IRC log for #koha, 2008-03-04

All times shown according to UTC.

Time S Nick Message
13:57 kados morning all
13:57 kados owen: can you rebase and doubls-check I've applied all your patches from this weekend?
13:59 owen Looks good to me
14:45 owen Hi atz
14:45 atz greets owen
14:45 owen Were you in Portland too?
14:45 atz yep
14:45 atz good conference, all around.  cool city too.
14:46 owen that's great. I wish I could have been there
14:47 atz columbus just barely lost out on hosting the next one
14:51 hdl hi all.
14:52 atz hello henri
14:52 atz are you safely back home in montpilier now?
14:52 hdl Yes after 24hrs trip
14:53 hdl was your trip back home OK ?
14:53 atz yes, I flew through Houston on the way back, without complication.
14:59 nengard hdl: 24 hours!! Yikes
15:00 nengard we hit snow but had no problems or delays - pretty lucky
15:01 hdl nengard:  yes : Chicago then Paris, Marseille  and Finally Montpellier
15:01 nengard hdl: you must be exhausted
15:01 hdl and jet lagged too... :D
15:01 hdl But it's ok.
15:01 nengard i bet
15:28 atz now that it's warming up, some sparrows seem to have finally found my bird-feeder
15:29 atrus123 Is there anything extra I need to do for my records to show up in the OPAC after having run the bulkmarcimport.pl script?
15:29 atrus123 running koha 3 git
15:29 atz atrus123: rebuild zebra?
15:30 atrus123 ah, that's it... this is the first time i've tried it with zebra, so i'm still figuring it out
15:30 atrus123 thanks
15:30 atz misc/migration_tools/rebuild_zebra.pl
15:30 atz no problem
15:32 atrus123 that was it!  thanks again.
16:08 atz owen: what is supposed to happen if the user uploads an oversized logo?
16:08 owen For opacsmallimage?
16:08 atz does it just get cut off, or shrunken?
16:09 owen I believe opacsmallimage will just get cut off
16:15 atz ok
16:21 kados owen: you around?
16:21 owen Yes
16:22 kados owen: I'm talking to Chris Catalfo about some issues we're having embedding biblios into Koha ... mainly some of the CSS difference between biblios and Koha are wreaking havoc
16:22 ccatalfo hey owen
16:22 kados isn't there some kind of 'reset' function for elements?
16:22 kados so we could potentially wrap the biblios app in a single <div> that has YUI.reset=1 or something?
16:23 kados to just remove all the style from that block?
16:23 owen So biblios is picking up CSS declarations for elements/classes that are common between them?
16:23 ccatalfo owen, that's right
16:24 ccatalfo i'm trying to wrap the biblios app up in a koha template
16:24 ccatalfo and so i'm including doc-head-open and doc-head-close
16:25 owen I wonder if this would work:
16:25 owen Put the YUI reset CSS into the biblios stylesheet, defining a specific div which it should apply to
16:26 ccatalfo this is the YUI reset css you're thinking of? http://developer.yahoo.com/yui/reset/
16:26 owen Yes. But that would only take care of HTML elements, it wouldn't cover any CSS classes/ids which biblios had in common with Koha
16:27 ccatalfo actually i think it is just the plain html elements: the tr, td, table etc
16:28 owen So where the reset CSS says "body,div,dl,dt,dd,ul,ol,li,h1,h2,h3,h4,h5,h6" etc., you'd alter it to "div#biblios div,div#biblios dl,div#biblios dt" etc.
16:29 ccatalfo yeah, maybe that would work
16:29 ccatalfo i'll give it a try now
16:46 ccatalfo Hhhm, I had forgotten that in the new version of extjs, biblios is not within it's own div.  it's actually rendering to the body el
16:46 ccatalfo let me see if i can change that behavior
16:47 ccatalfo 'cos that's how i eventually worked it w/ the prev. version of extjs....putting the koha menu/login area outside of extjs and then applying a few css fixes
16:50 kados makes sense
16:50 owen Okay, if a user's cart should be emptied when they log out, should they be warned? "Are you sure you want to log out? Your cart will be emptied."
16:51 kados not a bad idea, but I'd say definitely a low-priority one
16:51 owen Or should we assume that by logging out they really want to cover their tracks and don't bother asking
16:58 kados I think we can avoid asking for now
20:23 chris morning
20:33 atz greets chris
20:35 owen Hi chris
20:38 owen Hey chris, does your new boss know you're hangin' around in here? ;)
20:39 gmcharlt shhh
20:40 gmcharlt :)
20:40 chris heh yep
20:41 chris he's the president of the nz open source society, so he's pretty cool with it :)
20:41 kados hey chris
20:41 kados chris: can you just do a sanity check for me, rebase and see if there are any un-pushed patches left?
