If you need a new place in the Solr index to store your data, you can use dynamic fields (http://vufind.org/jira/browse/VUFIND-480?focusedCommentId=12510&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-12510).  Something like local_classification_str might work.


To actually extract these values, a custom BeanShell script might be in order – something that iterates over the 852 values and splits the strings at the appropriate places.  You might be able to use the existing import/index_scripts/dewey.bsh for inspiration.  More on BeanShell here:




Let me know if you have more questions.


- Demian


From: Christian Dabrowski [mailto:dabrowski.christian@fh-swf.de]
Sent: Thursday, February 21, 2013 9:35 AM
To: vufind-general@lists.sourceforge.net; vufind-tech
Subject: [VuFind-Tech] modify data during import


hi all,

for the browsing module I would like to use our classifications (due to fusions we have two). Both classifications has no special field in our ILS, but they're used in the callnumber:

first classification: Bta8 107:2
second classification: QSD1150(2)

The callnumber is in the MARC-field 852, subfield 9. The identification of the classification data is easy: I have to use the data (1) up to the first blank or (2) to the first number when there were three uppercase charakters before. My question is, where can I modify the script in order to create a new field during the import, for example 852x, and put there the classification data (Bta8 oder QSD)? If it is possible, could someone give me a hint where I have to make the changes (vufind 1.4)?