Created attachment 93886 [details] Importing Auth Record Match options When you import authority records using Tools->Stage MARC records for import the options for checking duplicates are for ISBN , ISSN and KohaBiblio. However, those fields make no sense for checking auth duplications.
It should _at least_ look for $1xx fields, but DGI would _love_ to have $001 control number field checking
Hi Cori, the matching rules are fully configurable. The ones you listed below are just some samples that Koha adds out of the box as sample data during instalation. Have you tried adding your own? The only tricky bit is knowing the index name. I found this helpful: https://bywatersolutions.com/news/match-points-in-koha I am changing to enhancement.
Let me rephrase then, Koha should come out of the box with a sample matching record for 1xx fields for loading Authority Records.
(In reply to Cori Lynn Arnold from comment #3) > Let me rephrase then, Koha should come out of the box with a sample matching > record for 1xx fields for loading Authority Records. Yes, I think that would make a lot of sense. It's just a bit more difficult than the standard rules. I think we would want something like: Score: 50 points Then 3 matchpoints for each 100 = 50 points 110 = 50 points 111 = 50 points So that each of those would make it reach the score needed to match.
Created attachment 93941 [details] Example of a 1xx Rule Attached is an example of a working 1xx Matching Rule for Authorities
I am surprised that the rule attached is working, I would have expected it not to because the Search index information is empty - could it be that it's using the keyword index for matching in this case?
(In reply to Katrin Fischer from comment #6) > I am surprised that the rule attached is working, I would have expected it > not to because the Search index information is empty - could it be that it's > using the keyword index for matching in this case? It does work. Should I add an index? If so, which one do you think?
(In reply to Cori Lynn Arnold from comment #7) > (In reply to Katrin Fischer from comment #6) > > I am surprised that the rule attached is working, I would have expected it > > not to because the Search index information is empty - could it be that it's > > using the keyword index for matching in this case? > > It does work. Should I add an index? If so, which one do you think? Hi Cori, sorry it took me so long. The indexing of authority records is a little bit of a mystery to me. We just had someone struggling with this on the German mailing list as well. I believe indexes would be: 100 = Personal-name 110 = Corporate-name 111 = Meeting-name As listed here: http://git.koha-community.org/gitweb/?p=koha.git;a=blob;f=etc/zebradb/marc_defs/marc21/authorities/authority-koha-indexdefs.xml;h=39b46025f7b28e80c02081e9636be90dc764668e;hb=HEAD If you get a chance and could try if adding these make a difference in your rule, I'd appreciate hearing about the results.