IRC log for #koha, 2005-06-15

All times shown according to UTC.

Time S Nick Message
13:33 paul_away kados, are you around ?
13:54 kados paul: paul yes
13:54 paul i'm trying to insert my iso2709 into zebra
13:55 kados great!
13:55 paul but can't find how to configure zebra.cfg
13:55 paul not great ;-)
13:55 paul could you explain how you do this ?
13:55 kados (phone call):-/
13:55 paul I may write a small SearchZebra.pm package for some tests if I succeed
13:56 paul ok, i'll wait.
13:56 paul (don't be too long, it's 9PM in France ;-) )
13:57 kados back
13:58 kados hmmm ... so here's how I did it
13:58 kados after install you should have a test directory
13:58 kados go to test/usmarc
13:58 paul right.
13:58 kados run ./test.pl
13:58 kados test.pl look in 'records' directory
13:58 paul test.sh
13:58 paul you mean
13:58 kados right
13:59 paul that's here :
13:59 paul recordType: grs.marc.usmarc
13:59 paul i think.
14:01 kados if you put an iso2709 file in records directory
14:01 paul indexing EMN...
14:01 kados it will index it (I hope)
14:01 paul it's on the way ;-)
14:02 kados good ... then it will automatically start a Z39.50 server
14:02 paul mmm...
14:02 kados (should take about 3 minutes)
14:02 paul 21:01:42-13/06 zebraidx(12685) [log] zebra_end_trans
14:02 paul Aborted
14:02 kados ahh ... probably because it's unimarc
14:02 kados :-(
14:02 kados k
14:06 paul http://blogs.zdnet.com/open-so[…]=feed&subj=zdblog
14:06 paul just for smiling
14:08 paul http://indexdata.dk/pipermail/[…]April/000203.html is the answer to our question
14:10 kados nice!
14:10 paul except that i get a
14:10 paul Search was a bloomin' failure.
14:10 paul in yaz-client :-(
14:10 paul still investigating
14:10 paul    [121] Unsupported Attribute Set -- v2 addinfo 'Bib-1'
14:12 kados hmmm ...
14:12 kados maybe you need a unimarc.abs?
14:41 paul kados ?
14:41 paul could you pls write a sample of a search on title
14:41 paul i try
14:41 paul find @attr 1=4 jauge
14:41 paul should have 1 answer (& only 1)
14:41 paul but get none
14:42 paul trying find title=jauge returns nothing either
15:09 kados paul: give me the server IP
15:09 paul opening firewall...
15:09 kados paul: I'll add it to the liblime z39.50 client
15:10 paul note it will be a sometimes-up server...
15:10 paul bureau.paulpoulain.com port 2100 as usual
15:11 kados paul: are you sure it's not port 9999?
15:11 paul should be up
15:11 paul zebrasrv -v 10 @:2100
15:11 kados ok
15:11 paul so should be on 2100 isn't it ?
15:11 paul try a query with title containg "jauge"
15:11 paul only 1 answer expected
15:11 paul or author=bellac
15:12 paul will return 4 results
15:12 kados paul: http://liblime.com/zap/advanced.html
15:12 kados seems unavailable
15:13 paul mmm... strange.... firewall open
15:13 paul (on port 2100 only)
15:14 kados I tried 2100 ...
15:15 paul iirc you added "VOYAGER" somewhere
15:15 paul i didn't do such thing
15:15 kados I used 'default' which is Zebra's default database
15:16 kados (unless you have different database listed in zebra.cfg
15:16 paul anyway, no log in my zebrasrv console
15:16 kados right
15:16 paul i've only :
15:16 paul profilePath: .
15:16 paul attset: bib1.att
15:16 paul attset: explain.att
15:16 paul recordType: grs.marc.unimarc
15:16 paul lockDir: lock
15:16 kados bureau.paulpoulain.com:2100
15:16 kados right?
15:16 paul setTmpDir: tmp
15:16 paul keyTmpDir: tmp
15:16 paul memMax: 4
15:16 kados yea ... all of that is right
15:17 paul yes, it's bureau.paulpoulain.com:2100
15:17 kados let's try by IP address
15:18 paul why do I get poul-poulain in
15:18 paul poul-poulain: unavailable
15:18 kados still not working
15:18 paul it's not poul but paul.
15:19 kados ahh ... typeo
15:19 kados yes it should
15:19 paul should change something ?
15:19 kados yep
15:19 kados now we get 'no hits' ;-)
15:19 kados you have a log?
15:20 paul yes
15:20 paul a long (launched with v 10 ;-) )
15:20 kados strange ... unavailable again
15:20 kados ok .. back up ;-)
15:20 paul i see new logs
15:20 kados still 'no hits'
15:21 paul i promise i havent done anything.
15:21 kados what is a keyword?
15:21 kados (maybe network timeout)
15:21 kados bellac?
15:21 kados no hits for it
15:21 paul a keyword you mean a subject ?
15:21 kados sure
15:21 paul theorie should be a subject
15:21 kados anything really
15:22 paul http://127.0.0.1:8080/cgi-bin/[…].pl?type=intranet
15:22 kados not finding anything it seems
15:22 paul with login test / test as usual. will give you usual Koha interface.
15:22 kados ok ... I'll log into my machine ;-)
15:23 paul & then go to bed ;-)
15:23 kados good idea
15:23 kados ;-)
15:24 kados hi David
15:24 kados need some help?
15:24 David yes please!
15:25 David I'm at the right place I take it?
15:25 paul for sure !
15:25 kados yep ;-)
15:25 David Having problems with Koha 2.2 on Windows XP
15:25 paul & you are speaking with previous & actual release manager !
15:25 kados hehe
15:25 David Getting a Premature end of script headers: isbnsearch.pl error when entering data
15:26 kados any mesages above that?
15:26 kados (that's just about as generic as error messages get ;-))
15:26 paul you mean you have 2.2.2b ?
15:26 David It says 'The server encountered an internal error and was unable to complete your
15:26 David request.
15:26 kados there should be some complaints on the lines above that line
15:26 kados how about above that?