20:42 chris will do
20:42 kados chris: also, darrell ulm says he sent some patches in, let me know if you don't see those in the queue
20:42 kados I'll do a bit more coaching for him
20:43 chris nope, no patches waiting
20:43 chris ill go check the mail
20:45 chris patches from galen, which i see have gon thru already, and one from owen
20:45 kados cool
20:45 kados I'll let Darrell know
20:53 kados owen: around still?
20:53 owen Yes
20:53 kados owen: re: bug 1912
20:53 kados I'm still having probs
20:53 kados http://opacdemo.koha.org/cgi-b[…]ac-search.pl?q=it
20:53 kados images are defined for my itemtypes
20:53 owen 1912 is talking about the intranet advanced search screen
20:54 kados ahh
20:54 owen But yeah, the problem in that link is still an outstanding bug
20:54 kados well, in the intranet, I used to get bullets for itemtypes for each item grouping
20:54 kados when I had item-level-itipes enabled
20:55 kados and one itemtype per bib when I had it OFF
20:55 kados now I see no bullets
20:55 kados was that removed somewhere along the line on purpose?
20:55 owen Are you looking at the search results page?
20:56 kados yea
20:56 kados so it used to be that the itemtype icon was configured to show up as a bullet for each item grouping
20:56 kados if that makes sense
20:56 owen Okay, valid issue, but again--not what 1912 is talking about :) 1912's about the advanced search form. But anyway...
20:56 kados oh, oh, oh
20:57 kados gotcha
20:57 kados sorry :-)
20:57 owen There's still the question of whether you can even display an item type image for a search result item if item-level-itemtypes is turned on
20:58 kados hmmm, because it's not giving you that information accurately?
20:59 owen In that opacdemo link you posted above, for example...
20:59 owen How do you display an item type icon if different items attached to one record have different item types?
20:59 kados well, the OPAC display doesn't have that capability yet
21:00 kados the staff client should though
21:00 kados ok, so there are two views in the OPAC you can control with a syspref
21:01 kados to determine if it's a 'summary' (like the link I posted' or if it's built the way the staff client display is built
21:01 kados with the full display option each group of items is separated out by location, itype, itemcallnumber, and status
21:01 kados so each item-grouping (below the bib level) can theoretically have a separate itype icon
21:02 kados and it did work a month or so back
21:02 owen Are you talking about OPACItemsResultsDisplay?
21:02 kados yes
21:03 owen I'm not sure about that... I'll check the history of that template
21:08 owen I'm not seeing any differences in that section of the template between today and the day the OPACItemsResultsDisplay syspref was added, except for replacing classification with itemcallnumber
21:09 kados huh, I wonder why the thumbnail icons disappeared
21:09 kados they used to show up as the style for the bullet
21:09 kados is that ringing a bell at all?
21:09 owen That's in the intranet, isn't it?
21:09 kados yea
21:26 owen 'nite all, I'm off
21:32 atz does koha store "magnetic media" item information?
21:37 atz i.e. RFID tag?
00:28 fbcit evening koha
00:30 fbcit ouch
00:30 fbcit chris around?
00:30 chris sorta
00:30 fbcit :)
00:30 chris at work :)
00:31 fbcit updatedatabase.pl has duplicate db versions
00:31 chris hmm i thought i fixed that for your one?
00:31 chris or was it before that?
00:32 chris ah yes 2 57's
00:32 chris thats gonna be hard to fix without causing a mess for everyone
00:33 fbcit right'o
00:34 chris actually not so bad
00:34 chris ill make the create table one be 60
00:34 fbcit It could be moved down with an IF NOT EXISTS and the ver bumped again.
00:35 chris yeah
00:35 fbcit you'll take care of it?
00:36 chris yep, doing it now
00:36 fbcit tnx
00:36 chris im waiting on the sysadmins to install my userpackage on a machine here so i have some down time :)
00:37 fbcit hrmmm... git rebase says: Dirty index: cannot apply patches (dirty: koha-tmpl/opac-tmpl/prog/en/l​ib/jquery/plugins/clickmenu.css)
00:37 fbcit ??
