IRC log for #koha, 2006-12-07

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

All times shown according to UTC.

Time Nick Message
11:22 owen kados: if you get time to look at the new z3950 search with me, give me a holler.
11:24 paul hello owen & good morning
11:24 owen Hi paul
11:24 What's new?
11:25 hdl hello
11:25 paul i'm adding some nice new features.
11:25 toins has ended the code cleaning.
11:25 the one i'm working on this afternoon is a simple one :
11:26 the library will be able to define what is shown in the result list, depending on itemtype of the doc.
11:26 for exemple, for a book <title>- <author>...
11:26 and for a website, only the href link
11:28 owen that sounds great
12:50 Anarka i sent a email do win32 ml asking on instruction on how to make z39.50 work on windows with 2.2.7 but still havent got any answer... should i use another ML for that question ?
12:50 paul Arnaka : you could/should try koha ML, even if it's win32 related. I'm not sure that the core developpers are reading win32 ml
12:51 (i'm not reading it)
12:51 Anarka eheh okis :)
12:51 owen Is someone actively working on the win32 version these days?
12:52 or is it just a matter of win32 people trying their best to shoehorn Koha onto windows?
12:55 Anarka it does work on windows, and you do have packages for it
12:55 the problem lies on the z39.50
12:56 the changelog of 2.2.7 makes it seem so simple but...
14:29 pecisk hi guys, quick question, does Koha has something for analytics?
14:30 it is meant for describing articles from journals, books, etc.
14:33 paul pecisk: I would answer yes, as you can define any framework for cataloguing.
14:33 pecisk I see
14:40 _paul (in which zebra config file I mean)
14:44 kados _paul: default.idx
14:44 _paul mmm... I don't have default.idx in my zebradb scope...
14:45 (hello kados)
14:45 http://www.indexdata.dk/zebra/[…]-and-charsets.tkl
14:45 as usual, indexdata doc is quite unclear to me
14:46 how do I explain that I want to use /home/paul/default.idx ?
14:46 kados profilePath
14:46 in your .cfg file
14:46 _paul kados rocks, as usual
14:50 paul it works, kados, you're a genius !!!
14:50 mmm... maybe not a genius, but at least a wonderful coder ;-)
14:50 a quick note before leaving : I think toins is gone with code cleaning.
14:51 he did a lot of cleaning on Biblio.pm
14:51 some things to clean in Reserves & circ probably, but we won't touch it yet.
14:51 now, he will work on OPAC, to synch librarian features on OPAC & check that everything is OK
14:52 dec 8th => SAN-OP stop cataloguing anything in their previous ILS
14:52 dec 14th => library closed, everybody works on Koha, for heavy tests.
14:53 dec 21th => library closed again, everybody doing last tests
14:53 dec 25-> dec 30 => library closed for holiday
14:53 jan 2th => still closed, except maybe for developpers
14:53 jan 3th => koha goes live on SAN-OP
14:53 I keep finger crossed...
14:53 kados there is a problem you should know about
14:54 two in fact
14:54 at NPL I don't currently have zebraop in production
14:54 the records are updated in a batch process
14:55 and zebraop is commented out
14:55 there are several reasons:
14:55 1. it's very slow to add a new record with zebraop
14:55 2. it can slow down searches too
14:55 3. we have seem cases where the index crashed (hasn't happened in a while and ID found and fixed some bugs so it may be OK now)
14:56 paul 1. => did you install the last release of zebra ?
14:56 kados 4. there is a problem with the way that the zebra connection is created -- in batch procedures it will fail
14:56 paul because it seems to have fixed a slowness in add iirc
14:56 kados yes, and I haven't tested yet
14:57 but we do run the latest version at NPL
14:57 of the 1.x series
14:57 2.0 needs more testing IMO
