The only real functionality that Lucene lacks that is supplied by other search engines is the wildcard prefix.
The most efficient way to do this is to, when indexing, add text to two fields, one the normal field, and another using an analyzer that reverses the text of each token. Then use prefix searches of the reversed field to efficiently implement prefix wildcards. This doubles the size of the index and slows indexing, but does the trick.
Doug
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
