bug seems to be that CheckVersion() skips updatedb blocks with $VERSION values that are more than 3dp
04:14 mtj so, for these 3 updates - the middle one gets skipped...
04:14 mtj 3.12.05.000
04:14 mtj 3.12.05.0001
04:14 mtj 3.12.05.001
04:15 mtj ...which is a gotcha if you are attempting to use 'local' version numbers, for your updates
04:15 cait eythian++
04:15 eythian mtj: 3.12.05.0001 == 3.12.05.001
04:16 eythian in version number arithmetic
04:16 eythian the dots aren't decimal places, they're simply breaking up sections, and it's an integer in each section.
04:20 mtj hmm, so whats a better string to give the 2nd 'local' update?
04:21 eythian are you supposed to add another section or something?
04:21 eythian well hang on
04:21 eythian you have anohter section
04:21 eythian so 3.12.05.002
04:22 eythian ah no, that's internal
04:22 eythian I think you can just chuck another section on.
04:22 eythian or increment the already existing minor number, but that's probably less ideal.
04:24 mtj yeah, well i thought i *was* adding another section on :)
04:24 eythian you need a . for that :)
04:25 mtj well, i thought the dots were simply ignored by koha (as you suggested?)
04:26 mtj ahh, actually you didnt quite say that before… did you :)
04:26 eythian nope
04:29 mtj fwiw - i did initially experiment with an extra '.' but got some weird error, so assumed i was doing it wrong :/
04:29 eythian I'm not sure if it's right, but it's how it should be done.
04:29 eythian oh, hi cait.
04:30 mtj morning cait :)
04:43 dcook 15 minutes until The Signal!
04:43 * dcook also waves to cait
04:44 mtj thanks for the info eythian - i'll poke about some more re: version numbers
04:45 eythian cool. let me know what you work out, too
04:45 dcook Oh noes...it's Monday...
04:45 eythian not for too much longer though
04:45 * dcook would also be curious as to what you find, mtj
04:46 dcook Well, still Sunday in Vancouver..barely Monday in Toronto..
04:46 eythian so behind the times.
04:46 eythian wahanui: new zealand
04:46 wahanui http://www.buzzfeed.com/daves4/hardest-parts-about-living-in-new-zealand
04:47 eythian wahanui: new zealand
04:47 wahanui http://www.buzzfeed.com/daves4/hardest-parts-about-living-in-new-zealand
04:47 eythian wahanui: new zealand is also in the future. We have jetpacks.
04:47 wahanui okay, eythian.
04:48 dcook hehe
04:49 dcook I should probably listen to local radio but...I like the host and the show in Canada :/
04:54 dcook argh..
04:59 dcook wtf Koha...
04:59 wahanui hmmm... koha is a free software ils see http://koha-community.org for more info
05:02 * dcook wonders if anyone uses Koha as an OAI-PMH server, because if they do...it's not working anywhere near correctly
05:03 dcook Or maybe every one of my patches fixes one thing and breaks another..
05:05 cait dcook: we had someone testing it for feeding their discovery interface
05:05 cait they noted some problems about it
05:05 cait the big problem was that there was no option to include item information
05:05 cait but they were also not so happy about the missing deleted records
05:05 dcook Those are optional things though
05:05 cait and there was something about paging giving you an error when there were no more records
05:05 dcook These are fundamental core problems
05:05 dcook Yep
05:06 dcook Actually, no, I haven't seen that
05:06 dcook The resumption tokens don't appear to work when using dates with times though
05:07 dcook Including item information would be easy
05:07 dcook Showing deleted records wouldn't be too hard either
05:08 cait can only tell you the problems we ran into
05:08 dcook Sorry
05:08 dcook I appreciate you telling me that :)
05:08 * dcook is just unreasonably frustrated and didn't mean to direct his frustration your way
05:09 cait it's ok ;)
05:12 dcook I guess I'll fix the resumption tokens...
05:13 * dcook feels like he's slowly becoming the oai guru...
05:13 mtj hmm, there is a bug around the version number stuff - its not just me being stupid
05:13 dcook Maybe because no one else cares about oai :p
05:19 dcook Now that I look at everything again...I think I finally understand...
05:19 dcook Wait...no, it still wouldn't have worked...
05:20 dcook Maybe bug 10824 isn't relevant if we consider Koha to just handle a granularity of YYYY-MM-DD...
05:20 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10824 major, P5 - low, ---, dcook, Needs Signoff , OAI-PMH repository/server not handling time in "from" argument
05:20 dcook But the resumption token thing is still busted..
05:24 dcook huzzah. Changing the colons to slashes fixes the resumption token issue..
05:26 cait mtj: but local database updates are evil...;)
05:26 cait dcook: yay!
05:30 mtj sure, but also unavoidable cait
05:30 cait hm it depends
05:31 * dcook tries to avoid local database changes these days
05:31 cait we have been able to avoid it so far - but we are very cautious about local changes
05:31 dcook Most of the time these days, I tell people that I'm happy to make changes to the database, but only after they've been vetted and pushed by the community first
05:32 dcook Not always possible though
05:34 cait args.
05:34 cait i really have to clean up my addressbook
05:35 cait [off] me still has limblime addresses in there for some people... email.
