IRC log for #koha, 2008-01-19

All times shown according to UTC.

Time S Nick Message
11:32 CGI232 hello
11:32 CGI232 any one is there
12:40 gmcharlt paul: about?
12:58 paul gmcharlt_away: i'm back from lunch
12:58 paul hello gmcharlt, good morning to you
12:59 gmcharlt good afternoon, paul :)
12:59 gmcharlt I've responded to your e-mail
13:00 gmcharlt I'm working on an alternative approach to the items issue that should maintain both the authority of the items and all of the subfields you need
13:00 gmcharlt I'll send you and hdl a patch to look at by Monday at the latest
13:00 gmcharlt again, profuse apologies for my delay
13:01 paul ok, sounds fair. Even if I think this patch will be quite hard to write & not that necessay (we could just change the rule to "items authoritative where the field exist, otherwise it's marc")
13:02 paul + in fact, I thought you wanted to have items table authoritative for issuing things, not for marc ones. Otherwise, I would have warned sooner
13:02 paul (as i've met the problem already in 2.0 !)
13:03 gmcharlt paul: we'll see about how hard it is to write :) (and having two sources of item data seems like more work IMO, but tastes vary)
13:04 gmcharlt paul: the one thing I would see is that if any of those subfield not currently mapped need to do more than just be displayed or indexed (i.e., if they activate circ or acq functionality), I think we need to establish new columns for those
13:05 gmcharlt paul: do you have a small (<10,000 bibs or so), complete UNIMARC test database I could have a copy of?
13:05 paul you want a koha DB or just 10 000 biblios ?
13:06 paul (iso2709 I mean)
13:06 gmcharlt paul: a koha DB
13:06 gmcharlt with items
13:39 fbcit g'morning koha
13:40 owen Hi fbcit
13:40 fbcit gmcharlt around?
13:43 gmcharlt fbcit: leaving, back in 15 minutes
13:51 kados g'morning all
13:51 kados hiya fbcit
13:51 fbcit hey kados
13:52 fbcit kados: any thoughts on this....
13:52 fbcit [Fri Jan 18 03:24:26 2008] Makefile.PL: Warning: prerequisite Text::CSV 0.01 not found. We have unknown version.
13:52 fbcit [Fri Jan 18 03:24:26 2008] Makefile.PL: Warning: prerequisite Text::CSV_XS 0.32 not found. We have 0.23.
13:52 fbcit [Fri Jan 18 03:24:26 2008] Makefile.PL: Warning: prerequisite Text::Iconv 1.7 not found. We have 1.4.
13:52 fbcit using README.debian
13:52 fbcit on a fresh install of debian
13:52 kados well for Text::CSV, I couldn't resolve that myself
13:53 kados I think the version string for that module must be incorrectly exported
13:53 kados what happens if you go:
13:53 kados misc/perlmodule_ls.pl
13:54 fbcit biblios:~/koha.git# misc/perlmodule_ls.pl
13:54 fbcit Algorithm::CheckDigits -- undef
13:54 fbcit Array::Compare -- 1.14
13:54 fbcit Barcode::Code128 -- 2.01
13:54 fbcit Biblio::EndnoteStyle -- 0.05
13:54 fbcit Class::Accessor -- 0.31
13:54 fbcit Class::Accessor::Chained -- 0.01
13:54 fbcit Class::ReturnValue -- 0.55
13:54 fbcit DBD::mysql -- 4.006
13:54 fbcit Data::ICal -- 0.13
13:54 fbcit Devel::StackTrace -- 1.15
13:54 fbcit DublinCore::Record -- 0.03
13:54 fbcit GD::Barcode -- 1.15
13:54 fbcit HTML::Template::Pro -- 0.68
13:54 fbcit IO -- 1.23
13:54 fbcit MARC-XML -- ???
