IRC log for #koha, 2006-03-30

← Previous day | Today | Next day → | Search | Index

All times shown according to UTC.

Time Nick Message
16:09 kados thd: check for a private msg
17:15 chris gotta love ppl who quote jello biafra as their sig
17:15 so i dont think i got this fuzzy thing right
17:15 kados huh
17:16 chris sorry, mike taylor's sig
17:16 for the fuzzy search im still getting
17:16 kados right :)
17:16 chris Search OK 2 1 1+0 RPN: @attrset Bib-1 @attr 5=100 @attr 6=1 @attr 3=3 @attr 4=2 @attr 2=3 @attr 1=4 distant
17:16 which should be 5=103
17:16 i think
17:19 russ hi there
17:20 kados woot
17:20 hi russ
17:20 how many do we have?
17:20 russ well these are just the ones from the website
17:20 kados right
17:20 russ so about 10 so far
17:20 kados cool
17:20 russ i know that paul and ineo are doing a direct mail campaign
17:20 kados if I get around to it I'll post a note to liblimes site
17:21 russ and i expect most of the registrations to come from that
17:21 kados and we're going to send some emails out too
17:21 russ most of the ones from the website have been from africa
17:21 kados tina's almost done with a graphic
17:21 interesting
17:21 we've just been so swamped here this month
17:21 we've got 5 migrations happening all at the same time :-)
17:22 all of which go live before the end of April
17:22 russ wow!
17:23 kados so I barely exist :-)
17:30 chris the other question
17:30 kados yep
17:30 chris do we want to do truncation by default
17:31 kados remind me what that does again?
17:31 chris or do we want the user to type something*
17:31 ok
17:31 current koha
17:31 if i search map
17:31 kados got it
17:31 chris ill find map, maps, mapping
17:31 kados right
17:31 chris with zebra, ill just find map
17:31 kados right
17:31 chris if i go map* tho
17:31 kados it's a tough one
17:32 what we need
17:32 is to make it an option
17:32 like a checkbox or something
17:32 chris yeah
17:32 kados so that NPL can decide what they want by default
17:32 during our next search meeting
17:32 their feedback was really helpful
17:32 since they use the system every day
17:32 chris cool
17:32 kados they kinda know what to expect intuitively
17:33 I'd say the same goes for fuzzy searching
17:33 if it could be a checkbox they coudl compare with and without
17:33 to see which does a better job
17:33 (is fuzzy another one of the ranking algorythms?)
17:34 chris dont think so
17:34 nope its not
17:34 kados cool
17:35 yea, so a checkbox might be nice
17:35 also, it'd be sweet if we could have a selection box that included rank1, rank2, rank3
17:35 so we could compare how they differ
17:35 chris im not sure you can change that on the fly?
17:35 kados and NPL could decide which to go with
17:35 no?
17:36 ahh , does it require mapping in the pqf file?
17:36 chris i have no idea how you change the relevance ranking
17:36 kados ahh ... there's a bit about it in the NEWS file
17:36 chris you can just tell it to do it
17:36 kados http://indexdata.dk/zebra/NEWS
17:37 there are a bunch of ranking algorithms in zebra
17:37 chris ill be happy if i can get this fuzzy to work
17:37 for a start
17:37 kados yea ... shelve the ranking for now
17:37 :-)
17:38 chris you might want to have a play with the latest pqf.properties and searchmarc.pm
17:38 its doing relevance ranking now
17:38 kados k ... will do
17:38 chris i dont think it was before
17:38 kados I thought it already was
17:38 chris nope
17:38 kados interesting
17:38 chris im sure it is now tho
17:38 @attr 2=102
17:38 kados so what was it doing before?
17:38 chris is being set
17:38 kados ie, what is kohatest.liblime.com 's relevance ranking?
17:39 chris it could have been doing it default
17:40 but by adding the /relevant
17:40 its now setting @attr 2=102
17:40 for the search
17:40 you can try it with or without the /relevant .. and see if the order changes
17:41 kados have you changed collection.abs lately?
17:41 chris it does for me
17:41 lemme check
17:41 kados (before I reindex)
17:42 cuz I need to merge those changes into record.abs if so
17:42 chris nope
17:42 only that melm 942$c            Material-type,Material-type:w
17:42 which i think you had already?
17:42 kados yep
17:42 k, I'm reindexing
17:42 chris you shouldnt need to
17:43 kados well, I already scraped the db
17:43 it only takes about 5 minutes now
17:43 chris ahh then you will :-)
17:43 cool
17:43 the /relevant does a make a different
17:43 kados zebraidx is crazy fast
17:43 chris difference even
17:43 kados cool
17:43 maybe it'll do everything NPL wants :-)
17:43 that would rock :-)
17:44 I'm gonna grab a drink while this index runs
17:44 bbiab
18:05 chris so if im reading this right
18:05 $p->option(record => content_of("/tmp/record.xml"));
18:06 we do need to pass the whole record
18:14 kados yea
18:14 I suspected that
18:23 chris hey, what the???
18:25 kados did it go and delete everything? :-)
18:25 chris naw nothing
18:25 still gettting
18:25 an error, ill figure it out, but now i gotta go do something else, ill be back looking into it in a bit
18:29 kados cool
18:29 thd: check pmesg
18:33 so when you get back, do I need to get rid of the 'relevance' ranking option to use the new relevance rank?
18:33 http://kohatest.liblime.com/cg[…]ha/opac-search.pl
18:33 search on 'the civil war' in interesting
18:33 s/in/is/
18:34 definitely different ranking than before
18:34 chris no
18:34 kados but still, the first record isn't 'the civil war' like NPL expects
18:34 chris relevance doesnt do anything
18:34 ie, relevance passes nothing through
18:34 kados k
18:35 chris so if you choose relevance, it wont sort by anything else
18:35 which means that now we have the /relevant  it will come back relevance ranked
18:35 kados search on 'Tango' doesn't pull up that record first
18:35 chris title or keyword?