05:37 cait [off] all gone... i hope
05:39 mtj so, are any koha devs/companies actually using some system for 'local' version numbers for DB updates?
05:40 cait we don't
05:40 cait and i have to run and import patron data :)
05:40 cait bbl
05:41 dcook mtj: Maybe catalyst?
05:42 dcook We track our db changes separately
05:49 mtj theres a bug in the TransformToNum() sub, that stops 'local' versions from being evaluated correctly
05:50 mtj # remove the 3 last . to have a Perl number
05:50 mtj $version =~ s/(.*\..*)\.(.*)\.(.*)/$1$2$3/;
05:51 mtj ….and its been about since version 3.00.00 :p
05:53 dcook :/
05:55 dcook bug 10974
05:55 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10974 normal, P5 - low, ---, dcook, Needs Signoff , OAI-PMH Resumption Tokens Do Not Handle Time
05:55 dcook If anyone wants a super easy sign off...
06:02 * mtompset grumbles about reindexing on 512MB of memory with no swap space and 53K+ records failing because of some malformed tag, and requiring using a script that inserts everything into the zebraqueue.
06:05 dcook Doesn't sound like fun, mtompset
06:05 dcook Although you'll be glad when you've cleaned up the marcxml :)
06:05 mtompset It's not.
06:05 mtompset Here's the thing.
06:06 mtompset It all indexes fine running this script.
06:06 mtompset It does not reindex nicely otherwise.
06:06 mtompset I should, at some point in time, try to determine the exact record it is barfing on.
06:06 * dcook nods
06:07 dcook Especially if you're using the XSLTs
06:07 dcook I've had sneaky bad bibs get in and it's not pretty
06:26 * magnuse waves
06:29 dcook hey ya magnuse
06:31 magnuse hiya dcook
06:34 Oak magnuse
06:34 Oak kia ora #koha
06:34 dcook hey Oak
06:37 magnuse Oak
06:37 * magnuse is too slow
06:37 Oak :)
06:37 Oak hey dcook
06:41 marcelr hi #koha
06:41 marcelr kf++ #qa
06:42 Oak kf++
06:42 magnuse kf++
06:44 * dcook succumbs to peer pressure
06:44 dcook kf++
06:45 magnuse hehe
06:46 marcelr hi magnuse dcook and oak
06:46 dcook hey ya marcelr :)
06:47 magnuse hiya marcelr
06:48 Oak Ahoy marcelr
06:48 gaetan_B hello
06:48 wahanui hi, gaetan_B
06:48 Oak \o
06:57 magnuse tee hee http://www.nerdmeritbadges.com/
06:57 magnuse @wunder boo
06:57 huginn magnuse: The current temperature in Bodo, Norway is 8.0°C (8:50 AM CEST on September 30, 2013). Conditions: Mostly Cloudy. Humidity: 66%. Dew Point: 2.0°C. Windchill: 4.0°C. Pressure: 30.12 in 1020 hPa (Steady).
06:57 magnuse @wunder marseille
06:57 huginn magnuse: The current temperature in Marseille, France is 16.0°C (8:30 AM CEST on September 30, 2013). Conditions: Fog. Humidity: 94%. Dew Point: 15.0°C. Pressure: 29.65 in 1004 hPa (Steady).
06:58 j_wright what's the state of Pg support in Koha?
06:58 magnuse j_wright: there is none
06:58 rangi not entirely true
06:59 magnuse oops, listen to rangi, not me
06:59 rangi with bug 8798
06:59 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=8798 enhancement, P3, ---, elliott, In Discussion , Add the use of DBIx::Class
06:59 rangi you can deploy to postgresql, and it mostly works
07:00 rangi once thats pushed we can clean out the last of the mysqlisms, there are very few left
07:00 dcook \o/
07:00 rangi it currently works fine on mariadb too
07:00 magnuse \o/
07:00 rangi pg support is nice, but its not a massive priority
07:01 j_wright well clearing out the mysqlisms would enable more than Pg support
07:04 rangi yep, there arent many
07:04 rangi but again, its not a massive priority
07:04 * dcook wonders what the massive priorities are
07:05 dcook Not that I disagree. I'm just curious :p
07:05 magnuse plack
07:05 magnuse dbic
07:05 magnuse stable master
07:05 rangi yep
07:05 magnuse is what i'd say ;-)
07:06 rangi id agree with that
07:06 magnuse yay! :-)
07:06 j_wright would be sad to be using DBIC and still be Mysql-bound
07:07 rangi we are transitioning to maria
07:08 rangi if we get pg support as an aside, all well and good, but its not one of my priorities
07:08 rangi because for the library user, it gains them nothing
07:08 reiveune hello
07:08 rangi and that should be our priority
07:08 dcook ^^
07:08 dcook rangi++
07:08 samueld hi everybody
07:08 dcook hey samueld
07:08 magnuse bonjour samueld and reiveune
07:09 reiveune \o/
07:09 rangi hi samueld and reiveune
07:09 samueld hi magnuse
07:10 j_wright pity mariadb doesn't work optimally unless you have ucontext
07:10 samueld is it normal that when i create a categorycode in koha, it is written automatically in upper case instead of lower case?