00:37 chris hmm did you get halfway thru a rebase before, or applied some patches?
00:38 chris do a git status
00:38 chris might tell you whats up
00:38 chris ok, patch sent to the rm
00:40 fbcit git somehow sees a new file: #       new file:   koha-tmpl/opac-tmpl/prog/en/l​ib/jquery/plugins/clickmenu.css
00:43 chris hmm
00:45 fbcit man you type fast :-)
00:45 chris heh
00:45 fbcit our email's nearly collided
00:48 chris yeah cant see it in git anywhere
00:49 chris :)
00:49 fbcit never get upset... just get a bigger hammer :-)
01:09 hurrells Hello. I'm a new koha-3.00/Zebra install. I have added two bibs and items and cannot search them in the OPAC. Ran rebuild_zebra.pl and running zebrasrv gives ERROR 109. Any ideas?
01:10 hurrells I'd be happy to give any information you need to help me get this going as I am sooo close.
01:11 irma I all.
01:12 irma I wonder if anyone is around to ask a question about bulkmarcimport.pl?
01:20 irma Good morning
01:20 irma I would like to ask a question about bulkmarcimport.pl
01:20 irma Is Chris around?
01:23 atz hurrells: check your koha-conf.xml for zebra username and password
01:25 hurrells Yes, There is a username "kohauser" and the password is correct. When I watch zebrasrv output it: loads zebra-biblios.cfg and zebra-authorities.cfg, starts listeners and then displays it's PID. (ubuntu 7.10 - btw)
01:27 hurrells Then I do a title search for "Dune" which is a valid item title it logins OK does the Search biblios and there it reports ERROR 109
01:27 hurrells I'll be back in 5min.
01:42 hurrells I'm back. I was thinking that I might have a perl/module issue. I didn't use INSTALL-CPAN.pl to get my modules but instead used apt-get and perl installs for most.
01:44 hurrells Is there any documentation on getting koha+zebra working that I can be looking at besides the 2.2 to 3.0 upgrade note?
01:59 hurrells Here is what zebrasrv is logging...
01:59 hurrells 20:58:23-03/03 zebrasrv(1) [request] Auth idPass kohauser -
01:59 hurrells 20:58:23-03/03 zebrasrv(1) [request] Init OK - ID:81/81 Name:ZOOM-C/YAZ Version:1.154/3.0.24
02:00 hurrells 20:58:23-03/03 zebrasrv(1) [request] Search biblios ERROR 109 1 1+0 RPN @attrset Bib-1 @or @or @attr 1=4 @attr 4=1 @attr 6=3 @attr 9=32 @attr 2=102 Dune @attr 1=4 @attr 4=1 @attr 9=26 @attr 2=102 Dune @attr 1=4 @attr 5=1 @attr 4=6 @attr 9=26 @attr 2=102 Dune
02:00 hurrells 20:58:23-03/03 zebrasrv(1) [session] Connection closed by client
02:00 chris when you ran rebuild_zebra.pl for the first time, did you do the --reset ?
02:00 chris (as well as all the other switches :))
02:02 chris that does the init of the zebra db .. which might not have happened
02:02 hurrells Just a minute..
02:05 hurrells I ran: "root@h33:/usr/share/koha/bin/migration_tools# perl -I ../../lib ./rebuild_zebra.pl -a -b -r -k".
02:07 chris i wonder if -r is the same as -reset
02:07 chris cant hurt to try that
02:07 hurrells Then I stopped and restarted zebrasrv. FYI. rebuild_ze eems to need that -I swich for C4/Context stuff.
02:07 hurrells OK. Let me try again... (--reset this time)
02:08 chris -reset just one - i think .. who knows try both :-)
02:10 hurrells I did (--reset) and it appears to work. FYI. I get these warnings: "21:10:26-03/03 zebraidx(303) [warn] No such record type: grs.marcxml.record"
02:10 chris i think it will silently succeed
02:10 chris if its not a switch that is matched, ie it will just ignore it
02:11 hurrells FYI. When I check the /tmp dir it makes I can see the MARC records for "Dune" there...
02:11 chris yep all that will be right
02:11 chris but if its not initialising the zebra db, then it might not be writing the indices
02:12 chris error 109 normally means database doesnt exist
02:12 hurrells Yes. That's what I read in the YAZ docco.
