IRC log for #koha, 2006-02-09

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

All times shown according to UTC.

Time Nick Message
11:16 kados |hdl|: could you go over the basic process for importing title/subject authority files?
11:17 |hdl|: can I bulkimport a MARC authority file into Koha?
11:17 |hdl| basic process is a bulkauthimport into the authority needed.
11:20 kados thanks
11:26 |hdl| file sent.
12:04 thd kados: bulkauthorityimport.pl is not even close to functional.  bulkauthorityimport.pl was initiated for an experiment that was abandoned.
12:09 s/bulkauthorityimport/bulkauthimport/
12:21 root Help: I am sure this is somewhere and I've probably looked right at it.  My Debian Sarge Install seems good BUT z39.5 just does not seem to show up anywhere.  It is started in the background but how can I test it?  Search?  Maybe I am wrong in thinking that it should help me in adding books etc?
12:33 |hdl| thd: the bulkauthimport I sent kados I used for an import.
12:34 root: have you launched z3950daemon
12:35 to achieve this : export KOHA_CONF=/path/to/your/koha.conf and PERL5LIB=/path/to/module
12:35 Verify you have a z3950-daemon-options file.
12:36 root Yes I did launch it via /etc/rc2.d/S99z39 which has this /usr/local/koha/intranet/scripts/z​3950daemon/z3950-daemon-launch.sh and a ps aux shows it is running as this www-data 16659  0.0  3.2 13232 8388 tty2     S    10:00   0:01 /usr/bin/perl /usr/local/koha/intranet/script​s/z3950daemon/processz3950queue /usr/local/koha/log
12:36 |hdl| in /path/to/koha/intranet/scripts/z3960
12:36 ok.
12:36 Have you set z3950 servers data in Koha ?
12:37 (see parameters module)
12:37 root Two things one where is parameters module and two
12:38 i thought it came preinstalled with congress link but since nothing worked I added two other places and marked them 1 to make sure they are searched
12:38 |hdl| launch koha
12:38 root alright
12:38 |hdl| go into intranet
12:38 get to parameters page
12:38 root alright
12:39 |hdl| get to z3950servers page
12:39 root alright
12:39 |hdl| what do you see
12:40 root I added this Library of Congress z3950.loc.gov 7090 voyager checker=1 rank=1
12:40 syntax marc21
12:41 |hdl| ok
12:42 relaunch the z3950daemon
12:42 root what is the best way to stop it and relaunch it is not /etc/init.d/???
12:42 |hdl| and then try z3950 search on biblio you DONOT have
12:42 root: should be
12:43 kados morning owen
12:43 |hdl| hi owen
12:44 root also it seems I had a button somewhere for z3950 search but now???
12:46 |hdl| root : bibliocreation
12:46 or edition
12:47 root okay it sais title or isbn now if I enter a title word say garden that is not in the local database it sais bibios in koha none
12:48 bibliois in reservoir none
12:48 duh
12:48 then there is the z395 search though it still sais nothing found
12:49 Am I wrong that this should search the Congress database for books with this word?
12:49 |hdl| you will probably have too many answers.
12:49 try with ISBN will be better.
12:52 root alright one moment I will try it
12:55 thd |hdl|:  Is the bulkauthimport.pl that you sent kados in CVS?
12:56 |hdl| Nope : It was a specific script for a client.
12:57 thd |hdl|:  Will it be in CVS any time soon?
12:58 root |hdl|so here is what i did...I went to the congress site and searched the isdn and it worked so then I entered that same isdn in the first page of Catalog-->Add Biblio -->ISBN
12:58 then after that
12:59 I went and clicked wait something just happened for the better
12:59 I changed the setting from Marc21 to unimarc and not the search brought it up???
12:59 |hdl| thd: could be.
13:00 root hdl I will write more later but it looks like it is working thanks for pushing me along
13:00 |hdl| root: didnt understand you latest sentences.
13:01 root well in Parameters --> Z359 --> I edited the Syntax from Marc to Unimarc (from the pull down menu) and now it worked
13:03 though I thought that the Koha install had this setup from default....I got this in a local Koha doc
13:05 thd |hdl|: If it is not otherwise in CVS, I would be very happy if you would send a copy to koha@agogme.com
13:11 |hdl| kados: the LOC Cataloging and Distribution Service, primary source of MARC file distribution in the US, just lost 40% of their staff.
13:11 |hdl| thd: sent
13:12 root : ok
13:12 thd: a shame
13:15 thd |hdl|: Thank you for the file.  Apparently, the money for the national Library in the US is now going to some other cause..
13:16 |hdl| Could it be Catrina's fault ?
13:16 or Irak War ?
13:25 thd |hdl|: I am not certain of the cause and actually the LC budget may not have been reduced.  However, MARC file distribution has definitely been downgraded within LC.  I am not certain what the consequence is except for a much smaller staff to administer the distribution of MARC files.  This just happened last month as part of unreplaced employee retirement.
15:55 |hdl|: Are you still there?
18:34 russ kados are you around?
18:35 kados russ: yep
18:53 russ: ?
18:58 thd kados: If russ is not present, could you explain about whether hdl sent you a file for building authority records from bibliographic records in Koha or for actually importing pre-existing authority records from an external source as might be done with a working bulkauthimport.pl?
18:59 kados hehe
18:59 he did in fact send me a file
18:59 but I have not had a chance to peek at it today
19:00 I spent the better part of the afternoon crafting a lovely little command-line report that checks for unused yet registered methods in the open-ils CVS repository :-)
19:01 thd kados: I think he may have confused building with importing unless he sent me something different from you by mistake ;)
19:03 kados: The Open-ILS methods could be used to interface between Koha and an Open-ILS system?
19:03 kados thd: no, they are just perl methods within open-ils