13:54 fbcit MARC::Charset -- 0.98
13:54 fbcit MARC::Crosswalk::DublinCore -- 0.02
13:54 fbcit MARC::Record -- 2.0.0
13:54 fbcit PDF::Reuse -- 0.33
13:54 fbcit PDF::Reuse::Barcode -- 0.05
13:54 fbcit POE -- 0.9999
13:54 fbcit Perl -- 5.8.8
13:55 fbcit Schedule::At -- 1.06
13:55 fbcit Sub::Uplevel -- 0.18
13:55 fbcit Test::Exception -- 0.26
13:55 fbcit Test::LongString -- 0.11
13:55 fbcit Test::Manifest -- 1.22
13:55 fbcit Test::NoWarnings -- 0.084
13:55 fbcit Test::Simple -- 0.74
13:55 fbcit Test::Tester -- 0.106
13:55 fbcit Test::Warn -- 0.10
13:55 fbcit Text::vFile::asData -- 0.05
13:55 fbcit Tree::DAG_Node -- 1.06
13:55 fbcit XML::RSS -- 1.31
13:55 fbcit I'm also having issues getting ZOOM to install
13:55 fbcit the parser is correct
13:55 kados hmmm
13:56 kados zoom has some dependencies
13:56 kados might want to just make sure you have yaz installed
13:56 kados yaz-client -V
13:56 fbcit biblios:~/koha.git# yaz-client -V
13:56 fbcit YAZ version: 3.0.20
13:57 kados hmmm
13:57 kados well, try installing Text::CSV from cpan
13:57 kados and paste in the ZOOM errors you're getting
13:58 fbcit t/2-ZOOM................ok 1/23                                              
13:58 fbcit #   Failed test 'connection to non-existent host 'no.such.host' fails'
13:58 fbcit #   at t/2-ZOOM.t line 20.
13:58 fbcit t/2-ZOOM................ok 5/23# Looks like you failed 1 test of 23.        
13:58 kados hmmm
13:58 fbcit t/1-Net-Z3950-ZOOM......ok 1/23                                              
13:58 fbcit #   Failed test 'connection to non-existent host 'no.such.host' fails'
13:58 fbcit #   at t/1-Net-Z3950-ZOOM.t line 22.
13:58 fbcit t/1-Net-Z3950-ZOOM......ok 8/23# Looks like you failed 1 test of 23.        
13:58 kados I wonder if someone introduced a test that includes a query to a non-existant host
13:59 kados fbcit: you on the yazlist?
13:59 fbcit mailing list -yes
13:59 kados I'd post your error there
13:59 kados for the ZOOM issue
14:00 kados Net::Z3950::ZOOM -- 1.14
14:00 kados is the version I'm running
14:00 kados and the latest is 1.21
14:00 kados so musta been something introduced between those versions
14:01 fbcit kados: I'll look into it more this afternoon. Gotta go now. tnx
14:30 lloyd_ if I set enrollment period to 0, is that forever?
14:34 kados lloyd_: I don't think so ... 999 will work though
14:34 lloyd_ ahh ok
14:34 kados lloyd_: so how do you like the 3.0 version ... compared to what you were running beore?
14:34 kados before I mean
14:35 lloyd_ one more, how do I change the category types. "Adult Person, Child Person, Pro, Org, Staff"
14:35 kados the types you can't change, but you can change the categories
14:35 lloyd_ I really like it..
14:35 lloyd_ mmmm
14:35 kados Administration -> Patron Categories
14:35 kados there are five 'types': Adult, Child, Professional, Institution, Staff
14:35 lloyd_ got time for a quick discussion on making it possible to change the types? :)
14:36 kados you can just change the categories
14:36 lloyd_ yeah.. but i need to change the types too
14:36 owen lloyd_: what do you mean you need to change the types?
14:36 owen What about it do you need to change?
14:36 hdl Can you give an example ?
14:37 lloyd_ sure... I dont want it to say adult patrol, child patrol etc etc etc as there will only be 2 types
14:37 lloyd_ Reader and Library Staff
14:37 lloyd_ so I just want to rename/remove some really
14:38 lloyd_ Reader, Organiation, Library Staff
14:39 kados just create a Reader category for type 'Adult'
14:39 lloyd_ nah i have reader categorys too
14:39 kados and a 'Library Staff' category for type 'Staff'
14:39 lloyd_ you're missing the point
14:39 kados oh
14:40 lloyd_ when somebody clicks new, on the patron screen I don't want adult, child, pro, org, staff to show
14:40 kados ahh
14:40 lloyd_ as we only have 3 types... normal readers, orgs and lib staff
14:40 kados OK, well, we can certainly modify that
14:41 kados it won't be free though :-)
14:41 lloyd_ that's fine.. pm?