18:35 kados title
18:36 at least it's on the first page :-)
18:36 chris http://kohatest.liblime.com/cg[…]tail.pl?bib=53677
18:36 that has tango twice
18:36 in title
18:36 once in 246 and once in 245
18:36 which makes it more relevant the one that just has tango once
18:37 that appears to be how it works
18:37 kados yea
18:37 that definitely makes sense
18:37 but probably as the second relevance consideration
18:37 chris you could try seeing if you could change the relevance
18:37 kados with proximity being first
18:38 huh, I wonder if I add the word tango like 10 times to the record
18:38 if it would come up first :-)
18:39 chris i bet it would, if you put it in the 246 field :)
18:39 owen Great, now we're search-engine spamming our MARC records!
18:39 kados hehe
18:40 mainly just testing to see how the ranking algo works :-)
18:42 now interestingly, that same search on NPL's current catalog
18:42 only has 18 results
18:42 I wonder why gloria used it as an example
18:42 chris i bet its searching less fields on the current koha
18:43 title how we have it set up,searches a lot more with zebra
18:46 kados well , there's some history here :-)
18:46 gloria's known for complaining that a search returns like 100 pages of results
18:46 so i would have expected all of her examples to do that on the old koha
18:46 chris ahh
18:48 kados bbiab, gonna head to the gym
19:03 chris holy crap it worked
19:07 rach wow joshua is actually not on irc :-)
19:07 chris heh
21:00 kados chris: got it working?
21:00 sorry, took a long dinner :-)
21:00 chris no worries
21:00 sure did
21:00 kados nice
21:00 chris if you update to the latest Biblio.pm and Search.pm
21:00 kados k
21:00 chris then try a delete
21:01 if it throws a server error, edit the url and change the bibid= to biblionumber=
21:01 kados heh
21:01 chris and then it oughta delete
21:01 kados k, I'll give it a shot
21:01 chris (template issue to fix)
21:01 kados yep
21:01 is it a template issue for head or for rel_2_2?
21:02 hmmm
21:02 well I'm not actually running the plugin
21:02 so none of the templates are really working
21:03 grrr
21:04 chris rel_2_2
21:04 i cant see that the delete will have actually worked in rel_2_2 anyway
21:04 addbiblio.pl expects biblionumber, it gets bibid
21:06 kados so you don't expect it will work for the plugin?
21:06 chris yeah it shold if you edit that url
21:06 kados k
21:06 chris im running it in the plugin
21:06 kados I'm gonna get the plugin going real quick
21:07 chris ack
21:07 see irma's email?
21:07 want me to respond to that?
21:07 kados yea
21:07 chris will do
21:16 email away
21:20 kados thx
21:21 chris it should work for head too (the delete) i havent tried it though
21:25 kados well, head is pretty broken at the moment
21:25 chris yeah
21:28 kados chris: did you copy over head versions of the tempaltes too or just the scripts for the plugin?
21:29 (and why is Amazon.pm required? was it throwing an error with the rel_2_2 version/)
21:32 chris i didnt touch templates
21:32 or scripts
21:32 the only thing i changed was Search.pm and Biblio.pm
21:33 so it should be the same instructions as setting up the plugin was before
21:34 i just added a get_xml_record to Search.pm
21:34 that hands back an xml rep of the data, given a biblionumber
21:35 which we can then use in Biblio.pm to get a record to delete
21:35 so get_record you get a MARC::Record   get_xml_record, you get the xml that zebra retursn
21:36 kados ahh ... cool
21:36 and that is how you're doing deletions?
21:36 (I'm writing a script to handle the plugin)
21:36 (hence my question about Amazon, template, etc.)
21:38 chris ahh right
21:39 so far i havent changed any templates
21:39 not in recent memory anyway :-)
21:39 the search ones will need to be changed
21:40 as will the marcdetail.pl as it passes the wrong variable to addbiblio.pl on delete
21:41 detail.tmpl too
21:41 so both moredetail.tmpl and detail.tmpl in the default templates
21:42 to work with the addbiblio.pl thats in head (ie that uses zebra)
21:42 kados maybe we need a plugin systempref :-)
21:47 chris: could you double-check this. I think these are the only files that need to be copied over:
21:48 MODULES:
21:48 Biblio.pm, Context.pm, SearchMarc.pm, Amazon.pm, Review.pm, Search.pm
21:48 INTRANET:
21:48 addbiblio.pl, additem.pl, detail.pl, MARCdetail.pl, ISBNdetail.pl
21:48 OPAC:
21:48 opac-detail.pl, opac-MARCdetail.pl, opac-ISBDdetail.pl
21:48 chris looks right
21:48 kados then modify addbiblio and additem scripts to grab the right template
21:48 (or maybe just create a symlink from cataloguing to acqui.simple, whaday think?)
21:49 chris that might work
21:49 if symlinks are allowed in the apache conf
21:49 kados the edit moredetail and detail to use the right variable
21:49 well, if they're doing a symlinked koha to begin with I assume they are
21:51 chris damnit
21:51 im sure this was working just before
21:54 2 secs
21:55 ah ha
21:56 the latest Biblio.pm (just committed) is the one to use
22:04 slef russ: hello?
22:04 russ hi slef
22:04 slef I've edited the screen.css, but the copy on the site hasn't changed. Do I need to release it somehow?
22:04 russ two secs let me look
22:06 yep ok what you need to do is check the box next to the file name
22:06 and then hit the notify button
22:06 and it will send a request for me
22:06 slef I wondered if that was it. Does the tutorial call it triggering?
22:07 russ yes
22:07 slef Why different jargon?
22:07 russ no idea
22:07 it predates my time with katipo and kea
22:07 slef Cool. Thought I was being stupid again.
