Hello Kohaers, I'm trying to set up Koha for our meager library here at Deep Springs, 25k volumes or so, and we're currently going through the process of building a digital register (essentially replacing an abandoned collection of poorly kept records)...which has been timely, but with lots of help we're more than half way through. However, this register was built on some assumptions that I'm hoping hoping hoping are true. 1) Is there is a scriptable way (even if i have to write it myself) of getting full MARC records into a database from a tabulated list of ISBNs, LCNs, Titles, Authors, Publishers, or some combination there of? This would essentially involving querying the Library of Congress database directly somehow. Through a perl module? hmm. There is an old MARC database lying around on a CD that has ~15k of the records, but we don't know which ones...heh heh heh. This place suffers from a horribly short institutional memory. But enough about us. So while this first step is a pretty big step, I am also very interested in getting this second step working indifferent of step 1 since we _do_ have the ~15k records in MARC. 2) Doing a Bulk import of MARC into Koha. This I assume you all know a lot about, and I just need to be pointed in the right direction. I'm a fairly proficient perler, so I should be able to figure it out...I've read about something called bulkmarcimport? is this in full operation? The first issue is one I would greatly appreciate help with, though it is something I need to figure out pre-Koha, and is not necessarily the forte of this list. Of course, any and all help anyone can provide would be much appreciated. Thank you, Johan Ugander Deep Springs College
Hello Johann, I think I can help you out with at least the first part of your question. You can get MARC records from the LoC by using the Z39.50 protocol. I have had success using the YAZ client ( http://www.indexdata.dk/yaz/ ) and Perl's Net::Z3950 on a linux machine. I have attached a simple script which uses Net::Z3950 to read ISBNs from a text file, look them up on the LoC, and then write the MARC records to a file. It works pretty well, but I threw it together pretty quickly. You can also use Perl's MARC::Record to do any processing you need to on the records. For your second question, I am not sure on the status of bulkmarcimport.pl, but in theory it should do what you want it to do. I know there are people who are more knowledgable with this than I. Anyone want to help me out? Mike
Mike Hansen wrote:
For your second question, I am not sure on the status of bulkmarcimport.pl, but in theory it should do what you want it to do. I know there are people who are more knowledgable with this than I. Anyone want to help me out?
For the 1.9.1/2.0 versions, bulkmarcimport is COMPLETLY out of date. It must be rewritten. If you want to begin this, take a look at C4::Biblio.pm It contains many interesting features. The NEWnewbiblio and NEWnewitem subs should help if you want to add MARC data and already have correct MARC parameter tables. imho, the most complex part of a bulkmarcimport is to separate biblio and items : in a MARC record, there can be 1 biblio and X items. You MUST split them and call 1 x NEWnewbiblio and X x NEWnewitem. Hope this helps. -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org)
participants (3)
-
Johan Ugander -
Mike Hansen -
paul POULAIN