14:42 kados we'll have to do that as a chance request, you can send a mail to Russel and we can get a quote for you on that :-)
14:42 kados and we probaly shouldn't talk about it on #koha much more soas not to bore the rest of the community :-)
14:43 kados s/chance/change/
14:46 owen It might even be a change you could do with custom javascript?
14:46 Brooke howdy
14:47 owen You could modify the DOM to replace the "new" drop down with your values to pass to the member entry script
14:47 kados owen: would that fit into the context of a usercss syspref?
14:47 owen userjs, yeah
14:47 owen Although I couldn't say for sure without really sitting down to try it out
14:48 kados hiya Brooke
14:48 Brooke :)
14:53 kados so guys, there was a post to koha-translate recently, proposing that we remove all ':' because it complicates translation
14:54 kados anyone have any thoughts on that?
14:55 paul kados : i've nothing against, except it will be a huge job to update french translation.
14:55 kados paul: I don't think so ... if that's the only change
14:55 kados paul: ie, you have already translated those strings, so it should just reduce the file size
14:56 paul yep, but i'll still have to update all strings one by one. but that's not a problem. just a long & boring stuff ;)
14:56 kados ?
14:56 kados just run update, shoudln't be any update of strings
14:56 paul all strings where the : disappear will be marked as "fuzzy"
14:56 paul and thus not parsed by the installer.
14:56 kados what is fuzzy?
14:57 paul hehe...
14:57 paul a string that has been modified (in place or content) is marked "fuzzy".
14:57 paul which means "check it again"
14:57 paul and all fuzzy strings are discarded by tmpl_process3.pl install
14:57 paul to avoid having a wrong translation
14:58 paul (at the price of having no translation at all)
14:58 kados hmmm
14:58 kados I still don't undersatnd :-)
14:58 paul so I explain again ...
14:59 paul we have a string like :
14:59 Brooke It's a QC measure
14:59 paul "enter the amount you want:"
14:59 paul which is translated to "Saisir la quantité voulue :"
14:59 paul if you remove the :, then the translator tools think it's a new stringe.
14:59 paul s/e//
14:59 paul and says :
15:00 paul "enter the amount you want" => MAYBE translated to "Saisir la quantité voulue : "
15:00 paul note the : have not been removed in french, as the translator don't want to do that himself
15:00 paul the "MAYBE" is written "fuzzy" in .po files
15:01 paul so, i'll have to manually say :
15:01 paul "enter the amount you want" => translated to "Saisir la quantité voulue "  
15:01 paul (remove the : & remove the fuzzy status)
15:01 kados but ... correct me if I'm wrong
15:01 kados if you already have a string:
15:01 Brooke it's a quality control measure
15:01 kados "enter the amount you want" => translated to "Saisir la quantité voulue "
15:01 kados it won't mark it as fuzzy
15:02 kados ie, if you're removing the ':' from one string, it should automatically find the other string that matches
15:02 paul tmpl_process3.pl update WON'T update the french string at all. the : will remain, so the process will mark the string as fuzzy.
15:03 Brooke because, presumably, you want a second set of eyes to see that the translation is accurate, yes?
15:03 kados strange, that's not my experience when I tested this
15:03 kados I'll have to test again
15:03 paul yes.
15:03 Brooke Because paul is clever :)
15:03 Brooke He does not want bad code
15:03 Brooke so he put in a quality control measure
15:04 paul it's not me, it's the translator tool & .po rules : when something changes on the left (original), mark the right (translation) as fuzzy for manual checking & confirmation.
15:04 kados Brooke: paul didn't write the PO standard ;-)
15:04 Brooke well, it's still clever
15:04 paul it's the case for any content change (event a case-sensitive change), or even, if I don't mind a location change
15:04 Brooke and so is paul
15:05 kados Brooke: :-)
15:05 Brooke he just doesn't get undue credit
15:05 Brooke sorry, paul, I have to steal your merit badge back
15:05 kados hehe
15:05 paul hehe
15:11 kados paul: OK, here's an example I will test:
15:11 kados #: ../../koha-tmpl/opac-tmpl/prog/​en/modules/opac-basket.tmpl:55
15:11 kados msgid "Authors"
15:11 kados msgstr "Auteurs"
15:11 kados #: ../../koha-tmpl/opac-tmpl/prog/​en/modules/opac-detail.tmpl:53
15:11 kados msgid "Authors:"
15:11 kados msgstr "Auteurs :"
15:11 kados #. %1$s: TMPL_VAR name=value
15:11 kados #: ../../koha-tmpl/opac-tmpl/prog/en​/modules/opac-sendbasket.tmpl:25
15:11 kados #, c-format
15:11 kados msgid "Authors: %s"
15:12 kados msgstr "Auteurs : %s"
15:12 kados so if I change opac-basket, opac-detail, and opac-sendbasket all to just 'Authors', what will happen to the french translation file on update? :-)
15:12 paul The 1st & 2nd one will be merged & I think you won't have a fuzzy.