14:58 I would recomment trying bulkmarcimport to see an example of the problem with the connection object
14:58 paul ok, i'll try to give it a try asap.
14:58 does it mean zoom is bugguy ?
14:58 or zebra ?
14:58 dewey well, zebra is #1 for me
14:59 paul you spoke of 2 problems ? what is the 2nd one ?
14:59 kados the connection object was the second
14:59 sorry :-)
15:00 paul OK. do you think it's a ZOOM / zebra / Perl problem ?
15:00 kados I'm not sure if it's zoom that's buggy, koha's method of creating the object (in Context.pm), or Zebra that is dropping it
15:01 paul are you using mod_perl ?
15:01 kados no
15:01 paul shame. because I would have suggested that the drop is in fact a timeout.
15:01 but without mod_perl, it can't be.
15:02 (as the process is a fresh one on every request)
15:02 kados it's on the command line so I think it's unrelated to mod_perl
15:03 paul it's strange that update_items works & koha don't...
15:03 kados bulkmarcimport you mean?
15:03 paul no, update_items.
15:04 (or i missed something)
15:04 kados bulkmarcimport doesn't work for me
15:04 update_items does
15:04 paul if I understand : updating zebra though koha don't work, but through update_items works.
15:04 kados updating through koha works sometimes, sometimes it doesn't
15:04 paul mmm... OK, understood : update_items don't use ZOOM to reindex
15:04 kados right
15:05 paul ok, I have to leave.
15:05 my wife is waiting for me for dinner, very impatiently
15:05 have a good day.
15:05 kados hehe
15:05 you too
15:05 paul_away (& see you tomorrow maybe)
16:23 mtnbkr Hello
16:23 dewey hello, mtnbkr
16:26 mtnbkr I am having a heck of a time installing Koha for a client.  The server is Gentoo linux (everything veryrecent)... The koha installer script tells me that GD::BarCode is not installed when it is, and there are  a couple red lines printed just as it starts checking perl modules:   http://rafb.net/paste/results/5MDnYo73.html   can anyone offer some assistance?
17:32 cm hi kados.  Just found your mail about bug 1215.  I checked to make sure we're still having the problem.
17:32 kados hey cm
17:32 cm hey.  :)
17:33 you're right; it probably is implementation.  ;)
17:34 kados working now?
17:35 cm nope.  All I see is a blank 'issues' when a book is checked out, & no previous issues section at all.
17:35 this is with an up-to-date npl template.
17:37 kados NPL's all committed so I don't get it
17:37 (there was one uncommitted change to Circ2.pm, but it's unreleated)
17:37 so let me just make sure I understand
17:38 cm ok
17:38 kados you have items with issues from, say, yesterday
17:38 sorry ... borrowers
17:38 with issues from yesterday or some time in the past
17:39 and the issues just show up in issues, not in previous issues
17:39 and today's issues show up in issues too, eh?
17:39 the code that splits them up is pretty simple to understand
17:39 it's in circ/circulation.pl
17:39 around line 200 or so
17:39 line 216 actually
17:40   my $issueslist = getissues($borrower);
17:40    # split in 2 arrays for today & previous
17:40    foreach my $it (keys %$issueslist) {
17:40        my $issuedate = $issueslist->{$it}->{'timestamp'};
17:40        $issuedate =s/-//g;
17:40 etc
17:42 cm yeah.  it's actually weirder than that now.  when you bring up the issuing screen for a patron,
17:43 and check out a book, it issues it, but it doesn't show up in the list of issues onscreen,
17:43 and it doesn't immediately update the number of books out either.
17:44 if you refresh, it asks you to confirm the issue because it's already out to that patron, and if you confirm it,
17:45 the number of items out is updated but nothing ever shows up under the Issues section, and the Previous Issues
17:45 section never appears at all.
17:45 I'll try it in something other than firefox to make sure it's not the browser.
