Details
Bug
Open
Minor
Description
Wildcard searches in VuFind are currently case sensitive. That is, if you do a search for "test*" you get results, but if you do a search for "Test*", you do not get results.
This is because Lucene bypasses the filter factories when applying wildcards, so wildcard searches don't get run through the LowerCaseFilterFactory in order to become consistent with index values.
For more details, see:
http://wiki.apache.org/lucene-java/LuceneFAQ#Are_Wildcard.2C_Prefix.2C_and_Fuzzy_queries_case_sensitive.3F
The solution for this problem would be to have VuFind perform the lowercasing of wildcarded terms prior to calling Solr. Doing this without interfering with uppercase boolean operators or causing problems in index fields that do not use the LowerCaseFilterFactory could get complicated.
This is because Lucene bypasses the filter factories when applying wildcards, so wildcard searches don't get run through the LowerCaseFilterFactory in order to become consistent with index values.
For more details, see:
http://wiki.apache.org/lucene-java/LuceneFAQ#Are_Wildcard.2C_Prefix.2C_and_Fuzzy_queries_case_sensitive.3F
The solution for this problem would be to have VuFind perform the lowercasing of wildcarded terms prior to calling Solr. Doing this without interfering with uppercase boolean operators or causing problems in index fields that do not use the LowerCaseFilterFactory could get complicated.