22:07 chris the live site lives on a different machine, triggering scp's it to that machine
22:07 russ but we could have used the word publish or update
22:07 chris only some users can trigger, others can only notify
22:08 russ there you go
22:08 kados chris: my plugin script seems to have worked like a charm
22:08 chris: I'm reindexing now
22:08 slef How can I track my file trigger requests?
22:08 chris so you might see notify, i might see trigger (you can set up privileges on a dir by dir basis)
22:08 kados chris: sometime today or tomorrow we should discuss directory locations for the plugin
22:08 chris slef, see your email address?
22:09 click on that
22:09 kados chris: I'm assuming an actual production system would have the zebradb on a separate partition
22:09 slef no, I don't see my email address at the moment.
22:09 kados bbiab
22:09 chris ahh you might not have permissions to do that too
22:09 i see modified by
22:09 and ur email under that
22:10 slef ah, by the file? got it
22:10 russ triggered now
22:10 chris that tells you the history of the file
22:11 slef it's a bit "mystery meat navigation" that
22:11 chris so you can see your trigger request, and then that russel triggered
22:11 sure is, we are in the process of a UI overhaul
22:11 russ slef : yep it sure is
22:11 slef ok... can I see a list of my/all recent changes?
22:12 russ two secs
22:12 drop down menu top right
22:12 http://koha2.edit.katipo.co.nz/kaka/findFiles.cgi
22:12 ah it doesnt do by user though
22:13 slef thanks - that drop-down menu doesn't work for me ;-)
22:13 russ interesting
22:13 slef koha2.edit.katipo.co.nz doesn't have script execute permission, which I guess that uses as there's no submit button
22:14 boogie! http://jigsaw.w3.org/css-valid[…]ss&usermedium=all
22:14 logbot: have a botsnack
22:15 russ woo - nice work slef
22:15 thanks for updating the css
22:15 slef hrm, 1 or 2 more nigglies
22:15 when you fix the UI, please put something useful in the <title>
22:16 navigating back multiple pages is a bit hit-and-miss atm
22:17 russ huh?
22:17 why is that
22:17 ?
22:17 slef they are all titled Kea: koha2
22:18 no easy way to pick between them in the history
22:18 russ ah right
22:19 cool thanks for the feedback
22:20 slef is it OK if I put a background-color on the header? AIUI, image users should just get the image instead
22:21 russ good idea
22:25 slef strangxajxo!
22:29 ok, I can't see why that doesn't work... it's not taking the colour
22:31 ho hum, now's not the time
22:34 kados chris: still around?
22:34 chris: indexing is done
22:34 chris: http://opactest2.liblime.com/
22:35 chris: for some reason I'm getting some errors, none of the searches are working
22:35 [Tue Mar 28 19:36:22 2006] [error] [client 70.106.189.214] no records found, referer: http://opactest2.liblime.com/
22:35 [Tue Mar 28 19:36:45 2006] [error] [client 70.106.189.214] QUERY : dc.title all /relevant "john" at /home/koha/plugin/koha/intra​net/modules/C4/SearchMarc.pm line 250., referer: http://opactest2.liblime.com/c[…]ha/opac-search.pl
22:36 19:38:25-28/03 zebrasrv(3) [request] Search OK 0 1 1+0 RPN: @attrset Bib-1 @attr 5=100 @attr 6=1 @attr 3=3 @attr 1=4 @attr 4=2 @attr 2=102 the
22:36 19:38:25-28/03 zebrasrv(3) [log] term="the" nn=0 type=rank,w=34,u=4 count=0
22:36 19:38:25-28/03 zebrasrv(3) [log] user/system: 0/0
22:36 19:38:25-28/03 zebrasrv(3) [request] Sort OK - (1)->1
22:37 so oddly, KOha's reporting that sort failed, but looks like zebra things it worked
22:37 am I forgetting to change something in opac-search.tmpl or something?
22:39 chris hmm dont think so
22:39 it shouldnt be sorting unless you told it to sort by title or something tho?
22:39 ahh
22:40 you might need to edit the opac-search.tmpl and put in the <option value="">Relevance</option> bit
22:40 but that shouldnt stop results being found
22:40 kados wft, can't find anything in yaz-client either
22:41 something strange going on here ... I indexed these records, ran commit operation
22:41 hmmm
22:41 index files are huge
22:42 I'm stumped
22:43 what the heck is going on here
22:43 chris itd be nice to have a command like status
22:43 that tells you how many records are in the collection etc
22:43 but thats by the by
22:44 kados yea
22:44 Z> list_all
22:44 Connected to         : localhost:2100/kohaplugin
22:45 I don't get it
22:45 chris weird
22:46 kados I ran:
22:46 zebraidx init
22:46 zebraidx create kohaplugin
22:47 zebraidx -g iso2709 -d zebraplugin update records -n
22:47 wait ... did a commit before that last one
22:47 then another commit
22:47 then zebrasrv localhost:2100
22:47 chris and it appears to have no records?
22:47 can u try bulkmarcimporting in a couple?
22:48 kados ahh ... I found it
22:48 chris what was it?
22:48 kados -d zebraplugin should be -d kohaplugin
22:48 heh
22:48 chris doh
22:48 of course
22:48 kados I wonder if I can just change my config file
22:48 yea, I'll bulkmarcimport in a bit
22:49 yep, that did it
22:49 chris cool
22:50 kados huh ... so maybe we don't even need to 'create'
22:50 chris dunno
22:50 i seem to remeber it not working if we didnt
22:51 but then, its hard to know, since we were flailing around a bit :-)
22:51 kados hehe
22:52 chris if we can verify this delete works, i think its just making some template changes (to do the boolean and fix the delete) plus some more fixes to SearchMarc.pm
22:52 and we are good to go
22:53 kados ok, I"ll give it a shot
22:54 you deleting from MARCdetail?
22:54 what did I need to change in the template again?
