Where do I find example data for MARC records to import into Koha?

I'm working in a online library and I need put some example Marc records in my Koha to make a demo. I downloaded some MARC records (.mrc, .mrk) but when I import them, there are no matching fields and therefore no matching searchs.

How can I determine what type of MARC record I'm handling? (I dont know if they are Marc21, Marc8, ...)

If I must set the rigth record fields before importing the MARC files (as there's a set of default record fields, like Books, DVD, ...), where can I download a spreadsheet of the right MARC record fields to import it? (Because a MARC record has many many fields, and I don't think that putting them in one by one was right)

The MARC records that I used can be found here and here.

I have Zebra running, and I followed the installation steps one by one, I'm stuck in this part and the Koha documentation doesn't say much.

user1403997

• ebooks
• marc
• koha

Comments

Answer by Jared

I'm not entirely clear what the problem is, but the section in the manual on importing records is at http://manual.koha-community.org/3.8/en/catalogtools.html#stagemarc

The example records you link to are all MARC21, since the United States uses MARC21. As for the encoding, I don't know, but I'd try UTF-8 first, and if that doesn't work (you get weird glitches on letters with diacritics), then try MARC-8. It is the responsibility of the importer to figure out what kind of data they are using, since the various standards don't provide any way to decisively identify format and encoding (supposedly encoding is indicated in the record, but that is accurate only 50% of the time, which means you are better off just choosing one, trying it, and switching if you don't like the result).

Koha handles MARC records natively so you don't need any sort of spreadsheet with mappings. You just need to follow the instructions linked above.

One further note: the UMich dataset is staggeringly large. You almost certainly do not want to import the entire thing. If you really want all 684k records, you'll need to chop it up because it is too large for the web server (not Koha, but Apache) to handle. I understand you can use MarcEdit for tasks like that.

P.S. For the future, you may have more luck asking questions about Koha on the Koha IRC channel.

Comments

• dsalo: Yes, I was going to suggest that the OP wasn't seeing results because the import failed halfway through. I don't know what Koha does by way of logging, but turning up the log level might help.