If you are building a query using the API, the WildcardQuery class will allow you to use a leading wildcard character. The QueryParser will not allow this, however, so if you're getting queries using the QueryParser a leading wildcard won't work.
I have successfully done substring queries through the API using code previously posted to the list: http://www.mail-archive.com/[EMAIL PROTECTED]/msg06388.html I haven't run into any performance problems because of these classes. There were a few minor changes that needed to be made to that code to make it work with the latest Lucene 1.4RC3 - I think it was just a matter of changing a constructor signature. Ben -----Original Message----- From: Terence Lai [mailto:[EMAIL PROTECTED] Sent: Tuesday, June 29, 2004 4:29 PM To: [EMAIL PROTECTED] Subject: Building query to match a sub-string of a field Hi Everyone, I am trying to construct a query which matches a sub-string of a field. As an illustration, I would like to search the following words by using the sub-string "test": - test - testing - contest - contestable I realize that Lucene does support wildcard searches using "*" and "?" in the custom query. Therefore, the query string "*test*" should give me the right result. However, the Lucene query syntax (http://jakarta.apache.org/lucene/docs/queryparsersyntax.html) does not allow the wildcard "*" as the first character of the search. Therefore, the query "*test*" is invalid. Does anyone have a solution to build the query to achieve the same result? Thanks, Terence --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