07:12 rangi yes, its supposed to be samueld
07:13 magnuse there is some js that transforms it to uppercase automagically, i think
07:13 samueld example, i write 'employee" and it is transformed in "EMPLOYEE", that's for why i've some problems to connect koha to the ldap
07:14 magnuse ouch
07:15 magnuse yeah, i was just thinking a couple days ago that forcing uppercase seemed a little heavy handed
07:15 samueld is it a bug?
07:17 dcook Time for me to run. Night/day all.
07:17 magnuse samueld: i would think the js was implemented as a feature, not by accident ;-)
07:17 samueld ok, so, it can be changed
07:41 magnuse cait++ bgkriegel++ for bug 10965 and bug 10969
07:41 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10965 major, P5 - low, ---, bgkriegel, Passed QA , Sample itemtypes can't load on new install
07:41 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10969 normal, P5 - low, ---, katrin.fischer, Signed Off , Fix sample itemtypes for translated installers
07:41 magnuse mysqlf-- for breaking sample item types
07:41 magnuse gah
07:41 magnuse myself--
07:44 marcelr interesting typo, magnuse: no fan of mysql ?
07:44 marcelr )
07:44 marcelr :)
07:44 magnuse :-)
07:45 magnuse mysqld is not too bad, but i positively hate mysqlf ;-)
11:07 paxed any dev got some time to help me resolve a problem? after install, can't go to maintenance.pl because i get "Can't use an undefined value as an ARRAY reference at /usr/local/lib/perl/5.14.2/DBI.pm line 2058."
11:19 paxed ergh. right, forgot port 8080, derp.
11:20 paxed that should really complain something else if the db is "missing" ...)
12:17 * magnuse waves again
12:20 oleonard Hi #koha
12:20 magnuse kia ora oleonard
12:25 magnuse ooh, shiny new buttons on the cataloguing page!
12:37 magnuse is it just me, or is marcflavour missing from the new, alphabetically sorted sysprefs.sql?
12:38 marcelr magnuse: isn't that a specific choice in webinstaller?
12:39 marcelr it should not be in that file
12:39 magnuse ah yes, that could be it
12:40 magnuse thanks marcelr
12:42 magnuse but at least marcflavour will never be "usmarc", right?
12:42 marcelr marc21
12:44 marcelr uppercase
12:44 marcelr normarc is one of the big three :)
12:46 magnuse w00t!
12:46 marcelr holmarc never made it ;)
12:47 magnuse aww...
12:47 magnuse Bug 10975
12:47 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10975 normal, P5 - low, ---, gmcharlt, NEW , addbiblio.pl checks if marcflavour equals "usmarc"
12:50 marcelr somewhere a title is not filled in the template
12:50 marcelr good catch
12:50 marcelr apparently, it does not hurt much
12:51 magnuse been like that since 2007, i think :-)
12:54 marcelr magnuse: the situation seems to corrected by a later line: $template->param( title => $record->title() ) if ( $record ne "-1" );
12:54 marcelr in the code more below the first line
12:55 marcelr without the usmarc check
12:55 kivilahtio__ Hi I was migrating items to Koha and realized that the itemnumber was not preserved. I used C4::Items::_koha_new_item() to migrate items. Would it be acceptable if I extended that function to push the itemnumber to DB as well if it is present?
12:55 kivilahtio__ just installing my git environment and figuring stuff out
12:56 marcelr magnuse: no sorry there is an exit in between
12:56 kivilahtio__ or maybe there is a more stylish alternative for migrating items using the Koha API?
12:56 marcelr so in a specific case it should not appear
12:57 magnuse kivilahtio__: why would you need or want to preserve the itemnumbers?
12:57 kivilahtio__ magnuse: so we can target our circulations and holds to our items
12:58 kivilahtio__ and fines
12:58 kivilahtio__ we migrate it all
12:59 magnuse kivilahtio__: use barcodes for the matching instead?
12:59 kivilahtio__ magnuse: our barcodes are not 100% reliable
12:59 magnuse marcelr: it's all yours if you want to fix it ;-)
12:59 kivilahtio__ magnuse: but that is a good idea
13:00 marcelr magnuse: after you please.. i have so much to fix already :)
13:00 magnuse marcelr: me too :-/
13:07 kf hm
13:08 kf not sure if the webinstaller creates the syspref or only updates it
13:08 marcelr hi kf
13:09 marcelr $installer->set_marcflavour_syspref($marcflavour); ?
13:09 kf guess you checked :)
13:09 kf sorry was afk until now
13:09 kf and hi marcelr
13:10 marcelr my $request = $self->{'dbh'}->prepare( "INSERT IGNORE INTO `systempreferences` (variable,value,explanation,options,type) VALUES('marcflavour','$marc_cleaned','Define global MARC flavor (MARC21, UNIMARC or NORMARC) used for character encoding','MARC21|UNIMARC|NORMARC','Choice');" );
13:11 marcelr kf: i did now
13:11 marcelr code from Installer.pm
13:11 marcelr so that is fine
13:14 kf yep
13:15 magnuse yay!
13:15 magnuse in other news, i would hesitate to call addbiblio.pl "fine"...