22:54 I got distracted by the index problem
22:56 chris just hit delete
22:56 and ull get a server error probably
22:56 (after clicking ok to the javascript prompt)
22:56 and a url that looks like
22:56 kados yep
22:57 http://kohatest2.liblime.com/c[…]delete&amp;bibid=
22:57 no id unfortunately
22:57 strange
22:57 chris cgi-bin/koha/acqui.simple/addbi​blio.pl?op=delete&amp;bibid=33
22:57 ohh
22:57 hmm interesting
22:57 what about from the normal view?
22:57 kados huh
22:57 [Tue Mar 28 19:59:30 2006] [error] [client 70.106.189.214] Can't locate C4/Log.pm in @INC (@INC contains: /home/koha/plugin/koha/intranet/modules /etc/perl /usr/local/lib/perl/5.8.4 /usr/local/share/perl/5.8.4 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.8 /usr/share/perl/5.8 /usr/local/lib/site_perl .) at /home/koha/plugin/koha/intranet/c​gi-bin/acqui.simple/addbiblio.pl line 30., referer: http://kohatest2.liblime.com/c[…]tail.pl?bib=58057
22:57 is that a new module in head?
22:57 chris must be
22:58 kados I'll try copying it over
22:58 chris better add that to the list
22:58 on the MARCdetail.pl page?
22:58 Biblionumber : 33 With Framework
22:58 does it have somethign like that?
22:58 oh and what templates are u using?
22:59 kados rel_2_2 templates
22:59 npl
22:59 chris ah
22:59 i havent tried
22:59 them
22:59 kados you're using default?
22:59 chris you could try default
22:59 kados damn it :-)
22:59 chris we will want to fix the npl ones anyway
23:00 anyway
23:00 this url
23:00 yoursite/cgi-bin/koha/acqui.simple/a​ddbiblio.pl?op=delete&bilionumber=33
23:00 will delete that record
23:00 kados k ... I'll try that
23:01 hmm
23:01 500 ing
23:01 chris sorry my typo
23:01 biblionumber
23:01 kados [Tue Mar 28 20:03:50 2006] [error] [client 70.106.189.214] \tHTML::Template::new('HTML::Template', 'filename', '/home/koha/plugin/koha/intranet/h​tdocs/intranet-tmpl/default/...', 'die_on_bad_params', 0, 'global_vars', 1, 'path', 'ARRAY(0x8de2d64)', ...) called at /home/koha/plugin/koha/int​ranet/modules/C4/Output.pm line 77
23:01 chris not bilionumber :)
23:01 kados ahh
23:02 ahh ... need to edit addbiblio.pl
23:02 and add that to the list
23:02 chris bibid ?
23:02 ohh the template path?
23:03 kados ahh ... so that means my symlink didin't work
23:04 # ls -l cataloguing
23:04 lrwxrwxrwx  1 root jmf 43 Mar 28 18:57 cataloguing -> /home/koha/plugin/rel_2_2/koha/acqui.simple
23:04 wonder why it's not working
23:04 chris i think it is
23:04 thats just for the script isnt it
23:04 and its finding the script
23:04 kados ha
23:04 I'm a dumbass
23:04 chris :)
23:05 the delete should just do a redirect anyway, not display a template
23:06 you should all things going well end up at the catalogue search page
23:07 kados huh ... for some reason it's trying for the default
23:07 chris it will do that
23:07 kados [Tue Mar 28 20:08:17 2006] [error] [client 70.106.189.214] HTML::Template->new() : Cannot open included file /home/koha/plugin/koha/intranet/htdocs/intran​et-tmpl/default/en/cataloguing/addbiblio.tmpl : file not found. at /usr/share/perl5/HTML/Template.pm line 1580
23:07 chris if it cant find the corresponding npl one
23:07 ie, it fails back to the defaul
23:07 t
23:07 kados ahh
23:09 well ... I think it worked
23:09 chris zebra log will tell ya
23:09 if you see
23:09 15:06:29-29/03 zebrasrv(8) [request] EsRequest OK: Done !
23:09 kados nopw, but we're close
23:09 [Tue Mar 28 20:08:17 2006] [error] [client 70.106.189.214] HTML::Template->new() : Cannot open included file /home/koha/plugin/koha/intranet/htdocs/intran​et-tmpl/default/en/cataloguing/addbiblio.tmpl : file not found. at /usr/share/perl5/HTML/Template.pm line 1580
23:09 woops
23:09 20:11:40-28/03 zebrasrv(9) [request] EsRequest  ERROR 224 only XML update supported
23:10 chris ah ha
23:10 get the latest version of Biblio.pm
23:10 i fixed that
23:12 kados h
23:12 uh
23:12 now I get:
23:12 20:15:03-28/03 zebrasrv(12) [log] delete grs.xml <no file> 0
23:12 20:15:03-28/03 zebrasrv(12) [warn] cannot delete record above (seems new)
23:12 20:15:03-28/03 zebrasrv(12) [warn] zebra_delete_record failed r=1
23:13 but the detail page for that record still works
23:13 strangely
23:13 chris interesting
23:13 i havent seen that error before
23:13 i get
23:13 kados and it's a 500 error too
23:13 chris 15:06:29-29/03 zebrasrv(8) [log] 1605 bytes:
23:13 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
23:13 <collection xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/MARC21/ ...
23:13 15:06:29-29/03 zebrasrv(8) [log] delete grs.xml <no file> 0
23:14 15:06:29-29/03 zebrasrv(8) [log] zebra_end_trans
23:14 kados huh
23:14 apache has:
23:14 [Tue Mar 28 20:16:27 2006] [error] [client 70.106.189.214] ZOOM error 224 "ES: immediate execution failed" (addinfo: "delete_record failed") from diag-set 'Bib-1'
23:14 lemme try another biblionumber
23:15 same deal
23:15 chris hmmm
23:15 kados 20:17:53-28/03 zebrasrv(16) [log] delete grs.xml <no file> 0
23:15 20:17:53-28/03 zebrasrv(16) [warn] cannot delete record above (seems new)
23:15 20:17:53-28/03 zebrasrv(16) [warn] zebra_delete_record failed r=1
23:15 chris whats before the delete bit
23:15 some xml?