17:48 yep, same thing in mozilla.  Unfortunately I don't have Opera installed right now for a completely different perspective.  :/
17:50 kados I think throwing some warns in circulation.pl is the way to troubleshoot this
17:50 lets take a look at what that timestamp looks like
17:50 my $issuedate = $issueslist->{$it}->{'timestamp'};
17:50 owen kados, haven't we had problems with items not displaying when an itemtype was missing or something like that?
17:50 kados add:
17:50 warn $issuedate;
17:50 after that
17:51 owen: ahh, good one ... yes
17:55 cm the warn would be in the log, right?  if so, then it said this:
17:56 [Tue Dec 05 15:47:07 2006] [error] [client 192.168.1.210] HERE : 8 / (99 for 863 at /usr/local/koha/intranet/mod​ules/C4/Circulation/Circ2.pm line 655., refer
17:56 er: http://circ.ccfls.org/cgi-bin/[…]borrower=25102119
17:56 [Tue Dec 05 15:53:43 2006] [error] [client 192.168.1.210] HERE : 9 / (99 for 863 at /usr/local/koha/intranet/mod​ules/C4/Circulation/Circ2.pm line 655., refer
17:56 er: http://circ.ccfls.org/cgi-bin/[…]borrower=25102119
17:56 kados hmmm
17:56 I don't see the warn
17:56 try this instead:
17:57 warn "ISSUEDATE".$issuedate;
17:59 cm um, same sort of output.
18:00 owen, do you mean the old itemtype or the ccode?  If it's the ccode, I don't have those yet, which may explain it.
18:03 kados cm: are you sure the error log is the right one?
18:03 cm: and you're issuing books ?
18:04 cm koha-error_log?  yeah, books.
18:04 kados and you're sure it's the right circulation.pl that you edited?
18:04 because you should see a ISSUEDATE2006blahblah in the log
18:07 cm yeah, pretty sure.  I put the warn on line 220 right after  my $issuedate = $issueslist->{$it}->{'timestamp'};
18:07 kados weird
18:09 cm indeed.
06:11 btoumi hi all
06:12 paul kados: sleeping or around ?
10:14 kados hi paul
10:14 paul hello kados & good morning
10:14 hdl hello kados.
10:15 kados hi hdl
10:15 toins hello kados
10:15 kados paul: did you have luck reproducing the problem with zebraop using bulkmarcimport?
10:15 hi toins :-)
10:19 paul no, I could not try it this morning.
10:19 but I hope working on this this afternoon.
10:20 kados : i wrote 2 mails on koha-devel, waiting for an answer from you (just in case you did not read koha-devel when waking up ;-) )
10:45 kados paul: responded to both
10:47 1. searching still works
10:47 2. the index can be completely rebuilt while the system is running, and quite quickly
10:48 paul last question : how do you know that index has crashed ?
10:48 kados because zebraop fails
10:48 paul so, you have something in koha logs, right ?
10:48 kados errors in koha logs, and on zebra logs
10:49 paul ok, thanks. we will hunt them next week.
10:49 kados yes, and some error like ES immediate execution failed
10:49 (maybe other errors too)
10:49 paul i'll drop a mail to SAN to warn them.
10:49 kados paul: now I have a question for you ;-)
10:49 paul: it seems that acquisitions is still broken ...
10:50 at least in dev_week with npl templates (which are fully in synch with rel_2_2)
10:50 do you have any libraries using acquisitions running 2.2.7?
10:50 paul yep
10:50 kados (ie, to prove they are working?)
10:50 strange
10:50 so maybe it's the templates
10:50 I'll investigate today
10:51 paul what kind of bug ?
10:51 kados when receiving an order
10:51 there are several problems
10:52 difficult to describe
10:52 koha.smfpl.org is a good place to observe it
10:52 there is one supplier, amazon
10:54 so suppose I order 4 items
10:54 when I receive a parcel, it doesn't handle the count properly
10:55 also, after saving the item, the page reloads and says 0 bibliographic items received
10:56 paul mmm... check for a tmplate problem, because nobody reported anything about this for me.
10:56 kados ok, will do

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

koha1