13:16 marcelr fine finer finest
13:21 marcelr cya
13:30 mtompset 12 hours reindexing... and still not done. Oh the pain!
13:34 smeagol Hi all, does anyone know if localuse syspref counts items that are checked in to koha to trigger a HOLD to waiting status? thanks, as always
13:51 Barrc Hey all - are odd Koha version numbers still considered unstable/dev releases? I ask cause I just ran a package install and then koha-install-log say "KOHA_INSTALLED_VERSION=3.12.05.000"
13:53 jcamins Barrc: the second number is the one that marks stable/development.
13:53 Barrc Of course it does........half asleep here, thanks!
14:26 banana wizzyrea++ -> "we would only include other Open Source projects"
14:31 kivilahtio__ hey I cannot duplicate the field 001 in "Administration" -> "Koha to Marc mapping". I need to explicitly define the field 001 to biblio.biblionumber, biblioitem.biblionumber and biblioitem.biblioitemnumber but I can apparently only assign it once
14:31 kivilahtio__ I would like to preserve our old database id sitting in $001
14:33 jcamins kivilahtio__: you should not be mapping biblioitem.biblionumber, and biblioitemnumber != biblioitem.biblionumber
14:34 kivilahtio__ jcamins: ERROR: Adding biblio failed: No biblionumber tag for framework ""
14:35 jcamins Personally, I wouldn't do that.
14:35 kivilahtio__ jcamins: hmm, I thought I understood that much from reading the bulmarcimport.pl code
14:35 jcamins You're also going to have to use a custom Zebra configuration.
14:35 kivilahtio__ default id is in $999, but we have it in $001
14:35 kivilahtio__ jcamins: wow I don't want to do that :D
14:36 jcamins That's what I thought. That's why I leave biblionumber in 999$c and biblioitemnumber in 999$d.
14:36 kivilahtio__ so if I want to keep our existing dabase id's I just need to sucker it up and migrate them in the $999?
14:36 rambuten jcamins: can you kick rambutan for me?
14:36 jcamins You can try that. I'm actually not sure how well it will work.
14:36 jcamins Done.
14:37 rambuten strangely it won't let me assume that nick now
14:37 kivilahtio__ rambuten: maybe it is still reserved?
14:37 kivilahtio__ rambuten: maybe wait a bit?
14:37 jcamins rambuten: rambutan may not have left the network yet.
14:37 jcamins rambuten: if it's registered, you can use /ghost rambutan, I think.
14:37 rambuten somebody has copyrighted it, but they never seem to be online
14:37 jcamins Oh.
14:38 rambuten I never have problems getting the nick, but it's no big deal
14:40 rambuten jcamins: you didn't kick ban, did you?
14:40 kf kivilahtio__: messing with the ids will bring you all kinds of headache later
14:40 jcamins rambuten: nope.
14:41 kf kivilahtio__: we never preserve old ids in migrations, but uor 001 are unique from a union catalog. We only store the ids from former systems in some fields so looking them up would be possible
14:43 kivilahtio__ kf: lol I cant set my mappings back to the defaults! Ok I'll try the $999 route
14:43 kivilahtio__ but this might be really hard when merging biblios and items
14:44 kivilahtio__ I just save my new mappings but nothing happens...
14:44 jcamins kivilahtio__: how often do you merge bibs and items?
14:45 kivilahtio__ jcamins: well it takes time to rewite the scripts. I was hoping to reuse some Koha parts which use the biblionumber.
14:45 kivilahtio__ jcamins: ok I just realized my logical mistake
14:47 kivilahtio__ bah ui misunderstanding
15:12 rambuten didn't the old koha-community have a search field on the main page
15:12 rambuten ?
15:12 jcamins I think so, yeah.
15:14 rambuten kinda annoying to have to do a google search to find something on a given site.
15:14 jcamins We could probably add it back below "Community Resources."
15:15 rambuten apparently no presentation schedule yet?
15:16 jcamins For the conference?
15:16 jcamins There is.
15:17 rambuten it's not exactly jumping out at me on any of the pages I see
15:17 jcamins Yeah, the KohaCon13 pages are kind of unusable. There's one somewhere, though.
15:18 rambuten http://wiki.koha-community.org/wiki/KohaCon13_Program
15:18 jcamins That's the one.
15:19 kf hackfest?
15:19 wahanui hackfest is awesome! Group motivated koha hacking is the best koha hacking :)
15:19 kf I will try to send out an email about the hackfest planning tomorrow or so I think
15:19 kf hackfest13 is http://wiki.koha-community.org/wiki/Kohacon13/Hackfest
15:19 kf hackfest13?
15:19 wahanui hackfest13 is http://wiki.koha-community.org/wiki/Kohacon13/Hackfest
15:20 kf jcamins: i think oyu are around the first day of the hackfest?
15:20 jcamins I am, yes.
15:20 kf :)
15:20 kf you could talk about the awesome query parser
15:21 kf or...
15:21 rambuten I'm gratified to see that coffee break sessions are will represented. I wonder if I could offer to bring/have purchased some better-than-your-average-bear coffee, like Eight O' Clock
15:21 rambuten will/well
15:31 jeff jcamins: Would you consider yourself a good person whose brain I could pick about Koha and Plack?