23:15 kados 20:17:53-28/03 zebrasrv(16) [log] 2189 bytes:
23:15 <record xmlns="http://www.loc.gov/MARC21/slim"><leader>00686     2200217   4500</leader><controlfield tag="001">105730</controlfield><controlfield tag="003">ACLS</contr ...
23:15 chris right
23:16 kados ahh ... I did modify my version of MARC::File::XML
23:16 to strip out the <collection> tag
23:16 chris ah
23:16 kados could that be it?
23:16 chris that might be
23:16 kados lemme uncomment that
23:16 chris i get
23:16 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
23:16 <collection xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/MARC21/ ...
23:16 kados right
23:16 chris i might be this bit
23:17 <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
23:17 thats needed
23:18 kados huh ... and how do I get that big?
23:18 bit even
23:18 chris no idea
23:19 lets just try with the collection bit
23:19 kados huh ... I can't seem to find that modif
23:19 guess I'll just delete it and reinstall
23:22 huh ... now my rmmod.pl script says it's not even installed
23:22 arrrg
23:31 I think I know what the problem is
23:31 chris whats that?
23:31 kados when I indexed those records
23:31 they went in with a <record> wrapper
23:31 cause I indexed them using zebraidx
23:31 which doesn't use a collection wrapper
23:31 I'll drop a replacement XML.pm file in C4
23:32 with <record> wrapper
23:32 course, that still might screw things up
23:32 chris hmm
23:32 kados cause somewhere we might need <collection> right?
23:32 chris could be
23:32 kados (zebra doesn't create a collection wrapper when you use zebraidx)
23:32 chris ahh right
23:33 kados arrg, my head hurts
23:54 chris: I've no idea what to do here
23:54 chris: maybe put a collection wrapper back in if one doesn't exist?
23:55 chris: zebra shouldn't use it anyway
23:55 so I'm confused about why it should matter
23:55 chris i have no idea im afraid, im not sure why it works on mine, but not there
23:55 kados is there a way I can delete in yaz?
23:57 chris hmmmm
23:57 not that i know of
23:58 kados well what else could it be
23:58 20:58:15-28/03 zebrasrv(24) [log] delete grs.xml <no file> 0
23:58 chris yep
23:58 kados 20:58:15-28/03 zebrasrv(24) [warn] cannot delete record above (seems new)
23:58 20:58:15-28/03 zebrasrv(24) [warn] zebra_delete_record failed r=1
23:59 could you paste in your log
23:59 starting with the xml bit
23:59 chris 15:22:27-29/03 zebrasrv(13) [log] delete grs.xml <no file> 0
23:59 15:22:27-29/03 zebrasrv(13) [log] zebra_end_trans
23:59 thats the bit right after the xml bit
23:59 no warns
23:59 it just does it
00:00 can u check ur koha-error_log
00:00 it should have the record in xml in it
00:01 kados I [Tue Mar 28 20:58:15 2006] [error] [client 70.106.189.214] heres the record to delete 58058 at /home/koha/plugin/koha/int​ranet/modules/C4/Biblio.pm line 1001.
00:01 [Tue Mar 28 20:58:15 2006] [error] [client 70.106.189.214] ZOOM error 224 "ES: immediate execution failed" (addinfo: "delete_record failed") from diag-set 'Bib-1'
00:01 I put a warn in with the XML
00:01 chris you could edit Search.pm
00:01 in get_xml_record
00:01 kados yea, I did
00:01 chris where it says ###$record
00:01 kados added a warn, also tried adding a <collection wrapper
00:02 chris make that ### $record
00:02 kados really? waht woudl that do?
00:02 chris thats using smart::comments
00:02 it will output it nicely
00:02 kados ok
00:03 chris should be just before that heres the record to delete line shows up
00:03 in the log
00:03 kados yea
00:03 it's there
00:03 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214] ### $record: '<record xmlns="http://www.loc.gov/MARC21/slim"><leader>00686     2200217   4500</leader><controlfield tag="001">105730</controlfield><controlfiel
00:03 <snip>
00:03 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214]   <idzebra xmlns="http://www.indexdata.dk/zebra/">
00:03 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214]     <size>686</size>
00:03 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214]     <localnumber>50800</localnumber>
00:03 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214]     <filename>records/npl.iso2709</filename>
00:04 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214]   </idzebra>
00:04 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214] </record>'
00:04 so I'm guessing that's different thank yours
00:04 chris hmm thats with ur warn eh
00:04 with smart comments you dont get all the stuff in front
00:04 just the variable
00:04 kados huh ...
00:04 chris all this
00:04 [Tue Mar 28 21:05:43 2006] [error] [client 70.106.189.214]
00:04 makes it easier to see :)
00:05 well i have no filename bit
00:05 kados what i pasted above is not my warn
00:05 cause I uncommented my warn
00:05 chris ah maybe ur logs just do that
00:05 kados yea, could be
00:05 chris <idzebra xmlns="http://www.indexdata.dk/zebra/">
00:05    <size>1454</size>
00:05    <localnumber>44</localnumber>
00:05  </idzebra>
00:05 </collection>
00:05 is how mine finishes
00:06 kados huhso mine has a filename
00:06 chris plus </record> instead of </collection>
00:06 kados and is wrapped in <record>
00:06 chris heres where my record bit is
00:06 </record>
00:06  <idzebra xmlns="http://www.indexdata.dk/zebra/">
00:06    <size>1454</size>
00:06    <localnumber>44</localnumber>
00:06  </idzebra>
00:06 </collection>
00:06 before it
00:07 kados huh
00:07 so maybe when I index I need to specify something special
00:07 chris you could try bulkmarcimporting 1 or 2 records
00:07 kados yea
00:07 I'll do that
00:07 chris and see if they delete
00:07 kados but we still need a way to deal with both types
00:07 because there's no way we're importing 150K records with perl-zoom
00:07 it'd take 4 days
00:09 chris yep
00:09 itd be good to know if that was the problem
00:09 though
00:10 kados huh ...