15:27 kados script names, line numbers ... all helpful
15:27 David Apache logs say 'Can"t use an undefined value as a HASH reference at c:/usr/koha222/intranet/modules/C4/SearchMarc.pm line 255'
15:28 paul just enter an empty word
15:28 paul in Koha >> parameters >> empty word
15:28 kados aha! that's the problem
15:28 paul should REALLY be in bold & everywhere...
15:28 paul as this question is asked almost once a week.
15:28 paul (not your fault but our...)
15:29 David Sorry - I'm not actually on the PC Koha is installed on - it's at work and I'm at home at the moment!
15:29 paul no prob, but 99% sure it's the origin of your problem ;-)
15:30 David Just explain - what do you mean by just enter an empty word?
15:30 paul the, a, an, in...
15:30 paul are empty words in english
15:30 paul le, la, une, ... are empty words in french
15:31 owen stop words
15:31 owen /cgi-bin/koha/admin/stopwords.pl
15:31 paul oups, sorry, in french, it's "empty" words.
15:32 David how does that relate to the error?
15:32 paul us ppl should light on TV...
15:32 kados David: they're called 'stop words' in English
15:32 paul if they care about M. Jackson suing...
15:32 kados ahh ... owen caught it already
15:32 kados ;-)
15:33 paul david : if you don't have a stopword, then Koha bugs.
15:33 David ok - i'll give it a go when i get to work!
15:33 paul because the stopword hash structure contains nothing (different from "" in Perl)
15:36 David Thanks for the help. Maybe it will all become clear when I try it on the PC in 1/2 hr or so. I'll report back then. Thanks!
15:37 paul you're welcome.I'll be in bed anyway ;-)
15:38 kados paul: how many records are in the database?
15:38 paul 13000
15:38 kados thanks ... writing an email to the list now
15:38 kados about Zebra
15:39 paul writing a mail to zebra list & going to bed
15:39 kados cool ... have a good sleep
15:57 kados paul_bed: Sebastial responded already ;-)
15:57 kados paul_bed: Sebastian even ;-)
16:07 kados chris around?
16:47 sanspach I searched logs and bugs but didn't see anything: has there been talk of isbn-13/ean ?
16:51 kados hey sanspach
16:51 kados what about isbn-13/ean?
16:51 kados (is that different that isbn?)
16:52 sanspach the isbn is moving to 13 digits and will now conform to standard for EAN
16:52 kados ahh ... no we haven't addressed any potential issues with that on koha-devel
16:52 kados might be good to bring it up
16:52 kados can you do that?
16:53 sanspach I was poking around for checkdigit (I want to do a different checkdigit algorithm)
16:53 sanspach and noticed ISBN for 10-digits
16:53 sanspach already have code for 13 digits/EAN as well as conversion
16:53 sanspach I'll try to write something
16:54 sanspach probably should make two different suggestions, though
16:55 kados cool
16:55 kados thanks!
16:57 owen When the question of isbn-13 was brough up on the Koha list on 3/28/2005, the response from Stephen was "the ISBN field can actually store 14 characters."
16:57 owen Currently Koha doesn't have any built-in ISBN checks
16:57 sanspach there are indications that anyone programming for 13 should go ahead for 14
16:58 sanspach but then there's the question of what to do with that extra character in the meantime
16:58 sanspach the isbn check subroutine isn't used?  (I'm very new to Koha)
17:00 owen No
17:02 sanspach ah, thus bug 906 requesting that there be a check (duh!)
17:21 kados chris around?
17:21 rach hello
17:21 kados hey rach
17:21 kados sorry about the bugsquashing session time ;-)
17:21 rach :-)
17:22 kados it's the only time I could get paul, slef, hdl, and genji to commit to
17:22 kados I hope the Search Meeting suggested time is better
17:23 rach but I've put in a couple of bug reports over the last week
17:23 kados yep ... saw those
17:26 rach There is a bit of a disconnect I think around acquisitions because the major developers/their libraries don't actually use that module
17:26 kados yea ... NPL doesn't use it at all
17:26 rach but I figure I should notify the things I think are bugs so that someone can scream if they don't think they are :-)
17:26 kados so I'm not really familiar with the issues
17:27 kados yep ... I suspect paul will be glad to optionize areas where libraries disagree
17:28 kados the Zebra Searching email is pretty important rach
17:28 rach I don't think any of pauls clients use acquisitions either - but if they do, then I'm keen to hear wether the problems that I think oculd arrise is actually do in real life
17:28 kados (just in cause you thought maybe it was just a techie thing)
17:28 kados if we go with Zebra we'll have to seriously look at what to do with non-MARC libraries
17:30 rach yep
17:32 rach idiot question perhaps but why? I thought (so I must be wrong on this) that non-marc libraries basically are just choosing not to see the Marc? But that the fields etc can still be linked up? or is that not right?
17:32 kados yea they can be
17:32 rach I thought they were by default?
17:33 kados but that may not be the best way to approach zebra
17:33 kados not sure as i don't have a non-marc koha installed anywhere
17:33 rach because even the people who don't want to see marc, still want to be able to export their data as marc in the future etc - I might be off base on that, or we may not have gone where I had in mind
17:34 rach but my thought/intention with MARC being on or off was that it should be largely just a matter of templates
17:34 kados yep ... though I'm not sure if that's the case
17:34 kados chris should know
17:35 rach you can turn marc off and on in a library fairly happily :-)
17:35 rach but yep the reality may not live up to my ideal :-)
17:35 kados hehe ... true
17:42 rach chris isn't in today unfortunatly
17:51 rach joshua are yo a list admin for the koha-devl list?
17:57 kados yea I can be
17:57 kados need something done?
02:35 paul someone testing ZAP! with paul-poulain at the moment ?
02:54 paul salut francoisl
02:54 paul je suis en train de rédiger un mail pour la liste infos.
02:55 paul marqué [important]...
04:28 David Paul: Thanks very much for the help this morning (my time). The problem was with the 'Stop words' being empty. Never would have found that myself!
04:28 paul right.
04:28 paul that's why we really should write it somewhere...
04:28 paul no, everywhere...
04:29 David Yes - it certainly would help!
04:29 paul have a good night if i don't mind ;-)
04:29 paul (almost lunch time for me)
04:29 David 9.30pm here (New Zealand) - where are you?