15:31 kivilahtio__ jeff :)
15:31 jeff jcamins: for starters, the "coding for plack" idea on the hackfest wiki page has me wondering what the current state of plack is in koha?
15:32 jeff kivilahtio__: greetings!
15:35 jcamins jeff: probably one of the best you're going to get, anyway.
15:35 rambuten no speed talks @kohacon13 ?
15:36 jcamins jeff: most things work under Plack.
15:36 jeff jcamins: http://wiki.koha-community.org/wiki/Plack and the referenced bug 7172 both seem somewhat dormant. Is that because the work's done, or because there hasn't been much progress to note lately?
15:36 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=7172 major, P1 - high, ---, paul.poulain, NEW , Omnibus for Plack variable scoping problems
15:37 jcamins At this point I think we've addressed all the variable scoping issues, mostly with terrible bandaids.
15:39 jcamins The only things that absolutely don't work under Plack are database export, and the progress bar for importing records.
15:41 jcamins However, Koha under Plack is still a bit brittle.
15:46 jeff jcamins: I understand that Plack has support for running (most, with some assumptions/fixes) CGI-style scripts under PSGI using Plack::App::CGIBin. Are you aware of anything that does the reverse -- takes a script that speaks PSGI and allows it to run under a CGI environment?
15:46 jeff jcamins: That last one isn't Koha-specific at all, sorry. :-)
15:49 jcamins jeff: I am not.
15:50 kf bye all
15:50 jeff Thanks. It was worth a shot. I had this dream of writing a script that talks PSGI and then in the README says "if you must run this under a CGI environment, do XYZ"
15:51 jcamins There may be something.
15:51 jcamins I know very little about native-PSGI applications.
15:52 kivilahtio__ hey there is something wrong with the bulkmarcimport.pl script. There seems to be no way of pushing the old biblio id to the biblionumber! The marc mappings don't do anything since the database accessor functions don't even touch the biblionumber column! in C4::Biblio::_koha_add_biblio()
15:53 kivilahtio__ biblionumber is picked fine up until that function, where nothing is done upon the biblionumber? I mean there should be a way to map the biblionumber?
15:54 kivilahtio__ anyway, I'm going home
15:54 kivilahtio__ Issues is that the biblionumber is correctly brought up to the database accessor function but it is not stored in the DB, but a new one is generated from the database primary key sequence
15:55 kf kivilahtio__: there is probably not
15:55 j_wright jeff: Plack::Handler::CGI
15:56 kf kivilahtio__: koha gives the bilbionumber, it's not supposed that you bring your own
15:56 kf kivilahtio__: it's not made to do what you want
15:56 kivilahtio__ kf: But that is what the bulkmarcimport uses from "Koha to MARC mapping"
15:56 kf the 999 are special
15:57 kf see bug 6113
15:57 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=6113 enhancement, P5 - low, ---, gmcharlt, ASSIGNED , enhancement to keep previous ids
15:57 kf i am still not sure it's a good idea
15:58 kivilahtio__ kf: if I can't have a biblionumber set in biblio-table, how can I link my items, when they have the biblionumber in them pointing to the parent record?
15:58 j_wright jeff: XYZ is, rename app.psgi to app.cgi, change the shebang line to #!/usr/bin/env plackup or the three lines from the POD
15:58 jcamins kivilahtio__: what do you mean by "link your items"?
15:59 jcamins That sounds like a one-time task.
15:59 kf you can have a biblionumber, of course you need one, but we let koha set them.
15:59 kivilahtio__ kf: I cant migrate holdings with the marc records
15:59 kivilahtio__ kf: just too much reworking our migration tool
15:59 kf you could write a script that looked up the number
15:59 kf that#s what we do
15:59 kf we store the former id in a custom field and then look it up
16:00 kf well for other migrations, for items + biblio we attach the items as 952
16:00 kf and import them with the marc records
16:00 jcamins I'd prepare MARC records that included the items prior to the import.
16:01 kivilahtio__ jcamins: well we have component parts with links from 773w -> 001 of the parent
16:01 kivilahtio__ we cant change the database id
16:01 kf and if you use an sql script... i would carry the number you need for linking in another custom field and adapt the sql accordingly
16:01 kf kivilahtio__: 001 has nothing to do with biblionumber
16:01 jcamins 773$w->001 doesn't use biblionumber.
16:01 kivilahtio__ but our legacy data has that linkage
16:02 jcamins Right, and that's not biblionumber.
16:02 kf we do exactly that in koha
16:02 kivilahtio__ jcamins: so where does 773w point in koha?
16:02 kf linkings from 001 to $w subfields
16:02 jcamins The 001 field.
16:02 kf the 001 of the host record
16:02 kf which is not the bilbionumber
16:02 jcamins Remember, Koha and EG treat records rather differently.
16:02 kivilahtio__ kf: but why cant they be the same? they are bu definition the same thing?
16:02 kivilahtio__ jcamins: I hope I understood that better
16:03 kf [off] https://hfjs.bsz-bw.de/cgi-bin/koha/opac-search.pl?idx=kw&op=and&idx=kw&op=and&idx=kw&limit=bib-level%3Aa&sort_by=relevance&do=Suche
16:03 jcamins Whereas EG does a lot at the database level, Koha's bibliographic relationships are controlled only at the MARC level, in Zebra.
