Diacritics and accents omeka solr

[hagud]

Good morning

We have one omeka 2.5.1 with solr 4.10 and search with accents or diacritics gives differents results, please may you advice how to fix it?, please

[kloor]

Is your issue that the SolrSearch wont find items with diacritics/accents if your search terms do not include those diacritics/accents?

[hagud]

Hi

Yes… We wish to find same results if we loof for historia and història for example

kindest Regards

[hagud]

Hi

sorry for asking, do you have any idea on how to search diacritiscs/accents even if they are not in the term… like historia and història, please?

[kloor]

You’ll probably need to alter the conf/schema.xml file in the Solr core for your Omeka installation. In that XML file, you’ll find a element starting with <fieldType name="text_en". That configures most of the text fields indexed for Omeka.

Inside the element are two <analyzer> elements, and you would probably want to update both by adding the following element inside of them:
<charFilter class="solr.MappingCharFilterFactory" mapping="mapping-ISOLatin1Accent.txt"/>

The one of type “index” will add an index that converts accents to ASCII, so if you search without accents the field will still match. The one of type “query” will remove accents from your search so it matches fields that do not have accents.

[hagud]

It works perfecte! thanks!