15:12 paul where the 3rd one will result in a fuzzy
15:12 paul (my bet)
15:13 Brooke hi atz
15:13 atz greets Brooke
15:13 kados ahh, so it's just when we hae a %s that we hav a fuxxy?
15:13 kados fuzzy I mean
15:16 kados paul: ok, you're right:
15:16 kados #: ../../koha-tmpl/opac-tmpl/prog/​en/modules/opac-basket.tmpl:55
15:16 kados #: ../../koha-tmpl/opac-tmpl/prog/​en/modules/opac-detail.tmpl:53
15:16 kados msgid "Authors"
15:16 kados msgstr "Auteurs"
15:16 kados #. %1$s: TMPL_VAR name=value
15:16 kados #: ../../koha-tmpl/opac-tmpl/prog/en​/modules/opac-sendbasket.tmpl:25
15:16 kados #, fuzzy, c-format
15:16 kados msgid "Authors %s"
15:16 kados msgstr "Auteurs : %s"
15:17 kados so we did reduce it by one string
15:17 kados without any effort to translation into french, right?
15:17 paul yep.
15:17 kados (except if you want to remove the ':'
15:17 paul nope
15:17 paul i MUST remove manually the :
15:17 kados ahh
15:18 paul otherwise, the fuzzy string will be discarded by tmpl_process3.pl install
15:18 hdl or unfuzzy the string.
15:18 hdl (if you want to keep ":" Some French librarian prefers :)
15:18 paul (hdl : in fact, removing the : automatically removes the fuzzy status in kbabel)
15:18 hdl ':' )
15:19 hdl yes... Because you updates string.
15:19 hdl But a Ctrl+U unfuzzy it without touching the string.
15:20 paul (hdl : i prefer removing all : if the english version removes them.
15:20 paul )
15:21 hdl So do *I*
15:32 kados hey toins!
15:33 kados long time no see :-)
15:33 toins hi kados
15:41 Brooke hi toins
15:42 Brooke hi foxnorth
15:50 kados hi there matthew
15:50 matthew Hello kados
15:50 matthew I'm wondering about getting the latest updates to 2.2.9. Should I download from cvs?
15:52 matthew I'm going to do a fresh install on another server and move our current database to a new machine.
15:53 matthew I want to make sure I'm using the latest stable version with any bug fixes that may be important.
15:55 kados matthew: it's very likely that 3.0 alpha is more stable than 2.2.9 :-)
15:56 matthew I'm very glad to hear that!
15:56 matthew I've been waiting to check it out.
15:56 matthew I am using it for production in a small library.
15:58 kados cool
15:58 kados you may find the no-zebra option appealing then
15:58 kados makes the setup easier
15:59 matthew thanks for the tips.
17:27 hdl wow.
17:28 hdl kados : I found a curious bug in Search.pm
17:28 hdl It has to do with encoding and MARC::File::USMARC::decode
17:29 kados hdl: yea?
17:29 hdl our UNIMARC fields may be supposedly in latin1 for him.
17:29 atz hdl: is there an alternative module?
17:29 kados encoding--
17:30 hdl And items value linked with authorised values with accents are blanked out.
17:30 hdl 2 solutions :
17:31 hdl - displaying non matching value as such
17:31 hdl i.e $locations{$item->{location}}?$location​s{$item->{location}}:$item->{location}
17:34 hdl - using utf8::decode when getting fields line 1137
17:35 hdl my opinion is that orphan authorised values should always be displayed.
17:35 hdl our customers are often questionning when values disappear when the do a change in authorised values.
17:35 hdl But it is my opinion.
17:36 hdl sending patch for solution 2
17:42 kados hdl: OK
17:43 hdl sent.
22:18 fbcit kados around?

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