16:03 kf kivilahtio__: I don't think there is a definitoin that your internal record number has to match the controlnumber
16:04 kf and they could be the same thing, but it doesn't matter for yourl inking, for your linking it's only important that you keep your 001 and $w subfields intact
16:04 kivilahtio__ kf: ok
16:04 kf kivilahtio__: the link above is a result list of articles
16:04 kf as an example
16:05 kivilahtio__ kf: but what is the handicap for using our legacy id's? I have a solution in mind for that already and I think I could du that bug which was posted
16:05 kivilahtio__ programmatically it is not a problem, but I feel there are some sinister drawbacks to not using the in-DB primary key sequence?
16:05 kf what happens if koha assigns the next number?
16:06 kf are your legacy ids strictly numeric?
16:06 kivilahtio__ kf: I cant link holdins records to the bibliographic record
16:06 kivilahtio__ kf: yes
16:06 kivilahtio__ starting from 1001 ->
16:06 kivilahtio__ well within the databse id range
16:06 kf some people have voted for that bug - i am not a fan, but i just think meddling with the numbes is prone to create headache later
16:06 jeff j_wright++ -- thanks! (re: Plack pointers)
16:07 kivilahtio__ kf: I am only afraid of zebra
16:07 jcamins I think making it possible to keep 001 as biblionumber would be nice, but not really necessary.
16:07 kf hm there is also a bug for that
16:07 kivilahtio__ jcamins: I understand. But for me it is easier to just keep the 001 as a biblionumber :D
16:07 kivilahtio__ and make this odd bugfix while at it
16:07 jcamins kivilahtio__: it is almost surely not easier.
16:08 kf bug 9921
16:08 huginn 04Bug http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=9921 enhancement, P5 - low, ---, nunyo, Patch doesn't apply , Make it possible to force 001 = biblionumber
16:08 kivilahtio__ jcamins: so what risks do you see in keeping the 001 matching the biblionumber and itemnumber?
16:08 jcamins kivilahtio__: nothing, the problem is making the biblionumber match the 001.
16:08 kivilahtio__ jcamins: we use a strict integer, 0<
16:09 jcamins If you have a single bad record, a single inconsistency, you are likely to trash your database.
16:09 jcamins Itemnumber is of course unrelated.
16:09 kivilahtio__ jcamins: why? we have staggered numbers, some records have been deleted on they fly, so our id sequence would be like 1001,1002,1004,1006,1007,1008,1009,1011
16:09 kf hm sorry have to leave
16:10 kf bye all
16:10 kivilahtio__ kf: Thakns for your help!
16:10 jcamins kivilahtio__: what happens when it turns out your previous system actually had a duplicate 001?
16:10 kivilahtio__ jcamins: well the biblionumber is primary key, so it cannot be dubplicated. Mysql deals with it
16:11 kivilahtio__ so the legacy record is not migrated
16:11 kivilahtio__ jcamins: it can be manually migrated later
16:11 kivilahtio__ jcamins: this happens occasionally and I dont see an issue with that
16:12 jcamins You'd have to make it optional, because data from Voyager, III, and most other ILSes that I have encountered does not consistently have 001 set correctly.
16:12 kivilahtio__ jcamins: or in our case we collect the erroes in a huge log and manually clean up our databsae as we proceed with the migration
16:12 kivilahtio__ jcamins: naturally
16:12 kivilahtio__ jcamins: Ill start working on that patch tomorrow
16:12 jcamins As I said, I have no objection to you adding this feature, I just think that it's likely to not work nearly as well as you imagine.
16:13 kivilahtio__ jcamins: but it is good to know there are no Zebra related dangers
16:13 kivilahtio__ jcamins: if the only problem is the purity of the 001, then it will work for us
16:13 kivilahtio__ jcamins: And I would rather have this work for us in the master Koha
16:13 kivilahtio__ jcamins: so others could benefit as well
16:15 jcamins The other danger I can see is primary key exhaustion.
16:15 kivilahtio__ jcamins: Agreed
16:15 jcamins 001 is often an arbitrarily large number.
16:15 kivilahtio__ jcamins: but the primary key can be compressed, it is a big script but doable when the time comes
16:16 kivilahtio__ just need to realign all the foreign key references
16:16 kivilahtio__ and force full reindex of Zebra
16:16 jcamins So if you import one record that has a 001 of 2147483646, you've just killed your database.
16:16 jcamins No, I mean in MySQL.
16:16 kivilahtio__ jcamins: we dont have that big id's
16:16 jcamins You'll want to make *very* sure of that.
16:17 kivilahtio__ jcamins: I am sure Mysql takes care of its boundary values. But I will write a warning about that to the help
16:17 kivilahtio__ jcamins: do you really think MySQL goes crazy if you push it's integer boundaries?
16:17 jcamins No.
16:17 jcamins But Koha will.
16:17 kivilahtio__ jcamins: ahhh :D
16:18 kivilahtio__ jcamins: well I got your Koha INT_MAX :D
16:41 oleonard *sigh* That was plenty of meeting.