00:10 i imported three records
00:10 chris yep
00:10 kados but I can't find them
00:10 chris hmmm
00:12 kados looks like they imported ok
00:12 21:11:39-28/03 zebrasrv(26) [log] 3053 bytes:
00:12 <?xml version="1.0" encoding="UTF-8"?><collection xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/MARC21/slim http://www.l ...
00:12 got a commit start and clean
00:12 chris hmm
00:12 kados so something definitely changed
00:12 chris weird
00:13 kados yea
00:14 I wonder if anything else works
00:14 like modifies or check-outs
00:15 huh ... member searches 500
00:15 wtf
00:15 Tue Mar 28 21:17:54 2006] [error] [client 70.106.189.214] Undefined subroutine &main::BornameSearch called at /home/koha/plugin/koha/intran​et/cgi-bin/members/member.pl line 80., referer: http://kohatest2.liblime.com/c[…]tail.pl?bib=58058
00:15 great ...
00:16 chris what the hell
00:16 ahh thats probably in Search.pm
00:17 kados it's in Borrowers.pm in rel_2_2
00:17 ahh .. head I mean
00:17 chris right, that will be the problem it must have been in Search.pm in 2_2
00:17 kados yup
00:17 that's gonna wreak havok
00:18 I've had enough for tonight :-)
00:18 heading to bed
00:18 chris what we will need to do, is add the sub routines to Search.pm for rel_2_2 not replace Search.pm
00:18 kados more beating my head against the wall tomorrow :-)
00:18 chris :)
00:18 i have no idea why the deletes wont work
00:18 but ill look into it more
00:19 kados thanks
00:19 chris its tricky as i dont seem to be able to replicate it
00:19 kados let me send you my install notes
00:19 so you can
00:20 chris i cant make it not delete
00:20 i mean
00:20 kados well, you imported differently than me too
00:20 which I suspect is the problem
00:20 chris yeah i did bulkmarcimport
00:20 but why didnt that work for you either
00:20 kados dunno
00:20 chris anyway, go get some sleep
00:21 kados before i go
00:21 chris clear head always helps
00:21 kados here's what you would need to do
00:21 record.abs is already committed
00:21 so just create a zebra/records dir
00:21 chris whats different between record.abs and collection.abs ?
00:21 kados copy the npl.mrc I gave you over to the records dir
00:21 chris anything?
00:21 kados and rename it as npl.iso2709
00:22 nothing different
00:22 chris k
00:22 kados to import
00:22 go:
00:22 # zebraidx -g iso2709 -d kohaplugin update records -n
00:22 # zebraidx commit
00:22 chris maybe we should just symlink them
00:22 kados where zebraplugin is your dbname
00:22 yea, good idea
00:22 we need both if we're going to zebraidx as well as bulkmarcimport
00:22 chris right
00:22 kados because MARC::File::XML uses a collection wrapper but zebra doesn't
00:23 anyway, that will get you up and running like me
00:23 chris zebraidx does create the koha stuff right?
00:23 doesnt even
00:23 kados it assumes you've got the koha stuff in there already
00:23 chris right
00:23 kados npl's data does
00:23 so meadville will import into 2.2
00:23 then export with the marc.pl script
00:24 then import into zebra with zebraidx
00:24 chris ah right
00:24 kados (I modified marc.pl to just dump to a file)
00:24 (which is what the npl.mrc file is)
00:24 (do you still have that btw?)
00:24 chris ok ill have a look see if i can get it to break
00:24 kados (I'll post is somewhere just in case)
00:25 chris id have to load all the npl data for it to be useful
00:25 ill just export a couple of the records ive already loaded
00:25 and then zebraidx them
00:25 and see if that breaks
00:27 kados all you need to do to modify the marc.pl export script
00:27 is comment out the first print
00:27 then you can run it like
00:28 ./marc.pl op=export start_bib=1 end_bib=3 > records.mrc
00:28 chris ta
00:28 kados k ... I'm out
00:28 chris cya
03:27 osmoze hello
03:40 pierrick is an old koha developer present ?
03:45 chris im not that old, im only 32
03:45 :-)
03:45 what do you need pierrick?
03:48 hdl hi
03:48 chris hi hdl
03:54 pierrick: im just off to wash the dishes, ill be back in a bit, if you want to ask anything about old koha
04:08 pierrick about bugtracker, I was wondering what meant the state "RESOLVED VERIFIED"? What is required to close the bug after this?
04:14 chris back
04:15 ahh, it should go resolved - fixed, or resolved - invalid, etc
04:16 then verified when someone has checked it
04:16 closed when its been released
04:17 so resolved - invalid, could go straight to closed
04:17 fixed, should be verified, then closed
04:17 at least thats how i use it
04:18 ideally a developer marks something as fixed, when he/she thinks they have fixed it, someone else verifies the fix, then the qa person closes the bug
04:18 if they are happy
04:20 pierrick OK, so every bug VERFIED on Koha 1.x should be CLOSED today
04:20 chris yes
04:21 pierrick thank you chris, I'll close som bugs today
04:21 chris thank you
04:21 pierrick chris, do you take care of koha.org ?
04:21 or is that russ work ?
04:21 chris some of it
04:21 russ does most of it
04:22 if you want something changed i can do that
04:22 pierrick yesterday, you said I made good suggestions about user oriented tools
04:22 chris ah yes you did
04:23 number 2 is already started
04:28 the extensions/templates is a great idea
04:28 ive been wanting to do something like that for a long time, where people can submit their templates
04:29 pierrick I wanted to discuss with kados about extension manager, because he said he wanted a kohazilla.org who would do the same
04:30 chris we could have
04:30 templates.koha.org
04:30 extensions.koha.org
04:30 etc
04:30 pierrick why only "templates" ?