19:03 thd: I suppose some day we could nab some stuff from them
19:03 but noone has time right now
19:04 thd: here, this should keep you busy :-):
19:04 for x in `grep -r "api_name.*=>.*" ILS/Open-ILS/src/perlmods/OpenILS/Application |awk '{print $4}'| sed "s/\"\|,\|)\|\;\|'//g"`; do echo " Method: $x "; grep "$x" ILS/Open-ILS/xul/staff_client/ch​rome/content/main/constants.js| wc -l| tr -d '\n'; echo -n " : "; grep "$x" ILS/Open-ILS/web/opac/common/js/config.js |wc -l| tr -d '\n'; echo -n " : "; grep -r "$x" ILS/Open-ILS/xul/staff_client/server/admin/ |wc -l|tr -d '\n'; done
19:05 thd kados: As if I were also not busy enough already still catching up from my long absence.
19:06 kados heh
19:07 thd kados: you use sed.  Perl is too powerful and consistent in implementation for you? :)
19:09 kados: or does Open-ILS use awk and sed much in preference to Perl or something else?
19:12 kados thd: no ... but I've been trying to use them whenever I have opportunity
19:16 thd kados: I can never remember what needs what type of quoting or escaping across languages.  However, sed has not been very consistent over time and between different implementations.  Yet, it still requires much less resources than a complete language such as Perl.
20:06 kados chris: here's the query overdue.pl is running:
20:06 select date_due,concat(firstname,' ',surname) as borrower, borrowers.phone, borrowers.emailaddress,issues.itemnumber, biblio.title, biblio.author from issues, borrowers,items,biblioitems, biblio where isnull(returndate)  && date_due<'2006-02-07'  && issues.borrowernumber=borrowers.borrowernumber && issues.itemnumber=items.itemnumber && biblioitems.biblioitemnumber=items.itemnumber && biblio.biblionumber=items.biblionumber  && (borrowers.firstname like '010-40203%' or
20:07 I recon returndate isn't null
20:10 chris hmm
20:10 what do you reckon it is then?
20:11 kados no idea
20:11 chris returndate is null if the book hasnt been returned, if returndate has a value
20:11 then the book isnt on issue
20:11 kados right ...
20:11 I'm confused :-)
20:12 chris do you know the borrowernumber?
20:12 id try a select * from issues where borrowernumber=whatever
20:12 and just see what they look like
20:12 kados k
20:12 chris or even
20:13 id try a select * from issues where borrowernumber=whatever and returndate is null
20:13 kados select count(*) from above gives me 14
20:13 chris and thats what circ reckons?
20:13 kados which is right according to the member page
20:13 yep
20:13 chris right
20:13 but the overdue report says how many?
20:13 kados 12
20:14 chris right
20:14 do we know the 2 that arent showing?
20:14 the itemnumbers or barcodes?
20:14 kados yea
20:14 well, one of them
20:14 let me find the other
20:15 chris do a select * from items,biblioitems,biblio where
20:15 ah ha
20:15 i see the problem
20:15 look at the overdues query
20:15 see this line
20:15 biblioitems.biblioitemnumber=items.itemnumber
20:15 BZZT wrong
20:16 that should be
20:16 kados heh
20:16 chris biblioitems.biblioitemnumb​er=items.biblioitemnumber
20:17 kados that did it
20:17 chris cool
20:17 kados nice catch ... thanks
20:17 I'll commit it right away
20:17 chris sweet
20:20 did you see yesterday that someone was in here asking about public libraries and koha
20:20 kados no I missed it
20:20 chris ahh, it'll be in the log, they were quite interested to talk to someon from NPL
20:21 kados I'll have to follow up on that
20:21 chris i said their best bet was between 9-5 EST
20:21 to catch owen or stephen
20:21 kados right
23:33 root ISBN Import question: How possible is it for a library to scan all their book's ISBN to a text file and then pass that info to a script that would One match the isbn number to the wanted book info and two import that bibio info into koha?
23:34 russ the library doesn't have any digital records?
23:38 chris its possible, wouldnt be too useful without some item data like barcode as well, so that koha can make items to attach to the biblio
23:38 im assuming this is for a library that has no library system already?
23:47 root They have no digital record other then the company they pay alot of money too but now they want to go independent
23:47 so
23:48 I was thinking they could quickly scan to a text file and then I would run the import at the end of the effort.  Later
23:48 I just set them up for using Z395
23:50 thd root: what sort of digital record do they have already with the company to which they pay a lot of money?
00:12 root records that could be exported to marc at about $1000 price
00:13 chris gotta love vendors that charge you for your own data
00:13 the toss up will be
00:14 can someone write a script and scan all the isbns, and load them in plus the other info it needs, like shelf location, branch location etc
00:14 for less than $1000
00:14 thd root: how many records are there?
00:17 root they have  5000 books and stuff maybe more that is a good question
00:20 thd root: chris put the economics quite well.  It is certainly a shame that your vendor would charge you for your own records as many do.
00:24 root: If $1000 is the price for all of your over 5000 records at less than $ 0.20 per record I doubt that you could do the work necessary to replicate them for less unless you consider your labour costs to be already paid in advance with time to spare for new record capture and editing.
05:18 |hdl| thd: I saw your question : Indeed, it is an authority creation from existing biblio.
05:18 You are right : it doesnot import auhtority file.
05:19 But the problem with such an import would be divided into 2 parts :
05:20 1) Import the authority data (it is a mere bulmarcimport)
05:21 2) link biblio with authority : you can use authoritysearch and input the $9 from that.

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

koha1