04:30 paul in France
04:32 David I've never been on this IRC channel before and only just tried using Koha, so it's the first time I've chatted with Paul
04:33 David My partner is getting suspious - me chatting online and all (he he...)
04:33 David I'd better go!
04:33 paul i have to go too
04:33 David Thanks again
06:02 kados paul: looks like you got your records indexed. congrats!
06:04 kados paul: also, note that I'm getting some hits from the Zap client
06:04 kados but the records aren't being displayed
06:11 paul kados : 'morning.
06:11 kados hi paul
06:13 paul how do you do to retrieve records.
06:13 paul because I see then in yaz-client
06:13 paul s/then/them/
06:14 paul Z> format xml
06:14 paul Z> show 1
06:14 paul Sent presentRequest (1+1).
06:14 paul Records: 1
06:14 paul [Default]Record type: XML
06:14 paul <unimarc><001>8058</001><010><  ><a>2222040264</a></  ></010><090><  ><9>7591</9><a>7591</a></  ></090><101><  ><a>fre</a></  ></101><200><1 ><a>Des ph\XE9nom\XE8nes critiques aux champs de jauge</a><e>une introduction aux m\XE9thodes at aux applications de la th\XE9orie quantique des champs</e><f>LE BELLAC, M</f></1 ></200><210><  ><5>CNRS Editions</5><a>Paris</a><c>CNRS Editions</c><d>1988</d></  ></210><215><  ><a>369 p</a><d>25 cm</d></  ></215><225><
06:14 paul Savoirs actuels</a></  ></225><410><  ><t>Savoirs actuels</t></  ></410><606><  ><9>3419</9><a>NUCLEAIRE</a></  ></606><606><  ><9>4229</9><a>PHYSIQUE NUCLEAIRE</a></  ></606><606><  ><9>5457</9><a>PHASE DE TRANSITION</a></  ></606><606><  ><9>4920</9><a>THEORIE DES CHAMPS</a></  ></606><606><  ><9>4689</9><a>RENORMALISATION</a></  ></606><686><  ><a>LIVR</a></  ></686><700><  ><9>8520</9><a>LE BELLAC, Michel</a></  ></700><990><  ><a>N 8
06:14 paul BEL</a><e>Nucl\XE9aire (Orange)</e></  ></990><995><  ><9>10795</9><b>CDI</b><c>CD​I</c><e>SL</e><f>INVENTAIRE 6 FAIT LE 29.07.03</f><j>012676</j><k>N 8 BEL</k><o>0</o><u>&lt;b&gt;OUVRAGE RECHERCHE&lt;/b&gt;</u></  ></995>
06:14 paul <idzebra xmlns="http://www.indexdata.dk/zebra/">
06:14 paul    <size>841</size>
06:14 paul    <localnumber>7598</localnumber>
06:14 paul    <filename>catalogue/EMN.iso2709</filename>
06:14 paul  </idzebra>
06:14 paul </unimarc>
06:14 paul which is correct.
06:14 paul do you know how to retrieve iso2709 files ?
06:15 kados sure ...
06:15 kados there are two stages to a z39.50 request
06:15 kados search
06:15 kados present
06:16 paul the "show 1" you mean ?
06:16 paul (in yaz-client)
06:16 kados yes
06:16 paul I tried : format usmarc
06:16 paul show 1
06:16 kados except that you can 'show 1-50'
06:16 paul and got :
06:16 paul Z> format usmarc
06:16 paul Z> show 1
06:16 paul Sent presentRequest (1+1).
06:16 paul Records: 1
06:16 paul [Default]Diagnostic message(s) from database:
06:16 paul    [238] Record not available in requested syntax -- v2 addinfo ''
06:16 paul nextResultSetPosition = 2
06:16 paul Elapsed: 0.000816
06:17 kados ahh ... yes ...
06:17 paul any idea welcomed
06:17 kados http://liblime.com/zap/try.html
06:17 kados put your server:port/database (default is fine)
06:18 paul works fine, except in usmarc format :-(
06:18 kados right
06:18 kados well remember that zebra
06:19 kados relies on the data as a file
06:19 kados and retrieves it that way
06:19 paul ???
06:19 kados zebra creates its own index
06:19 kados but still retrieves the records from the file
06:20 kados (you can have zebra store internally the data but not by default)
06:21 paul and then ?
06:21 kados http://www.indexdata.dk/zebra/[…]ating-records.tkl
06:21 paul (is there something related to the show format ?
06:21 kados maybe ...
06:24 kados http://www.indexdata.dk/zebra/[…]tocol-support.tkl
06:25 kados http://www.indexdata.dk/zebra/doc/formats.tkl
06:25 kados this last one maybe
06:26 kados we may need to setup a .map schema ... but not sure
07:23 paul kados : got it !
07:24 paul strange...
07:24 paul works fine in explicit usmarc format in try.html
07:25 paul but still wrong in advanced.html
07:27 paul in try.html, the following display are OK : SUTRS, USMARC, XML
07:27 paul the other display are empty : GRS-1, danmarc, HTML
07:28 paul kados, do you want that I commit some of my files into CVS ?
07:38 kados sure
07:38 paul any idea on try.html working & not advanced.html ?
07:38 kados even better would be a README for how to setup Zebra with a Koha catalog eh?
07:38 paul ;-)
07:38 kados it's very strange
07:39 paul tomorrow, i'll try to share some SearchMarcZebra.pm package.
07:39 kados paul: one important thing to consider is switching to utf-8
07:39 kados paul: that'd be great!
07:39 paul for sure !
07:39 paul I begin to understand how zebra config files works. should not be too hard.
07:40 paul 1- query zebra
07:40 paul 2- retrieve records
07:40 paul 3- find biblionumber
07:40 paul 4- do as usual.
07:40 kados yes ... perfect for a test case
07:40 kados but eventually I think we should do:
07:40 kados 1- query zebra
07:40 kados 2- retrieve records
07:40 paul will be a P.O.C., as it's silly to retrieve records then retrieve them again in Koha ;-)
07:41 kados 3- display to user
07:41 kados right ;-)
07:41 kados the only thing MARC doens't have is status info
07:41 kados which can be easily obtained from koha tables
07:41 paul should not be to hard to embeed in the MARC record.
