Time  Nick  Message
21:54 kados chris: this might be something: http://indexdata.dk/zebra/doc/protocol-support.tkl
21:54 kados chris: The supported query type are 1 and 101. All operators are currently supported with the restriction that only proximity units of type "word" are supported for the proximity operator.
21:55 kados when I search I'm getting :
21:55 kados 20:54:06-19/03 zebrasrv(1) [request] Sort  ERROR 207 (1)->1
21:55 kados according to this: http://www.loc.gov/z3950/agency/defns/bib1diag.html
21:55 kados 201 Proximity not supported with this attribute combination attribute list
21:56 kados so maybe our problem is that we've got types other than word specified?
21:56 kados anyway ... something to consider, not sure exactly how we should modify our data model
05:44 russ  paul are you there?
05:44 paul  yep.
05:44 paul  hello
05:44 russ  hi how are you?
05:45 russ  has a price for the discover koha day and the koha user group day been decided?
05:45 russ  i have 15 euro for discover (3rd)
05:45 paul  not definetly, but it will probably be free at the end.
05:45 russ  and 50 euro for the 4th
05:45 russ  ah ok
05:45 paul  (for some legal reasons we can't charge)
05:46 russ  ah right
05:46 russ  ok
05:46 paul  1- ENSMP can't host a meeting if it's not free.
05:46 russ  i'll take that off the form then :-
05:46 russ  ;-)
05:46 paul  2- we don't have a NPO structure yet, so who could take the funds ?
05:46 paul  but that's not definetly sure, still waiting for ineo position.
05:46 russ  ok it can always go back later i guess
05:47 paul  right
06:03 paul  hdl ?
06:03 hdl   yes
06:03 paul  did you create dns entries o1.bureau.paulpoulain.com ...
06:03 paul  ?
06:03 paul  is your bloophone OK now ?
06:04 hdl   DNS entries : Not yet. Should create it today
06:04 paul  ok
06:04 hdl   bloophone : waiting for nerim response
06:04 hdl   still out of order.
06:05 hdl   Should be a firewall problem.
06:05 hdl   But it is in DMZ.
06:05 paul  open everything, you should see quickly if there is a problem there ;-)
10:16 kados paul: you around?
10:16 paul  yep
10:16 paul  hi joshua
10:16 kados hi :-)
10:16 kados I just did a small test
10:16 kados changed zebradb entry in my config file to point to external server
10:16 kados indexdata.dk:210/gils
10:17 kados http://opactest.liblime.com/cgi-bin/koha/opac-search.pl
10:17 kados searching works ... but, the results don't display!
10:17 kados I suspect this is because koha does not use the results from zebra to display title/author, etc.
10:17 kados is that correct?
10:17 paul  this DB is not an marcxml isn't it ?
10:18 kados dunno
10:18 paul  opac-detail.pl don't use marc datas, so don't use zebra, you're right.
10:19 paul  (i'm almost sure gils is NOT a marcxml DB)
10:19 paul  MARCopac-detail.pl could work, but gives an internal server error
10:22 kados ok
10:23 kados if we ever want to have a federated search we will need to start using results from zebra (MARC, XML, Dublin Core, etc.) for opac-detail and opac-searchresults.pl
10:23 kados opac-search.pl I mean
10:25 kados not important for rel_2_2 though
10:25 kados (except for testing sorting of external server)
10:25 paul  kados : right.
10:26 paul  what we will need soon will be something like a XSL stylesheet to transform any source into a "koha-xml" ;-)
10:26 paul  the question being "what means "soon" " ?
10:26 paul  I think it won't be 3.0
10:34 kados you may be right
11:06 paul  (pb with nerim ?)
11:06 paul  (I have some today... 2 disconnects a day, very rare problem)
11:17 kados thd: got your email, thanks
11:17 kados thd: "The link to edit existing items seems to link to a blank form."
11:17 kados thd: it seems to work OK for me
11:17 thd   your welcome kados
11:18 paul  (rel_2_2)
11:19 thd   kados: I tried the link link for editing an item from the items editor after submitting an item.  I did not test otherwise just now.
11:19 kados paul: w00t
11:20 kados thd: 942 only has one field?
11:20 kados thd: itemtype?
11:21 kados thd: just pmesged you a link to a catalog with the new framework
11:21 kados thd: it seems only 942 is showing up...
11:21 kados sorry ... only 942$c
11:23 kados thd: you have visibility set to '9', I think it should be '0'
11:24 kados I'm wrong, it's set to -5
11:24 kados no idea why it's not showing up
11:26 kados ahh ... -5 is !Editor
11:26 kados I'm pretty sure we want visibility to be '0' in this case
11:26 kados thd: when you're off the phone could you confirm that?
11:31 kados thd: 942 $c should be mandatory, if an item doesn't have an item type in Koha it can't be issued for instance ... we need to ensure that it's filled with a value
11:33 kados thd: why have you marked the record-level (biblioitem-level) classifications' as obsolete?
11:34 kados thd: I believe these are still used in searching and for display
11:38 thd   kados: well the obsolescence is intended to conform with our discussion of changing the templates to use items.itemcallnumber instead of whatever they use now
11:39 kados I see
11:39 kados that will probably work fine for the detail view
11:39 kados but it won't work for the initial search results
11:39 kados item fields aren't available to opac-searchresults.tmpl
11:40 thd   kados: It will work if the search template is changed would it not?
11:42 thd   kados: Or is the problem that the column is not passed to the template for use and so some Perl would need to change to pass the column?
11:42 kados thd: correct, the column is not passed to the template
11:42 kados thd: even with some perl that kind of behaviour would not be desired for rel_2_2
11:43 thd   kados: Is that not a simple change of adding the column?
11:43 kados thd: as there are more than one item per biblioitem and you could only select one for the displa
11:43 kados thd: no
11:43 kados thd: it's quite complicated because currently opac-search doesn't look at the items table at all
11:43 thd   oh
11:43 kados thd: we can't do it
11:44 kados thd: so that will need adjusting in the framework ... sorry :/
11:46 thd   kados: Then the value of biblioitems.dewey which is the only one needed should be set to 050 $a if the user is using LCC or 082 $a if the user is using DDC or alternately some script should fill the correct value in 942
11:49 thd   kados: actually the script could divide DDC numbers at the decimal point for filling 942 while LCC is too complicated for division.  LCC could be separated at the first space after the letters and before the numbers.