04:30 OK, better with "extensions"
04:30 Imagine customers sharing their biblioframeworks as well...
04:31 chris true that would be great
04:31 coordinating this, and looking after these sort of things are the kind of things id like to see a foundation take charge of
04:32 pierrick when I searched this kind of software for PhpWebGallery, I found nothing, so a team member wrote something
04:32 we are going to make it free independant software, usable for any project
04:33 chris sweet
04:33 pierrick it is based on extensions, revisions of extension, compatibility with the extended software. For example, revision 2 of template "darksky" is compatible with Koha 2.2.5 and 2.2.6
04:34 you come to the extension manager and ask template compatible with Koha 2.2.5, "darksky" will be listed
04:34 chris ooohh that sounds ideal
04:35 pierrick did you work with Joshua about an extension manager ?
04:35 chris no, it was just in the idea stage
04:35 we didnt get any further
04:35 (no time)
04:37 pierrick OK, will you or russ have time to install something on koha.org in April
04:38 chris yep ill make time
04:39 osmoze someone knows if a module inventory is in todo list ?
04:40 chris i think tumer has written some code to do stock inventory
04:40 i seem to remember an email he sent paul about it
04:40 but i havent seen it working
04:40 osmoze ok, i will investigate :)
05:52 pierrick reading koha-devel archive, I've found a suggestion on user tagging feature : http://lists.nongnu.org/archiv[…]-10/msg00011.html Was this feature added? Is someone interested in doing it?
05:53 chris it seems like something interesting, but no one has worked on it as far as i know
05:54 pierrick OK :-)
05:55 are there features like suggestions to borrowers already been planned ?
05:55 chris suggestions to borrowers?
05:55 pierrick I mean like on Amazon. Based on what you already look and buy, Amazon suggests to look at other items.
05:55 chris ahhh
05:56 pierrick tagging would do this in some way
05:56 chris we could do something like that
05:56 pierrick based on other users experience
05:56 chris that would be cool
05:58 pierrick "see" and "see also" are already a first suggestion way, but very static. In fact, I don't know very well how "see" and "see also" work... I have to investigate
05:58 chris yeah, letting users suggest others
05:59 would be good
05:59 pierrick so the user tagging feature is interesting :-)
06:00 chris yep
06:00 pierrick but does not exactly suggest... it's something else...
06:00 chris yeah, grouping
06:01 hmm suggestions, how to do it nicely
06:02 pierrick the Amazon way ?
06:02 chris well we'd have to collect more info
06:02 the trick is
06:02 that in order to be useful we need to track what users look at
06:03 which is scary in the USA with the patriot act
06:03 pierrick lol
06:04 chris only being slightly joking, if you dont track what users look at, you cant hand that info over to the FBI
06:04 pierrick I have the same problem with PhpWebGallery navigation logs
06:04 paul pierrick ?
06:05 pierrick we don't need to make logs associated to a borrower. We just need to know that "a" borrower did this and then that.
06:05 paul, yes ?
06:06 paul you should read carefully the wiki page about KohaCon, may 3rd.
06:06 chris ohh true
06:06 paul i've added you as a speaker !
06:06 chris good point pierrick
06:06 hi paul
06:06 paul hi chris.
06:06 pierrick paul, OK, interesting to speak in public :-)
06:07 osmoze what is "vendangeur" for pierrick Paul  ?
06:07 paul hi osmoze.
06:07 pierrick will explain itself !
06:07 osmoze hi :)
06:10 chris cool
06:13 pierrick chris, about user tagging, what would be the difference with virtual shelves? I think user tagging is just improving virtual shelves functionnality
06:14 chris hmm you could be right
08:34 osmoze paul ?
08:34 paul yep
08:37 osmoze une suggestion, dans les transferts
08:38 a chaque livre il faut remettre la branch
08:38 c est pas tres pratique quand on transfert 500 livres
08:38 qq un a deja fait cette remarque ?
08:40 (je m explique, je passe 500 livres de laf à vazerac, a chaque fois je dois remettre vazerac...Une tres grosse perte de temps pour nous
08:40 )
08:40 je reprend
08:40 paul tu veux dire que la branche de destination doit être resaisie après chaque transfert ?
08:40 osmoze oui
08:41 il serai bien de faire comme pour le pret, tu mets ta branches et hop, ca reste
08:43 tu vois mon idée alors paul ?
08:44 paul (au tel)
08:51 oui je vois l'idée
09:07 osmoze tu penses qu une petite modifs pourrait etre faite pour la 2.4 ?
09:07 paul je vais la noter.
09:07 je pense que ca devrait être simple
09:09 osmoze cool :)
09:22 paul hdl ?
09:23 hdl oui
09:23 paul tu as réservé l'hotel à partir du dimanche 7 mai à Marseille ?
09:23 hdl Oui.
09:23 paul ah, ok, je me demandais si c'était une erreur de date.
09:23 hdl Tu pense que je devrais réserver à partir du 8 ?
09:24 paul du tout, tu passeras le dimanche soir avec nos hotes étrangers
09:24 (et pierrick, qui a réservé le 7 au soir lui aussi)
09:24 hdl Il me semblait que ce serait mieux d'être sur place la veille.
09:38 kados paul: yes, I was trying to call them rjust now
09:38 paul: I never recieved an email from them
09:39 paul: after I submitted the web form
09:39 paul they told me they sended one to request your credit card number
09:39 kados hmmm, I did not recieve it
09:39 paul they at least have recieved your mail
09:39 kados great, can they send it again?
09:39 paul if you don't succeed to get them, let me know.