07:41 paul (in UNIMARC recommandation 995, there is a subfield for this)
07:41 kados really! that's good news
07:42 paul (unused yet for koha users)
07:42 kados although I do want to preserve transactions in koha tables
07:42 paul http://www.adbdp.asso.fr/outil[…]ogestion/r995.htm
07:42 kados I think we have two databases:
07:42 paul right. we will have to invetigate to see if it's better to have transactions in Koha or in zebra.
07:42 kados koha database handles information on the movement of items
07:43 paul anyway, at 1st glance, my feeling is that we still need raw marc in Koha.
07:43 kados zebra handles textual elements of MARC records
07:43 kados I hope transactions in Koha will be better
07:43 kados (otherwise, what's the point of using koha ;-))
07:43 paul ;-)
07:44 paul in reco995, issuing informations are in :
07:44 paul $m - Date de prêt ou de dépôt
07:44 paul $n - Date de restitution prévue
07:44 paul $o - Catégorie de circulation
07:44 paul (issuing date / return date / issuing category)
07:44 kados cool
07:44 paul my idea could be to have :
07:45 paul * zebra for search & results list
07:45 paul * koha for MARC editor and when MARCdetail.pl called.
07:45 kados why when marcdetail called?
07:46 paul because we already know the $bibid, so no need to query zebra
07:46 paul + we already have issuing status in mySQL
07:46 kados but ... we already have the whole marc record too
07:46 kados so why re-query mysql at all?
07:47 paul yes but no. where is the marc record coming from zebra once you have sent opac-searchresult.pl to the user ?
07:48 paul as we are in http, what you have found is in a process that is dead.
07:48 kados not if we use DOM
07:48 paul so, when the user clics on "see detail", you need to find the MARC data again isn't it ?
07:48 kados nope ...
07:48 kados just have a 'tab'
07:48 kados when they click on it, it displays
07:48 kados no reload
07:49 paul and when you get 150 results from the query ? you DOM'ed the 150 biblios ?
07:50 kados here's an example:
07:50 kados http://catalog.cincinnatilibra[…]5Ewords+or+phrase
07:50 kados (not sure if this link works ... if not:
07:50 paul works
07:50 kados great
07:50 kados click on catalog record
07:51 kados oops ... now I understand what you mean
07:51 paul (in konqueror, the marc is always here ;-) works fine in ff)
07:51 paul ;-)
07:51 kados and I forgot that your default view is MARC
07:51 kados whereas NPL is not
07:52 paul my problem is not here.
07:52 kados so I thought you meant to re-query after details page is already loaded
07:52 kados but you mean on the initial search results page _> details page
07:52 paul it's that when you do a search, you 1st get a result list, then a detail
07:52 kados right ...
07:52 paul http://catalog.cincinnatilibra[…]MAIN/59160050/5/0
07:53 paul then >> clic detail (& read the MARC record once again)
07:53 kados what is the search term?
07:53 paul lord ring in title
07:54 kados (the link is session-based and so it didn't work)
07:54 kados ok
07:55 kados well maybe we should re-invent this function
07:55 kados do search
07:55 kados return result list
07:55 kados user clicks on detail
07:55 kados details pop down below record
07:55 paul even in this case, you can sent all the records in 1 html page.
07:55 kados avoiding page reloads
07:56 kados no ... only the first 10
07:56 paul (in case there are only 20, you will have html rendering very long !)
07:56 kados hmmm
07:56 paul & http traffic too heavy on a low bandwidth
07:57 kados are you sure?
07:57 kados a marc file is quite small
07:57 paul the html MARC editor page (for a single MARC record) can be >50KB
07:57 paul (with all the authorised values lists I agree)
07:58 paul but my opinion is that we will have a better way (& an easier one) to use zebra for what it's done for : indexing & searching.
07:58 kados yep ...
07:58 kados (above this I was just brain-storming ;-))
07:59 paul right. we all are brain storming ;-)
07:59 kados hehe
07:59 paul (beware the headake)
07:59 paul headake ?
07:59 paul (wrong spelling i bet)
07:59 kados headache I think
08:00 kados :-)
08:03 kados paul: do you run multiple Koha's on the same machine regularly?
08:04 paul no : always.
08:04 kados right
08:04 paul go to bureau.paulpoulain.com : at least 4 public Koha. + 22+head + 2 private atm
08:04 kados and all of them from CVS repos?
08:04 paul yes
08:04 paul (from rel_2_2)
08:04 kados do you use installer.pl to create a new install?
08:05 paul (except head, of course)
08:05 kados or do you manually install?
08:05 paul no more installer.pl on my computer.
08:05 paul but when doing a public install, I use installer.pl of course
08:05 kados right
08:05 kados I'm thinking it would be nice to modularize Koha a bit more
08:05 paul + mysqldump the DB from my server & mysql <db.sql on the dest server.
08:06 paul like ?
08:06 kados the current intaller is pretty limited
08:06 kados well /etc/koha.conf
08:06 kados /etc/koha-httpd.conf
08:06 kados always go in the same place
08:07 kados and I can't have say /etc/httpd/conf.d/koha-client1.conf /etc/httpd/conf.d/koha-client2.conf, etc
08:07 paul why not. the 1st test to remove in this case is the "stop installer if /etc/koha.conf exists)
08:07 kados right
08:07 paul should not be too hard to do. But a little bit tricky for updater...
08:07 kados but also we should have an installer.conf
08:07 paul we already have
08:07 paul (auto_install file)
08:08 paul where you can set all parameters & have an automatic install
08:08 kados right ... but still some limitations
08:08 paul only 1 that i see
08:08 kados like I can't choose where to put koha-httpd.conf koha.conf (or what to call them)
08:08 kados I can't choose log file names, etc.
08:08 paul no way to specify sql files to upload.
08:09 paul right, but it's not auto_install limit, it's installer.pl limit !