02:12 chris which is why i suspect it wasnt initialised
02:13 chris there is a line in the script that if it sees reset then it initialises
02:13 chris before building the indices
02:14 chris i think it might have just ignored -r and --reset .. so if it still doesnt work after this try .. try with -reset just to be sure
02:16 chris if you did -k you can do -s on this one, to skip reexporting the data
02:16 hurrells It did not work. So I stopped zebrasrv; ran rebuild_zebra.pl with (-a -b -k -reset) and restarted zebra. Still no "Dune". FYI. OPAC is at: http://koha.hurrells.ca
02:18 hurrells I still get ERROR 109. Is there some place in the mysql db where I can see the indicies or are they external somewhere?
02:19 chris they are in zebra
02:21 chris so nothing to do with mysql im afraid, we export the data from mysql and build indices in zebra which we then search
02:23 chris you could try using yaz-client and trying to search the zebra db's
02:23 hurrells OK. So where are they? In /etc/koha/zebradb/... or somewhere else? I just want to see timestamps for action.
02:23 chris no idea
02:23 chris thats a good place to start looking though
02:23 chris :)
02:24 chris zebra is cool, but it does give the most useless error messages
02:24 hurrells Back in 2min.
02:30 hurrells OK. yaz-client....I think I need syntax help please. How do open to my db on localhost?
02:45 hurrells OK.  I tried "yaz-client koha.hurrells.ca:210/koha" and I think that 210 is not my z39.50 port. So where/how is this port selected?
03:02 hurrells OK. Nothing in /etc/koha/zebradb appears to be today's date so I suspect that the indicies are not being made. Back to rebuild_zebra.pl again.
03:11 hurrells OK. I symlink'd /usr/share/koha/lib/C4 into /usr/local/share/perl/5.8.8 so that rebuild_zebra can find these C4 modiles in @INC. Then...
03:11 hurrells I re-ran "./rebuild_zebra.pl -a -b -k -reset" and noticed the following:
03:12 hurrells 1: warnings about a missing record type " No such record type: grs.marcxml.record"
03:13 hurrells 2: Records: 0 i/u/d 0/0/0 which seems to me to indicate that no records where found.
03:15 hurrells I restarted zebra and still get the ERROR 109. I'm doing all of this as user "root" BTW..
03:16 hurrells As far as yaz-client goes I think that I don't have a z39.50 server going which I think should be up when zebrasrv is running right?
03:19 hurrells Possibly my problem may be from the bibs/items I created. All I did was create a bib, use z39.50 to pull in a record for Dune and then created an item/holding off it. Perhaps there is some crutial attribute that is not set in that way.
03:34 hurrells OK. Was just reading that after starting zebrasrv then you should start zebraqueue_daemon.pl. This causes zebrasrv to go crazy with endless "EsRequest  ERROR 224 update_record failed"
03:38 hurrells OK. Thanks for all your help. If something occurs then just jot it in here and I'll catch it tomorrow morning. Thanks again.
03:45 atz hurrells: if your root process is creating files, then it shouldn't be surprising that your later operations can't modify them
04:00 atz check permissions on your zebra files
05:37 CGI550 hello!
05:38 CGI550 I want to know more about Koha
05:58 CGI550 hello
06:03 CGI550 we are using ALexandria right now in our library. Can we ask help from the KOHa team if you could helps us transfer our date from Alexandria to KOHA?
06:07 CGI550 *our data
06:27 CGI550 hello
06:35 CGI550 ello everybody
06:35 CGI550 Whose the right person who can talk to about transfering our data to KOHA
06:49 CGI550 hello
09:55 hdl bug squashing session 1 in 5 minutes....
10:00 hdl bug squashing session 1 in 1 minute....
10:02 hdl hi there.
10:02 hdl is there anyone around ?
10:03 js_ i think you are alone ;)
10:05 hdl I guess so. (I was wondering if US liblimers would get up so early just after getting back home.)
10:17 kados hey hdl
10:17 hdl hi how are you ?
10:17 kados hdl: it will be just me :-)
10:18 kados :)
10:18 hdl in front of 300 bugs ;)
10:19 kados hehe
10:19 hdl I took all the bugs for HEAD and rel30 and ordered them by severity.
10:20 hdl http://tinyurl.com/2s32nt
10:20 kados looking
10:21 hdl hi frederic.
10:21 frederic hello
10:21 kados 298 bugs
10:22 hdl I merge one with an other and there were actually 299 bugs.
10:22 hdl So that is the correct count.
10:23 kados if I create a new vendor, I get a page that has a blank search!
