Problem to search 'valide' word on custom property

[ribz33]

Dear community,

When i search "valide" word on custom property LuceneQueryParser use LowerCaseFilter as TokenStream so Lucene transform my word "valide" as "valid" so i have more results than expected…

I there a way to configure Lucene to avoid this ? Is there a way to force VerbatimTokenFilter ?
If i define <tokenised>false</tokenised> in my model for my custom property, alfresco will be able to search again on this property and do you think that this can solve my problem ?

Thanks in advance.

[andy]

Hi

Your issue is related to stemming. The tokeniser can be changed to alter this behaviour. However, you may want this exact behaviour in other searches. The StandardAnalyser does not do any stemming except for simple English plurals and indication of possession.

If you wish to treat the whole text as a single identifier then your suggestion is correct.

Unfortunately, the analyser can only be specified at the type level and not per property.

Andy

[ribz33]

Thanks for answer.

you say that analyser can be specified on type.
Analyser for a type is configured in xml file ?  If yes, which xml file ?

Thx in advance.

[rsweeten]

I think Andy is referring to the <deployed_web_app_home>/WEB-INF/classes/alfresco/model/dataTypeAnalyzers.properties file.