18:53 cait1 permissions--
19:19 rambuten Is the NCIP protocol specification finalized?
19:24 jeff i'm not sure which of the last three words to mockingly place in quotation marks.
19:24 jeff (sorry, snark is often not helpful)
19:24 jeff rambuten: there are at least three currently published versions of the NCIP standard, and I suspect you could consider all of them finalized. see http://www.ncip.info/the-standard.html
19:24 rambuten Yea, I was just looking at that site.
19:24 oleonard Why do you ask rambuten? RDA is far from final but people are already "using" it
19:24 rambuten We funding a PAC development, and the dev had some questions.
19:24 rambuten We're
19:24 oleonard Can you say more?
19:24 jeff The other aspect to consider is that the NCIP standard defines a protocol of sorts, which is used to imeplement one or more Application Profiles. Some of those Application Profiles have various "flavors" in the wild, when it comes to interacting with specific software.
19:27 oleonard Even in the open source world, people are often mum
19:28 jeff What aspect of NCIP relates to your PAC development (or vice versa)?
19:30 rambuten Well, the PACs will authenticate to the Koha (and we hope to incorporate/test w/ Evergreen too) db server.
19:30 rambuten SIP is a given, and probably NCIP will be included too.
19:30 jeff ah. just to be clear, can you define what you mean by "PAC" in this context?
19:31 rambuten Public Access Computing - public Internet computers
19:32 jeff Since both Evergreen and Koha do a good job of speaking SIP2, I'd almost suggest leaving it at that. Is there any particular reason you're thinking about supporting NCIP? Do you want to talk to other things that do talk NCIP and do not talk SIP2?
19:33 rambuten No, but "more is better", right? And there may be weird things out there that don't come to mind immediately where admins would prefer to use NCIP.
19:33 jcamins You should ask dyrcona about NCIP.
19:34 rambuten OK, next time I see him online I'll ping him.
19:34 jcamins Not to say that jeff doesn't know about NCIP. He might.
19:34 jcamins But I know dyrcona was working on NCIP a few weeks ago.
19:34 jeff Keep in mind that neither Evergreen nor Koha currently speak NCIP "out of the box". If you don't have a particular need for it, I'd say leave room/hooks/whatnot for future enhancement, but leave it at that. :-)
19:34 rambuten Wish KohaCon13 was having speed talks, I'd love to throw up some slides.
19:35 mtompset rambuten: Sometimes less is more. :)
19:35 jcamins rambuten: are you staying for the hackfest?
19:35 rambuten no, sadly
19:35 rambuten gmcharlt: about? Any NCIP comments?
19:36 cait1 rambuten: why not drop nancy a mail suggesting that?
19:36 rambuten hummm
19:36 cait1 there might be some changes in the program last minute, you neve rknow
19:36 gmcharlt rambuten: briefly, because I'm in the weeds at the moment, you might want to consider joining http://lists.mvlc.org/listinfo/ncip-dev
19:36 jeff There is an implementation of an NCIP 1.0 responder (intended for use with Evergreen and III's INN-REACH) here: https://github.com/jonscott/iNCIPit (with at least one more up to date fork available) and I believe Dyrcona and rangi and possibly others are collaborating on code here: http://git.evergreen-ils.org/?p=working/NCIPServer.git;a=summary
19:37 jeff there's a developers-only list at the url gmcharlt posted (though others can supposedly lurk, as i understand it), and a document describing it at...
19:38 jeff ah, here: NCIP Responder Design: https://docs.google.com/document/d/1NuV47145SnEV1f-9BrY2d9jpX8Ij_VxK2a51lotgSsU/edit
19:40 * jeff sends sympathy gmcharlt's way -- good luck in the weeds
19:41 gaetan_B bye !
19:43 * jeff returns to making iNCIPit.cgi more useful
19:52 mtj jeff, have you spotted rangi's ncip repo on github? -> https://github.com/ranginui/ncip-server-chris
19:53 rambuten Does everybody have an NCIP dev/blog going on?
19:55 jeff mtj: I hadn't, thanks. It seems to be a currently outdated additional remote for the repo I linked earlier, http://git.evergreen-ils.org/?p=working/NCIPServer.git;a=summary
19:56 mtj ah, ok… i just spotted it on #koha yesterday :)
19:56 * jeff stars/watches
19:56 jeff mtj: thanks. :-)
20:32 rangi jeff: thats just a mirror
20:32 rangi http://git.evergreen-ils.org/?p=working/NCIPServer.git
20:32 rangi its a collab between koha and evergreen people
20:33 cait1 very cool :)
20:37 rangi my repo runs a little ahead, then gets merged
20:38 rangi but its getting there
20:38 jeff rangi: thanks for the clarification. that's what it looked like, but i wasn't certain.
20:39 jeff well, i suppose "wasn't certain" isn't entirely accurate, but I'm not going to worry about being that precise. :P
20:41 jcamins Ooh, good news everyone! My library school has just added a course on Dialog to their schedule for the fall!
20:41 cait1 wow
20:41 cait1 was dialog that query langauge for expensive databases?
20:41 rangi ?
