[EMAIL PROTECTED] wrote:ehatcher 2003/11/27 18:03:14
Modified: src/java/org/apache/lucene/analysis StopFilter.java
. CHANGES.txt
Added: src/test/org/apache/lucene/analysis TestStopAnalyzer.java
Log:
Use position increments to account for removed stop words
I finally remembered the problem with this. Consider the phrase query "Lord of the Rings". What does QueryParser do with this if "of" and "the" are stop words? I think it turns it into "Lord Rings". But with this patch, that phrase query will no longer match text containing the text "Lord of the Rings". Or have I missed something?
*ugh* - you're right, of course. I'll revert it. Just to clarify for others, that means we'll be back to QueryParsing "Lord Rings" matching "Lord of the Rings" that was indexed.
So either this patch should be pulled, or we need to add position-increment-like support to PhraseQuery. I plan to do the latter in the next few months (for a contract I'm working on) so perhaps we should just pull this patch until PhraseQuery is updated, at which time we can consider updating QueryParser to take advantage of this feature.
Sounds good to me. I can't wait to see the new and improved PhraseQuery!
Erik
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