08:09 kados right
08:09 kados I should just read through installer.pl ... but I suspect there are some PATH muckings in there as well
08:10 paul maybe. I must say i'm not a .sh pro
08:10 kados also ... a question
08:10 paul so i just created auto_install, using what had been written before.
08:10 kados would it be a good idea to run multiple Kohas off the same script files?
08:10 paul ???
08:10 kados (changing templates of course)
08:11 kados I know if sounds crazy ;-)
08:11 kados s/if/it/
08:11 kados I'm just thinking that it might be nice to have:
08:11 kados /build/koha/opac
08:11 paul (no, it's just i don't undestand. not crazy ;-) )
08:11 kados /build/client1/opac-tmpl
08:11 kados /build/client2/opac-tmpl
08:12 kados /build/client1/etc/koha.conf
08:12 kados /build/client2/etc/koha.conf
08:12 kados /var/lib/mysql/client1
08:12 kados /var/lib/mysql/client2
08:12 paul that's already possible (manually) to share .pl & .pm but have spectific templates for each library, as well as parameters & sql, of course
08:12 kados but all of the clients use the same /build/koha/opac, intranet, etc.
08:12 paul (as templates path is in /etc/koha.conf iirc)
08:13 kados right ... just wondering if that's something you've played with
08:13 paul no for templates, as i only deal with css in opac
08:13 paul (& a stylesheet if requested)
08:14 kados i would also like to create an installer for the CVS repo
08:14 kados actually ...
08:14 paul you have too many ideas...
08:14 paul ;-)
08:14 kados it might be nice if we had installer.pl and cvs-installer.pl
08:15 kados :-)
08:15 paul (you will have to delay some of them imho)
08:15 kados cvs-installer.pl would grab the latest cvs repo and install from there
08:15 kados (of course)
08:21 kados FrancoisL: hi there Francois-Laurent
08:21 kados FrancoisL: what's new with you?
09:03 kados slef: evening
09:04 kados slef: are any of the times paul listed in his response to the Searching Meeting email good for you?
09:15 FrancoisL Hi kados ! How are you ?
09:16 paul hi again francoisl ;-)
09:16 FrancoisL paul: salut
09:17 FrancoisL kados: I've been with the SAN last week - things are clear now
09:17 FrancoisL I have a list of all modifications they want (not a long list !)
09:17 FrancoisL We are hiring a Koha developper to help them.
09:18 paul did you plan to be here for "Koha 2.4 Searching Group Meeting" ?
09:18 paul VERY important, imho.
09:19 kados yes
09:19 kados hi FrancoisL
09:19 kados :-)
09:19 FrancoisL I agree, but 1-I'll be in Grenoble for a congress 2- 11pm is much too late for me :(
09:19 paul the date will change, for sure.
09:19 kados yep
09:19 kados what time can you make it FrancoisL ?
09:20 FrancoisL OK so I'll make an effort. Kados: any time before 10pm local time I can manage...
09:20 FrancoisL I'll talk to the wife :)
09:20 kados ok ;-)
09:20 paul (we are GMT+2 joshua)
09:21 FrancoisL (we are you+6 joshua)
09:21 kados right ;-)
09:22 FrancoisL Moday is a no-no for me as I teach in an architecture school (5 hours without pause...)
09:22 FrancoisL s/Moday/Monday
09:22 kados Thur 16? Fri 17?
09:23 paul thur16 ok, but not before 9PM GMT
09:23 kados Tue 21?
09:23 FrancoisL Nope : I'll be in Grenoble (16-20th)
09:23 kados so it looks like Tue 21
09:23 FrancoisL ..and so well be Paul :)
09:23 kados for Paul and FrancoisL
09:23 FrancoisL OK for Tue 21. What time ?
09:27 kados FrancoisL: http://tinyurl.com/8mbfc
09:42 slef kados: finding email.
09:44 slef kados: all look possible.
09:45 kados slef: great ... thanks!
09:51 owen Bug Squashing this morning?
09:56 slef http://raw-output.org/20050614[…]ranslators-insane
10:00 kados BugSquashing Session
10:01 kados hi everyone
10:01 kados who's here?
10:01 kados owen and slef I know
10:01 kados paul still with us?
10:01 kados hdl?
10:01 kados here's the list of bugs:
10:01 kados http://tinyurl.com/as66z
10:02 owen Where do you want to start?
10:03 kados well last time we did the '900s'
10:03 kados but I think pauls method is better
10:03 kados start with the ones that are higher priority
10:03 kados I have 893 listed first
10:03 owen You mean by severity?
10:03 kados yep
10:03 kados http://tinyurl.com/as66z
10:04 kados does that link not auto-sort for you by severity?
10:04 kados slef joining us?
10:04 owen Yes
10:04 owen It does
10:04 kados cool
10:04 kados paul joining us?
10:04 paul (on phone)
10:04 kados it won't be much of a bug squashing session with owen and I ;-)
10:05 paul (was on phone with francois L. it's done now)
10:06 kados sweet ... ok ... I guess that makes quorum
10:06 kados so
10:06 kados http://tinyurl.com/as66z
10:06 kados list of bugs
10:06 slef also, tinyurl.com links aren't obvious what they are
10:07 slef so I mostly ignore them
10:07 kados true
10:07 owen slef, aren't we among friends here? Couldn't you give kados the benefit of the doubt?
10:07 kados maybe I'll annotate them from now on
10:07 kados anyway ...
10:07 kados we've got 66 bugs
10:07 kados bug 893: Can't get past biblio info page in simple acquisitions
10:07 slef owen, it was a quick glance at scrollback once I got back online.
10:08 owen I don't see bug 893 happening for us... Can anyone duplicate it?
10:08 paul i think it's a MARC=OFF problem
10:08 slef 893 is marked Windows 2000
10:08 paul so, it's for chris ;-)
10:09 owen Oh, right.  I see
10:09 slef I guess we need info from the server log.
10:09 owen She said 'simple acquisitions,' so I assumed she meant acqui.simple
10:09 slef paul: has anyone got a MARC=OFF system to hand?
10:09 slef erm, not sure why that said paul:, sorry
10:09 paul in new zealand, yes.
10:10 kados so I'll try to get more info from chris on this one
10:11 kados 857 is assigned to paul
10:11 kados paul: is that still right?