20:42 * cait1 should just look it up before asking stupid questions
20:42 jcamins Dialog _is_ the expensive database.
20:42 jcamins I only know of one thing that it is still used for regularly: citation searches for tenure applications.
20:43 jcamins Spending a small fortune per search somehow doesn't appeal to most people.
20:43 cait1 oh
20:43 jcamins Weird, eh?
20:43 cait1 it just sounded vaguely familiar
20:43 cait1 possible other databases use their syntax or something?
20:44 cait1 i remember carefully preparing searched than trying to be done really really fast with less possible searches from library school
20:44 cait1 searches
20:44 wahanui searches are rendered client-side, particularly if you click the knowledge trail maps on the right.
20:44 jcamins Yeah, that's Dialog.
20:44 cait1 searches then... typos... :(
20:44 cait1 glad you still got it
20:44 cait1 :)
20:58 magnuse rangi: did you ditch Dancer for NCIP, or is that another piece of the puzzle?
20:58 rambuten http://www.wired.com/wiredenterprise/2013/09/gendarmerie_linux/
21:11 eythian hi
21:11 wahanui que tal, eythian
21:13 rangi its another piece of the puzzle
21:14 rangi so, there is the NCIPServer.pm
21:14 rangi which is called by bin/start_server.pl
21:14 rangi that runs raw
21:14 rangi then NCIPResponder.pm
21:14 rangi is for the mod_perl implementation
21:16 rangi NCIP needs to do both raw and http(s) to be a full implementation
21:16 rangi im going to do a dancer wrapper too
21:17 wizzyrea the authority magnifying glasses - can you turn those off?
21:17 cait1 css
21:17 wizzyrea ok cool ty
21:17 cait1 :)
21:18 wizzyrea didn't know if that was a syspref or a hack your way around it thing./
21:20 cait1 hm i think the css here is intended
21:20 cait1 for that purpose
21:20 cait1 instead of adding another pref
21:20 wizzyrea yep that's fine
21:20 wizzyrea Just needed to know :)
21:20 cait1 so... does someone know why i have no log files appearing in my koha-dev/var/logs? :(
21:20 wizzyrea gitified?
21:21 cait1 no, stlil haven't got around to try that
21:21 wizzyrea bc if it's a gitified dev environment the logs will be where the packages keep them.
21:24 cait1 rangi++
21:24 cait1 restartedapache, now it works
21:28 cait1 anyone feeling like they really want to do me a big favour? :)
21:28 cait1 and order and receive an something on current master?
21:30 cait1 eythian: guess now you will have to come back here and buy him a beer? ;)
21:31 eythian cait1: guess I will :)
21:31 cait1 :)
21:43 bag manual?
21:43 wahanui manual is, like, at http://www.koha-community.org/documentation
21:43 bag anybody able to get that to load for them ^^^^
21:44 jcamins @isitdown koha-community.org
21:44 huginn jcamins: I'll give you the answer just as soon as RDA is ready
21:45 eythian bag: not working for me
21:45 jcamins bag: Looks like it's down.
21:45 cait1 hi bag
21:45 bag thanks for me too
21:45 eythian wizzyrea: is that something you can poke at?
21:45 wizzyrea yep
21:48 bag hey cait
21:53 rangi gmcharlt: you around?
21:59 wizzyrea that should be back
22:01 rambuten gmcharlt is in the weeds
22:01 wizzyrea oh look you're all checking to see if I'm telling the truth >.>
22:02 rambuten http://www.koha-community.org/documentation <- works for me
22:19 wizzyrea Someone just slamming it with Joomla attacks. Silly people, that won't work it's wordpress. :P
22:22 wizzyrea well it obviously *did* work as a denial of service heh.
22:33 rambuten koha-community you mean?
22:34 wizzyrea yep
22:36 bag thanks wizzyrea
23:29 bag @later tell cait - http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=10486 updates added like requested :)
23:29 huginn bag: The operation succeeded.
23:29 huginn 04Bug 10486: new feature, P5 - low, ---, jweaver, Needs Signoff , Allow external Z39.50 targets to be searched from the OPAC
23:42 dcook Hmm, anyone have any back-ups of the LoC website? :p
23:43 jcamins dcook: yeah, see code4lib. :)
23:43 jcamins Well, except for Z39.50. You're SOL there.
23:44 dcook Where am I looking for code4lib?
23:44 jcamins The mailing list.
23:44 wahanui the mailing list is at http://koha-community.org/support/koha-mailing-lists/
23:45 jcamins http://stuff.coffeecode.net/www.loc.gov/marc/
23:46 dcook Intriguing
23:46 dcook Thanks for the link
23:46 dcook I actually don't see anything in the mailing list. Hmm. *shrug*
23:46 dcook Well, except the system halt message
23:47 jcamins It's the last message in the HEADS UP thread.
23:47 dcook Ah, it probably just hasn't come in my mailbox yet
23:48 dcook As with all things, I get the digests
23:58 dcook "The Evergreen and Koha integrated library systems now express their record details in the schema.org vocabulary out of the box using RDFa. "
23:58 dcook Ummm, what?
23:58 dcook I don't recall seeing RDFa anywhere in Koha...
23:59 jcamins Run a detail record through Google's rich snippets tool.