09:40 kados I have attempted to dial +33 4 91 33 65 64
09:40 but it doesn't seem to connect ...
09:41 paul that's the correct #
09:41 kados I'll try again
09:41 paul you can also try http://www.europe-hotel.fr/fr/[…]tacts/contact.php
09:43 kados ok
09:43 paul: quick question
09:43 paul: in the US, often a hotel does not charge extra for more persons in a room (up to 5)
09:43 is this the same in France?
09:43 (we will add one to our number)
09:44 paul it depends on the hotel.
09:44 who will you add ?
09:44 kados Sara, our junior programmer :-)
09:44 programmer in training ... she attends the local college
09:44 paul for europe hotel, it seems they only have single / twin.
09:44 kados and 'double' whatever that is
09:44 paul twin is for 2, same price if you are 1 or 2.
09:45 kados right
09:45 paul twin is : 2 1 person bed
09:45 double is : 1 2 person bed
09:45 kados right
09:45 paul they have a 'triple'
09:45 (75€)
09:45 you could ask them.
09:45 kados yes, I will
09:46 paul but 1 man & 2 women in the same bedroom will make ppl looking at you as a strange guy...
09:46 kados hehe
09:46 ok, message sent
09:46 paul how much cost a flight ticket Columbus => paris ?
09:47 (& back to columbus)
09:47 kados approx $700 USD
09:47 paul that's what I was thinking. Quite the same for us paris > columbus > paris
09:48 my student arrive in 20mn
09:48 youpi !
09:49 kados woohoo
09:51 paul: are there many riots in Marseille?
09:51 paul riots ?
09:51 kados revolutions :-)
09:52 small revolutions == riots :-)
09:52 paul do you want to laugh a little :
09:52 the police says that yesterday there was 28 000 ppl in the street (against the CPE)
09:52 kados yes
09:53 (in US they reported 200K in Marseille alone) :-)
09:53 paul and how many for ppl organising the défilé (don't know the word in english) ?
09:53 they said 250 000
09:53 usually police says X and ppl say 2X or 2.5X
09:53 kados hehe
09:53 paul but in Marseille, there is always a *8 or *10 factor !
09:53 kados he
09:54 so which is it? 28K or 250K ? :-)
09:54 pierrick that's why Marseille is so much appreciated...
09:54 kados hehe
09:54 paul to answer your question : don't be afraid and don't hear what NYT says : france is safer than Badgad !
09:54 pierrick let's say 50K
09:54 paul s/hear/believe/
09:54 I would say 2 things :
09:55 * police has a constant method to cound ppl : they take a photo from the sky, and use a grid.
09:55 100 in a grid & 100 grids filled => 10 000 persons.
09:55 kados right
09:55 paul organisators counts ppl passing on a given point.
09:56 so, it's not strange to have a x2 or x2,5 difference.
09:56 but I bet that Marseille want always to be the largest & highest & tallest & biggest & ...est city in france.
09:56 so, it's not enough to have 30 000 ppl.
09:56 250 000 is better.
09:56 so let's say we were 250 000.
09:57 pierrick to conclude I would say that riots should never take place in forrest...
09:57 paul (note that the city is 800 000 ppl)
09:57 pierrick trees would prevent from counting!
09:59 kados paul: I finally got through to the hotel ... they are writing me another email right now
10:00 paul great !
10:06 kados chris got deletions working last night
10:06 well, partly
10:06 they work for records imported via perl-zoom
10:07 paul kados : my student is here.
10:07 kados but not yet for records imported with zebraidx (like Tumer's example)
10:07 paul and I still don't have anything from you !
10:07 kados paul: tell him hello for me :-)
10:07 ha
10:07 paul (you told me you had some suggestions to do)
10:07 kados yes ... I can tell you right now
10:07 paul but is here if you prefer a chat
10:07 kados ok ... in fact I'll update the wiki
10:08 paul great
10:08 kados http://wiki.liblime.com/doku.p[…]catalogingproject
10:15 paul kados : can I add sara to the 2nd week list ?
10:16 kados paul: yes, though I'm not 100% sure she will spend all time with developers
10:16 paul no prob
10:16 kados paul: but certainly some :-)
10:29 paul: http://wiki.liblime.com/doku.p[…]catalogingproject
10:29 paul: I will expand it later today
10:29 paul: I need to head out for a bitt
10:29 paul: back in about an hour
10:40 pierrick I have the same problem Owen encountered 3 months ago : http://koha.org/cgi-bin/logs.p[…]:18&search=Search
10:40 branchcode in aqbookfund
10:40 not supported in HEAD it seems. Do I suppose it is a result of a too violent merge from rel_2_2 ?
10:44 hdl, you worked on this, can you help me? http://lists.nongnu.org/archiv[…]-01/msg00120.html
11:13 paul hdl ?
11:13 hdl oui
11:47 kados paul: hotel confirmed :-)
11:47 paul great.
11:47 for 3 ?
11:47 (in the triple I mean)
11:49 kados paul: how far is it from the hotel to the dev meeting location?
11:50 paul something like 15km
11:50 kados ahh
11:50 quite far then
11:50 paul yes, but hotel is in marseille center.
11:50 while CMI is in the north.
11:50 kados right
11:50 it's best I think
11:51 makes it easy to 'see the city'
11:51 after work :-)
11:51 paul & marseille center is interesting, while the marseille north is something to avoid generally !
11:51 kados hehe
11:51 (yes, the triple)
11:57 paul: about XUL MARC editor
11:57 paul: do you have a moment to discuss it?
11:58 paul yep ?
11:58 no
11:58 kados ok
11:58 paul I write a mail, & leave.
11:58 kados I will email you
11:58 paul i'll be there tomorrow, but not on friday.
11:58 kados ok
11:58 paul due to riot, antoine has a last course tomorrow, so he will begin really on monday

← Previous day | Today | Next day → | Search | Index

koha1