10:11 paul yes
10:11 kados ok ...
10:11 paul I think i've a simple solution to fix it.
10:12 paul i'll let the url field with http://site1.com|http://site2.org
10:12 paul and split in opac-detail to have correct URLs
10:12 paul so, now you can expect a fix in 2.2.4
10:12 kados ok cool
10:12 kados (can't make it for 2.2.3?)
10:12 paul no, because 2.2.3 is for friday on monday ;-)
10:13 kados ok ;-)
10:13 kados bug 472: It should not be possible to delete an item which is on issu
10:13 paul it's ready, translation to french is OK, i'll install a french library in 2 hours for RC tests
10:13 paul nothing to add, it's a bug to fix ;-)
10:13 kados and assigned to paul
10:14 paul yep. I'll take care of it... (nervous smiling, gun in hand...)
10:14 kados hehe
10:14 slef bugs.koha.org just went dead on me for a minute, sorry for being quiet
10:14 kados Bug 670: "Waiting" status often inaccurate
10:15 paul owen ?
10:15 owen Yes?
10:15 paul 670 is yours ;-)
10:16 paul should we let this bug open ? or close it ?
10:16 kados owen: do you consider it fixed?
10:16 owen I think there's one last major place where this bug needs work: Circulation
10:16 owen We need to get information into the circ template about holding branch versus pickup location
10:16 kados who was responsible for the fixes on the opac (does anyone know?)
10:18 owen I made some changes in some places...not sure if I worked on the opac or not
10:18 kados owen: do you want to investigate further?
10:18 paul + working on .pm will impact opac as well as librarian interface.
10:18 owen I'm at the limit of my abilities with circulation.pl
10:18 owen I need someone else to take a look
10:19 kados ok ... I'll assign it to myself
10:19 kados Bug 915: opac-MARCdetail.pl incorrectly shows request status
10:20 paul owen, is it fixed ?
10:20 owen Not the last time I checked, but let me look again
10:21 kados ok ... in the meantime
10:21 kados 984: Bug in installer appears to not correctly escape password
10:22 kados slef: your bug right?
10:23 slef yes
10:23 kados Bug 834: Add Category Fields Need Descriptions
10:23 slef I think that's a fix, but
10:23 slef oh, never mind then :-/
10:23 kados ;-)
10:23 kados so is it fixed?
10:24 kados sorry, I thought maybe you were still working on it
10:24 owen Bug 915 is not fixed
10:24 slef I think that's a fix and committed to HEAD, but not 2.2 and submitter seems to have died.
10:24 kados submitter?
10:25 slef Cory, who reported the problem.
10:25 kados right
10:25 kados ok ... (thought you were talking about submitter software ;-))
10:25 slef paul: can you consider that for 2.2?
10:25 kados (and hopefully Cory's not dead ;-))
10:25 paul yes, i'll give a try.
10:26 kados great ...
10:26 kados paul look at 915 too?
10:26 paul yes
10:26 kados Bug 834: Add Category Fields Need Descriptions
10:27 slef kados: why are you jumping around the bug list?
10:27 owen yeah, that's skipping ahead I think
10:27 kados hum
10:27 slef not just this one, but lots
10:27 paul not skipping for me.
10:27 kados not for me either
10:27 paul bugs ordered by severity
10:28 paul bug 834 : there is an online help, should be enough imho
10:28 kados ok ... so we close it with a pointer to the online help?
10:28 owen I think the normal bugs aren't getting sorted consistently, that's all.
10:29 owen kados, you lead us through the list
10:29 kados ahh ... probably not ... good point
10:29 kados ok
10:29 paul that's my opinion, we can close 834
10:29 kados ok ... go ahead and close it but include a note about online help
10:29 paul you do ?
10:30 kados no
10:30 paul I do ?
10:30 kados yes (if that's ok)
10:30 paul done
10:30 kados ok ... next on my list is
10:31 kados 857: Cannot link to multiple web site addresses from 856 tag
10:31 paul already done ;-)
10:31 kados (I reloaded my page and now everything's ordered differently)
10:31 kados sorry
10:31 kados :-)
10:31 owen Thanks, Bugzilla! :)
10:31 paul pffff... not fun...
10:31 kados 856: subscription renewal changes the start date in OPAC
10:32 paul it's fixed (2 days ago)
10:32 kados great ... please close it
10:32 paul done
10:32 paul (& you proofread the page)
10:32 kados (I will)
10:32 paul (no, you already proofread the opac page)
10:32 kados 878: updatedatabase breaks parameters (author and title searches
10:33 kados (oh ... yes I remember now) ;-)
10:33 paul fixed.
10:34 paul if I don't mind.
10:34 kados right ... go ahead and close it
10:34 kados 885:; Non-ISO8859-1 charset support broken in CSS theme in 2.2
10:35 paul bugs.koha.org off...
10:35 slef sounds interesting to me, but bko just gone off again
10:35 kados yep I notice it too
10:35 paul up again...
10:36 slef anyone up for replacing bko for 2.4? ;-)
10:36 paul 858 requireds a lot of work.
10:36 paul the best solution being to really move to utf-8
10:36 kados bko?
10:36 paul bugs.koha.org
10:36 owen bugs.koha.org
10:36 kados I agree ... utf-8 is probably the best way to go
10:37 kados slef: what do you propose?
10:37 paul (on phone)
10:37 owen Would an upgrade improve things?
10:37 slef kados: give me 5 or 10 minutes at end of bugsquash? (or next time it goes offline)
10:37 kados sounds good
10:37 kados so 858 ...
10:37 slef How do we go utf-8?
10:38 kados no idea ... :-)
10:38 kados my knowledge of charsets is pretty poor
10:38 kados Encode is one way I've done it in the past
10:38 slef I guess we add the http-equiv and I make the next installer set the headers in koha-httpd.conf
10:38 owen Wouldn't the same bug require utf-8 in the database?
10:39 owen Is the bug complaining about the templates, or the database, or both?
10:39 slef yes. I guess paul will have an opinion about that
10:39 kados slef: have you ever used Template::Toolkit?
10:39 slef bug is complaining about templates afaict?