10:23 kados You searched on vendor 2, 0 results found
10:24 kados bug 1841 specifically is fixed
10:25 kados hdl: how do you want to proceed?
10:25 kados step through each bug?
10:25 hdl at least major, blo cri...
10:26 hdl Leaving normal bugs to the big bug squashing session ?
10:26 hdl Testing and debugging will take time.
10:26 kados sure
10:26 kados so next is 1812?
10:26 hdl frederic, are you in for bug squashing session too ?
10:26 hdl yes.
10:27 kados we have inconsistent mechanisms for display of itemtype images
10:27 kados there are two modes currently, with item-level_itypes ON and with it OFF
10:27 frederic Well, I'm curious to discover what 'squashing' mean :)
10:27 frederic Listening...
10:27 frederic yes, if there is a smallish bug...
10:27 hdl (killing, assigning for resolution)
10:28 hdl kados : how can we solve that ?
10:29 frederic kados: I don't want to interfer to your bugs 'squashing' session, but I sent a patch to translate facets labels in opac/staff search result page. This patch didn't go through. Any reason?
10:30 kados frederic: I sent you a message about that patch, you didn't receive it?
10:30 kados sent twice in fact
10:30 frederic no
10:30 frederic yes, I resent it :)
10:30 kados really strange
10:32 kados hdl: so for all of your customers, you use biblioitems.itemtype for both icons and circ rules?
10:36 kados frederic: email was sent to Frederic DEMIANS <f.demians@tamil.fr>
10:36 kados frederic: on Feb 28
10:37 hdl kados : yes we aez.
10:37 hdl are
10:38 frederic kados: this address is correct but I didn't receive your message... Could you please resend it?
10:38 hdl at the moment.
10:39 kados frederic: sent
10:39 hdl But we have a customer that is willing to use itemlevel circ rules.
10:40 kados biblioitems.itemtype and items.itype are used for too many things atm IMO
10:40 kados it should be for circ rules only, not for icons, etc.
10:40 kados but for 3.0, we'll just fix the display probs
10:41 kados I will take 1812
10:41 hdl next 1892 : I can take if galen has not coped with it yet.
10:42 hdl 1898 even
10:42 kados yep
10:42 kados go for it
10:43 hdl Next: what are those reserve enhancements about ?
10:43 hdl 1532
10:44 kados yep, it's on the list for 3.0 to at least look at kyle's holds queue for dev_week
10:44 kados galen will take that
10:45 kados one prob is as Owen describes:
10:45 kados If the reserve list can't be limited by
10:45 kados branch  it is unusable to us.
10:45 kados but also, the holds list in the official koha should be able to 'target' locations, not just do a broadcast to every location
10:46 hdl (Are you IndependentBranches ? or not ?)
10:46 kados either way
10:46 kados NPL it's OFF
10:47 hdl I think it will be the same thing in France.
10:47 kados in fact, our larger US libraries have custom code to do it already
10:47 kados like stow (PHP), NPL (PHP)
10:48 hdl Can we reuse this code partly ?
10:48 kados galen is going to investigate this
10:48 hdl Or is it all hardcoded and quick coded ?
10:48 hdl so that is for bug 1824.
10:49 kados and 1532 also
10:49 hdl OK.
10:50 hdl next is suggestion management.
10:50 hdl We spoke about it with Nicole.
10:50 hdl I can cope with this one.
10:50 kados OK
10:51 kados 1803 is fixed
10:52 kados 1629 is next
10:52 hdl This is two folded bug :
10:52 kados IMO it's a very poorly implemented feature :(
10:52 hdl a) use opac advsearchapi.
10:53 hdl b) problem of storing files on the disk.
10:53 kados IMO, we shoudl not store any files on the disk
10:53 kados why should apache have write access just for rss?
10:53 kados IMO it's just an alternative 'view' of the results page
10:54 kados so it could be just a template
10:54 kados instead of a separate script
10:54 kados hdl: what do you think?
10:55 hdl it is not the same page at all as opac-searchresults.
10:55 hdl Maybe could be produced via an additional parameter RSS.
10:56 hdl (say outputformat for instance)
10:56 kados format=rss
10:56 kados snap :-)
10:56 hdl but we definitely need it to run and be ok.
10:57 hdl (And maybe would be great in conjunction of storing searchresults in the db)
10:57 kados why do we need it to store the results?
10:58 kados RSS is really just an alternate view of the same content
10:59 hdl you're right.

| Channels | #koha index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary