IRC log for #koha, 2006-04-21

All times shown according to UTC.

Time S Nick Message
12:00 kados sec
12:01 kados so long as the templates in both head and npl have been updated to use the new syspref, yes
12:03 paul_away bye guys.
12:03 paul_away I'm away until next thursday.
12:04 paul_away i'll read my mails but won't be on this chanel (except maybe on the evening)
12:04 paul kados: could you tell me where is you hotel In Paris ?
12:04 paul i have to reserve one too
12:04 paul so, I could reserve the same as you
12:17 paul ok, kados is not here :-(
12:17 paul i'm leaving...
12:19 kados paul_away: back
12:19 paul_away hehe...
12:19 paul_away i'm lucky !
12:19 kados :-)
12:20 paul could you just remind me your hotel in Paris ?
12:20 kados the hotel in paris ...
12:20 kados unfortunately, only russ knows :-)
12:20 kados as he made my arrangements :-)
12:20 kados I will find out and email you tonight
12:20 paul great, thanks
12:20 kados np
12:37 thd kados: please check you email
12:38 thd s/you/your/
12:40 kados thd: got it, thanks!
13:47 kados wow, owen, you're really nailing those bugs
13:48 owen Just some simple stuff that never got taken care of, or which have since been taken care of.
13:48 kados that's great
13:49 kados I've got a very good feeling about 2.4
13:49 owen I thought the bug squash was very successful. I hope we get lots of follow-through
13:49 kados yea ... I'm planning on attacking my bugs later this week
13:50 owen We need to also go through and reassign stuff that's assigned to missing people like pate and mwhansen
13:50 kados yep
13:51 kados good point
13:52 tumer kados:One problem I'm having with MFX is that after a query the search term put on the result page is very weird. It looks in some other encoding than utf8 or iso.
13:52 tumer oops where did the smiley come from sorry
13:53 kados hmmm
13:53 kados could it be transformed twice by mistake?
13:54 kados my idea is just to transform it once, on its way into the database
13:54 tumer kados: I hav a utf8 database. Utf8 zebra, utf8 templates so I donno whats going on
13:54 kados from there, everything is done in utf8
13:54 kados so here is how i have been doing it
13:54 tumer kados yes. And the results are all correctly displayed but the search term
13:55 kados ahh
13:55 kados I bet that is the font
13:55 kados try changing the font in the template
13:55 tumer never thoght of that but does not do it with old version
13:55 kados old version?
13:55 kados you mean before MFX?
13:56 tumer old MFX
13:56 kados hmmm
13:56 tumer have you tried any saerch with high ascii?
13:57 kados you have MARC21 records?
13:57 tumer yes
13:57 kados and they are marc8 encoded before MFX?
13:57 kados (no, I've not tried with high ascii)
13:57 tumer well before they enter koha yes. In koha they are utf8
13:58 kados and how do you convert them to utf8?
13:58 kados during import?
13:59 tumer In the old method my char_decode does that very nicely
14:00 tumer I checked everything in db char by char. All looks good. export them out and dump tehm all looks good. As soon as I start with MFX all conversion is OK but this new weird problem
14:01 tumer Oh all but some chars dont get converted with MFX as I said before
14:02 tumer It still could be windows problem but i have to sort this out.
14:04 kados hmmm
14:04 owen So is a bug filed about the old Koha web page fixed because it was redesigned or invalid because the old site doesn't exist anymore?
14:05 kados hehe
14:05 kados I'd say 'wontfix'
14:05 kados tumer: we need to examine MARC::Charset to see if we can add the chars that aren't getting converted for you
14:05 kados tumer: i will have time to work on this later in the week
14:06 kados today I'm finishing up two migrations
14:06 kados and shipping some servers off
14:06 tumer kados:I'll look into that as well
14:07 tumer bye for now
16:08 owen chris: you around?
16:09 rach hes at tech meetinmg 4 another hour or so
16:12 kados they rach
16:12 kados hey even
16:12 rach hey
16:13 kados :-)
16:14 owen Mine never let me sit still long enough to do that
16:15 rach he's not super keen
18:08 kados hey chris
18:10 chris cool
21:05 Jo Russ: are you here.
21:06 Jo can you look up please and see if our website is loading. Had another complaint that our website has been down for the last 3 days.
21:06 chris works for me from home jo
21:06 Jo thanks Chris.
21:07 chris joshua are u about?
21:07 Jo I dunno whats happening - 2 completely seperate people complained
21:07 chris do you know what isp's they are on?
21:07 Jo no - but i shall ask next time!
21:07 chris cool
21:09 Jo Cjris: it wouldn't be loading a cached file for you would it?
21:09 chris i did a search just to test
21:09 chris worked ok
21:09 Jo thanks.
21:09 chris and i got walter to check
21:10 chris he is on a different isp to me
21:10 Jo ok.
21:10 Jo cheers.
21:10 Jo coincidence
21:14 chris if may have been really slow jo, if lots of ppl were looking, and using up the bandwidth
21:14 chris if=it
21:14 chris so they might have thought it was down
06:43 tumer thd:around?
06:46 thd tumer: yes I am here
06:46 tumer thd: thanks for the framework. I've commited a small script for interim use of remapping
06:47 tumer thd:Regarding LC sorting my problem is with zebra
06:48 tumer May be due to  windows platform sometimes it times out when there is too many updates in quee
06:48 thd tumer: your quite welcome.  What is the name of the script file you committed?
06:48 tumer thd:marc_rebuild_newframework
06:49 tumer in head /misc
06:51 tumer thd: It is ofcourse much better that we use only one field for LC sorting after all the padding. I'll commit some script that does this
06:52 thd tumer: I have no good information about Zebra.  I am still trying to fix things for 2.X and had a large gap of lost time at the end of last year.
06:53 tumer thd:I'm not enquiring about zebra. All I'm saying is due to problems with it I decided to use LC sorting on 2 fields, but I believe you weer saying of
06:53 tumer thd:combining them to one single field. Am I correct
06:53 thd tumer: I have at least tried to concentrate attention on some issues that would be carried over into 3.0.
06:54 thd tumer yes I suggested combining the two elements into a third subfield
06:55 tumer thd:are we have to have this 3rd field reflected in mysql db as well?
06:55 thd tumer: I did not understand why two would be faster but id you actually find that two was faster?
06:56 thd tumer: well you can always add something to the Koha DB tables.
06:57 tumer thd: 2 is not faster.Definitely slower. It is zebra finding a looong word timeconsuming to sort and giving out Time-Out errors
06:58 tumer thd:If I can sort that out than offcourse definitely sort on 1 field
06:59 thd tumer: The use of the existing mapped Koha DB columns is that chris had written code for Dewey sorting and searching.
07:00 thd tumer: searching by call number also expects Dewey numbers without padding.
07:01 thd tumer: padding may solve the sorting problem but then a call number search many not match the Koha DB columns for LC call numbers.
07:01 tumer thd:Getting confused now:(
07:02 tumer thd:I am talking about pure LC Call numbers Alpha.Decimal.Alphanumeric
07:05 thd tumer: If you search by call number in the OPAC in Koha 2.X it searches the call number columns specified in my bibliographic framework revision.
07:33 pierrick are misc/migration_tools/ and misc/migration_tools/ still relevant?
07:33 pierrick I would like to make zebra work on my HEAD copy
07:48 hdl pierrick: I havenot tested these last two weeks. But last time I did, it was.
07:49 pierrick which one? or
08:36 hdl pierrick: both of them. But rebuild_zebra_idx is faster. (Sorry Was on sthg else)
08:39 pierrick creates many warnings/errors. I've used and it only fails at the very end :-/ But, My HEAD copy seems to work better (I can add orders to my basket now).
08:40 pierrick At the end of, the error I had was "ZOOM error 224 "ES: immediate execution failed" (addinfo: "only XML update supported") from diag-set 'Bib-1'"
08:45 hdl pierrick: rebuild_zebra_idx requires a new directory called biblio created in your koha/zebra dir.
08:46 hdl pierrick: do you have one ?
08:47 pierrick yes I do
08:58 hdl pierrick: is it not filled when launching rebuild_zebra_idx ?
09:04 pierrick How could I know zebra is filled? Zebra is a blackbox for me
09:23 hdl pierrick: you can have a look at zebra dir and you should have mf files which are big
09:23 hdl hi paul
09:23 paul hello all
09:24 hdl To me, I had to tell zebra on which base I commited things.
09:24 hdl zebraidx -d databasename
09:25 hdl That commited changes to my base.
09:25 hdl If not so, I had problems.
09:31 pierrick hello paul
09:31 paul hello pierrick
09:32 pierrick hdl, I have *.mf files, but there are not very "big"
09:32 pierrick the biggest is 13KB
09:32 pierrick (with EMN dataset)
09:32 hdl too small. IMHO
09:36 kados morning owen
09:36 kados hi all
09:36 owen Hi
09:42 paul pierrick: my dataset with EMN is something like 100MB !