10:39 paul (on phone)
10:39 kados paul: we'll wait for you
10:39 slef kados: Probably. Don't know.
10:39 kados let's move on to 849: Error when generating barcodes
10:39 kados till paul gets back
10:40 owen Hmmm... I haven't tried that in quite a while
10:40 owen No idea if it's still current
10:40 kados I must admit I've never tried to print barcodes
10:40 owen Anyone else playing around with barcodes?
10:41 kados i can look into it
10:41 kados It's something I should know anyway
10:41 kados 926Many form tags lack method and/or action attributes
10:41 owen We talked about that one last time.
10:41 kados right
10:41 owen Isn't it assigned to me now?
10:42 kados I think so
10:42 kados bugzilla down it seeems ;-)
10:42 kados actually, it might be NZ bandwidth issue too
10:42 paul back
10:42 kados great ... so back to 858 then
10:42 slef sorry, I just got a call on a line I thought I'd switched off
10:43 paul utf-8 is a 3 way plan i think :
10:43 paul * utf-8 templates
10:43 paul * utf-8 datas
10:43 paul * utf-8 translations
10:43 paul moving templates to utf-8 requires use of a tool like html tidy if i don't mind.
10:44 owen utf-8 datas is the biggest question mark isn't it?
10:44 paul plus some manual changes
10:44 kados paul: have you used Template::Toolkit?
10:44 paul No kados. utf-8 datas means, I think : dumping datas, changing mySQL & reloading datas
10:44 kados right
10:44 paul for translation, it's not clear.
10:44 paul anyway, it's not a 2.2.x task.
10:44 kados but we still need to figure a way to have utf-8 for templates
10:45 kados in addition to data
10:45 kados in fact, it might be possible to "pretend" we have utf-8 in the data
10:45 kados by using an encode when we pull data out
10:45 kados so maybe this bug is really a feature request?
10:46 kados (or at least it suggests a feature request)
10:46 slef or maybe it's a WONTFIX until 2.4?
10:46 kados right ...
10:46 paul maybe we can have a solution to fix it without everything in utf-8
10:46 kados so there's a discussion I'll have to schedule on encoding
10:47 paul i think we need acli opinion.
10:47 kados acli?
10:47 owen The bug reporter
10:47 kados right
10:47 kados ok ... so slef would you look into this one?
10:47 slef either way, that's more than one bug
10:47 paul acli is also tmpl_process3 author (the tool for translations)
10:48 slef kados: yes.
10:48 paul we have to investigate correctly this question, because iso2709 char encoding is poor for instance
10:48 kados ok ... go ahead and assign it to yourself
10:48 paul (awful in fact ;-) )
10:48 kados yes!
10:48 slef iso2709 = marc?
10:48 kados it's really abd
10:48 kados bad even
10:48 slef paul: Is that OK with you?
10:48 paul iso2709 is the technical form of a MARC file.
10:48 paul it's OK.
10:49 kados ok ... shall we move on?
10:49 paul ok for me
10:49 kados I can write an email to the list about encoding issues later today
10:49 kados 247: we should split city and state in borrower info
10:49 paul nice.
10:49 kados is this fixed?
10:49 owen 247 is really part of a larger overhaul of the borrower database
10:50 kados ok ... so it's still a prob
10:50 kados owen: can we link 247 with other bugs?
10:50 owen Paul weren't you interested in changing the way the borrower's table looked?
10:50 paul it's not a bug for frenchies ;-)
10:50 paul right owen.
10:50 paul I think we need to clarify borrowers table
10:50 owen paul: right, but it's a larger issue in terms of i8n
10:50 paul because fields names are not clear.
10:51 kados so ... another discussion on koha-devel it sounds like eh?
10:51 paul yep.
10:51 kados I'll start that one tomorrow ;-)
10:51 paul just 1 example :
10:51 owen Well, kados, it's something to tackle along with NCIP
10:51 kados 871: Rename OPAC CSS template 'default'
10:51 kados owen: right ... I'll talk to Art about it
10:52 paul Art ?
10:52 owen If we're going to have a new default template for 2.4, then 871 is moot
10:52 paul right.
10:52 paul so "wontfix" ?
10:52 owen But right now the default opac template is unusable
10:52 kados why's that?
10:52 paul yes, I should remove it from official releases.
10:53 kados right ...
10:53 kados also: remove sample data from official release paul
10:53 kados the sample data is for 1.2 Koha
10:53 paul that's done since 2.2.1 iirc
10:53 kados ok ... just checking
10:53 owen Have we replaced it with something?
10:53 kados paul: change 871 to WONTFIX ?
10:54 paul no owen & that's a shame I agree.
10:54 paul if someone has a sample & working MARC21 english DB, let me know !
10:54 kados I could contribute some data I think (maybe in time for 2.2.3)
10:54 kados ok ...
10:54 paul put it in misc/sql-datas/english
10:55 kados 612: Renewal failure gives no feedback
10:55 paul & the user can upload it.
10:55 kados paul: ok
10:56 kados paul: did you already fix 612?
10:56 paul no (reading it)
10:57 paul no, i did nothing on this.
10:58 kados is it interesting to you?
10:58 owen I'm wavering on whether this is still relevant
10:58 owen We've managed to get the renewal checkboxes not to appear in the templates
10:58 paul quite a complex one to re-play.
10:59 paul ;-)
10:59 kados ok ... does that mean you'd like to fix it if it's a bug?
10:59 kados s/like to//;
10:59 paul if nobody else want it yes :-(
10:59 kados I'm actually pretty familiar with that code
11:00 kados so I can take it on (leave other stuff for you to take ) ;-)
11:00 kados right
11:00 kados 892: data in a closed order appear differently than data before c
11:01 paul i'll take care of it.
11:01 kados ok
11:01 kados 585: Using 'document.forms[0]' notation prevents use of other for
11:01 kados owen: could you take 585?
11:02 paul not a fixable bug, but some coding rules.
11:02 slef owen reported it ;)
11:02 kados 753: Apostrophes and commas cause marc_word search to fail
11:02 paul that are not respected (i'm a dummy in html, I agree)
11:02 kados 753 is fixed right?
