[Koha] Matching rules on record import
Cab Vinton
bibliwho at gmail.com
Fri Feb 20 04:54:35 NZDT 2009
>> It would be nice if during the import process Koha could check a
>> sequence of matchpoints when checking for duplicate holdings (e.g.,
>> first check ISBN, then LCCN, etc.) This is currently not possible,
>> however, right?
I'm informed by Joshua & Galen that one can check a sequence of match
points simply by adding additional match points to a rule set. (Thank
you!)
So, I've created a new matching rule based on the ISBN rule, adding a
match point for the LCCN. I based the match point on the ISBN match
point, changing the tag to 010 & entering LCCN as the normalization
rule.
My expectation was that this new rule would check the ISBN first for
matches, & then the LCCN. A test case failed, however.
I created a duplicate record of an item in our system & then deleted
the 020 ISBN tags before attempting to stage the record.
Unfortunately, the system did not pick up on the 010 LCCN match.
I thought this might because, at least in our system, the LCCN numbers
all seem to be preceded by 3 empty spaces. So I changed the Offset for
the LCCN match point to 3. But this didn't work either.
Does anyone see anything wrong with following setup for this new rule?:
1. Matching rule code: ISBNLCCN
2. Description: 020_010
3. Match threshold: 1000
Matchpoint 1
1. Search index: isbn
2. Score: 1000
1. Tag: 020
2. Subfields: a
3. Offset: 0
4. Length: 0
5. Normalization rule: ISBN
Matchpoint 2
1. Search index: lccn
2. Score: 1000
1. Tag: 010
2. Subfields: a
3. Offset: 0 [or 3]
4. Length: 0
5. Normalization rule: LCCN
[No match checks]
Perhaps the problem is coming from the normalization rule. Not sure if
this field is mandatory, how many normalization rules there are, etc.
It would probably too easy to simply enter "OCLC" as a normalization
rule when trying to match on the OCLC number ...
Many thanks,
Cab Vinton, Director
Sanbornton Public Library
Sanbornton, NH
More information about the Koha
mailing list