09:42 pierrick hi kados & owen
09:43 pierrick paul, yes that's the size of my MySQL database, but one day I'll make this ***** zebra work (maybe... I hope)
09:48 pierrick why are currencies manageable on acqui/ The management page isn't admin/
09:48 pierrick (I'm trying to simplify acquisition screens)
09:50 kados pierrick: I'm not sure that rebuild_zebra works
09:51 pierrick kados, :-)
09:51 pierrick kados, I'm feeling tired :-/
09:51 kados pierrick: yep, still lots to do before zebra's integrated
09:52 pierrick kados & tumer, do we have an estimated percentage of integration today?
09:52 kados pierrick: I'd say we're about 65% there ...
09:52 kados pierrick: just a rough estimate :-)
09:53 pierrick OK, for me it means everything that touch biblio is broken
09:53 pierrick once you know it, you avoid some screens :-)
09:53 kados well ... as I envision it, a good way to play with zebra right now is the perl-zoom plugin
09:53 kados for rel_2_2
09:53 kados as it's mostly stable
09:54 kados (stable as far as I know, it could use some testing)
09:54 pierrick kados, I've tried to use the plugin on 2.2 last friday and... couldn't make it work (I really thought it was not so hard to make work)
09:55 kados really?
09:55 kados it works fine for me ... what did you have trouble with?
09:55 pierrick yep, maybe I'm underskilled?
09:56 pierrick trouble with export first, two exports did not produced the same result :-/
09:56 pierrick I suppose I had wrong versions of MARC::* modules. I updated them today from sourceforge CVS
09:58 kados pierrick: the instructions on are almost a step-by-step
09:58 kados pierrick: I'd be surprised if they didn't work
09:59 pierrick at least they are not complete
09:59 pierrick I made you a remark last friday, don't remember which
10:00 pierrick yes, it was about the export script was bugged on rel_2_2
10:00 kados missing zebrasrv localhost:2100 I think
10:00 pierrick (I commited a fix)
10:01 kados ahh, yes the from HEAD is I think the right one
10:01 kados brb
10:01 paul I'm really not close to the soccer stadium. But have my window open.
10:01 pierrick so, you have to modify kohadocs
10:01 paul And we hear the OM supporters very well !
10:01 paul (semi-final of french soccer cup since 1mn)
10:02 paul "coupe de France"
10:02 paul If there is a goal, we should hear it !
10:03 pierrick I would like a clarification about Koha terminology on acquisitions
10:03 paul hmmm...
10:03 paul goal for OM !
10:03 paul (at 1st mn, we thought it was for the beginning of the match, but it was for the 1st goal !)
10:04 paul 17seconds after the beginning of the match
10:04 paul (for non-french ppl : OM = Olympique de Marseille)
10:04 pierrick When we say "add to order", it means "add to basket". In the database, an order is a basket line
10:04 paul pierrick: there are things more important than Koha !!!
10:04 paul :-D
10:04 pierrick paul, you like soccer?
10:05 paul yes. And you must know that it's IMPOSSIBLE to live in Marseille & not like soccer.
10:05 paul really !
10:05 pierrick (happy not to live in Marseille, I hate soccer)
10:05 pierrick (I hate stupid supporters in fact)
10:06 paul at least you're not a PSG supporter. that's better hating soccer than supporting PSG :-D )
10:06 pierrick (the one that destroy a gas station when their team loose a stupid match)
10:07 pierrick happy for this productive conversation about soccer, acquisitions?
10:08 paul kados &all, a good news :
10:08 paul Your project registration for Gna! has been approved.
10:08 paul Project Full Name:   OpenCataloger
10:08 paul Project System Name: opencataloger
10:08 paul Project page:        /projects/opencataloger
10:08 kados paul: great!
10:08 pierrick :-) Gna! is fast
10:08 kados very fast it seems
10:08 paul pierrick: + it has SVN support !
10:09 kados excellent
10:09 pierrick paul, of course I know :-)
10:09 pierrick (use it every day... night in fact)
10:10 paul ToinS is working not as fast as a long-time Perl programmer, but what he does seems nice to me. I hope I could commit something next week
10:10 paul pierrick: is there somewhere a nice tutorial in french for svn ?
10:10 paul (as i'm a real newbie)
10:11 pierrick the reference is the
10:11 pierrick personaly I own the french paper version
10:14 pierrick but SVN works very like CVS
10:15 pierrick svn co module; vi module/; svn status module; svn diff module; svn commit module/
10:15 pierrick what you can do more is : svn move; svn merge; svn copy (create branch)
10:45 paul hurlement au vélodrome...
10:45 paul 3-0 for Marseille !

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