11:02 paul I think
11:02 kados I know it is
11:03 slef ideally, javascript needs to be removed, but it's not a 2.2 option, is it?
11:03 paul you change it fixed ?
11:03 slef Is non-javascript browser support in the 2.4 roadmap?
11:03 kados yep
11:03 kados (to paul)
11:03 owen I don't think removing javascript is an option period.
11:04 slef If you want a fat client, that shouldn't be the web browser.
11:04 kados slef: non-javascript support is definitely important for 2.4
11:04 kados imo
11:04 kados but I also think use of javascript is a 'good thing'
11:04 kados for browsers that support it
11:05 slef as helpers, perhaps
11:05 kados even for major functions as long as there are other ways to perform the functions
11:05 kados but that's a topic for another discussion
11:05 owen I agree
11:06 kados owen can you handle 585 (either close it if there's nothing to fix and alert Stephen to include that recommendation in coding guideelines)?
11:06 owen Yes
11:06 kados I'll skip 950 ... paul still working on that I asume
11:06 kados 717: Quotes in MARC subfield cause blank line in addbiblio.tmpl
11:07 kados anyone know if this is still a problem?
11:07 paul about 950 : i'm waiting for owen answer
11:07 owen Uh oh... :)
11:08 paul 717 : dunno
11:08 kados I'll assign it to me and I'll find out
11:08 owen One thing about 717
11:08 kados I'll skip 992 (I'm still working on it)
11:08 kados owen: ok
11:08 owen We've got a problem with #'s
11:09 kados ok ... can you add it to the description?
11:09 owen Sure.
11:09 kados 834: Add Category Fields Need Descriptions
11:09 kados sorry ... we've done that one right?
11:10 paul already done 1 hour later
11:10 kados 844: renewstatus subroutine should check for reserves
11:10 kados paul: do you like the fix?
11:10 kados paul: or should it be customizable?
11:11 paul it's ok for me
11:11 kados ok ... I'll skip 975
11:11 kados 782: Admin interface must not require javascript
11:12 kados owen: can you handle 782?
11:12 owen that's part of our larger javascript question, I think
11:12 owen I'm still on the fence, despite my comments there
11:13 kados ok ... so I'll put that on the list then for discussion on koha-devel
11:13 kados 868: OPAC MARC view doesn't appear in IE6
11:14 paul i'm working on this
11:14 kados ok ...
11:14 kados 879: bugs on OPAC
11:14 paul (unless owen want to fix it...)
11:14 kados I'm assuming operator error here
11:14 kados and I'm closing it out
11:14 owen Paul, I can take a look
11:14 slef kados: close as worksforme?
11:14 paul ok, thanks. I'm not so good in css...
11:14 kados slef: yep
11:15 kados 452: Payment due flags not showing up for children
11:15 kados owen: so I'll assign 868 to you
11:16 kados owen: ahh ... already done
11:16 kados right
11:17 kados ok ... let's start that then
11:17 kados slef have a proposal?
11:17 slef Is this discussing bugs.koha.org?
11:17 kados yep
11:17 slef 452 looks like an enhancement, by the way
11:18 kados slef: can you mark it thus
11:18 slef paul: how long are you there for?
11:18 paul i have to leave in 5-10 mn
11:19 slef Basically, bugs.koha.org showed us some of its weak spots today.
11:19 kados because
11:19 slef There are also security updates out for its version of bugzilla.
11:19 kados 1. timeouts (or general processing slowness)
11:19 kados 2. limits for sorting
11:19 slef and a whole later minor version is out too (2.16.x) if I understand it correctly
11:20 kados ok ... so should we bug katipo to upgrade?
11:20 kados (it's probably just running whatever woody's got)
11:20 slef There's a couple of other things which irritate with bugzilla: registrations to submit bugs; inability to send email commands
11:20 kados (so updating would require pinning, etc.)
11:20 slef kados: depends if they're planning to move to sarge soon
11:21 kados has anyone played with the sourceforge bug reporting system?
11:21 paul sarge is out ?
11:21 paul :-D
11:21 slef paul: since 8 days
11:21 paul was a joke. Even martians know this ;-)
11:21 slef kados: yes. Normal sourceforge problems
11:21 kados right ...
11:21 paul like slowness...
11:21 kados just thought I'd throw it in the mix
11:22 slef kados: it works and probably is less of a single-failure-point than bugs.koha.org right now, but on the whole, I'd prefer to stay at katipo
11:22 kados right
11:22 slef do we ask katipo to upgrade bugzilla, or do we look to move to a competitor for 2.3/2.4 and just use bugzilla for 2.2?
11:23 slef is bugzilla mostly doing what you want and it's just me?
11:23 kados hmmm ...
11:23 kados well since this is pretty much the largest dev project I'm involved with
11:23 slef (My attempt to get routine buglists out of bugzilla didn't work, as you can probably tell by the lack of them ;-) )
11:23 paul i'll have to leave now.
11:23 kados I'm not really clear on the options
11:23 paul unless there's a critical bug to see immediatly
11:24 kados paul: ok ... thanks!
11:24 slef paul: thanks
11:24 kados paul: nope ... we'll get the rest next time
11:24 kados paul: :-)
11:24 paul you're welcome. Read you soon
11:24 kados slef: so what are our options for bug reporting systems these days?
11:24 kados I seem to recall that the one bug I reported for Fedora was running on a nice reporting system
11:25 kados I guess they've got bugzilla too
11:25 slef kados: bugzilla, sourceforge, http://www.mantisbt.org/ (but that's python), http://www.bestpractice.com/rt/ http://www.agstools.com/bugtrack/
11:25 slef Did you have to register to report the bug in Fedora?
11:26 kados yep
11:26 kados that bestpractice link isn't working for me
11:27 slef I think that puts people off. I don't really understand why bugzilla does that, apart from maybe spamming.
11:27 slef sorry
11:27 slef http://www.bestpractical.com/rt/
11:27 slef was typing from memory
11:27 kados yep
11:27 kados shoot .. phone brb
11:29 slef well, owen isn't returning anyway...
11:30 slef ...I guess I'll move this to koha-devel.
11:30 kados sounds good
11:30 kados thx

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