Thanks a lot guys.
Uwe: thats why I was asking ;) I had no proof it was the TokenStream
API, that just seemed a likely candidate - I'm not familiar with that
filter, but it worked with a version of Lucene right before the
TokenStream improvements patch, and then started failing after.
When I traced what was going on, I saw:
In Set:false String:the
Set:[BIG, the, and, it]
Which is pretty weird - and why I asked here what it could be.
Thanks for looking into this, both of you !
Robert Muir wrote:
that makes perfect sense
On Thu, Aug 6, 2009 at 11:31 AM, Uwe Schindler<u...@thetaphi.de> wrote:
I have seen ur mail, but this bug should not be related to the new Token
API, it should occur with old API, too.
Maybe the problem is an unrelated change:
https://issues.apache.org/jira/browse/LUCENE-1762
This issue changed the default length of the termBuffer in
Token/TermAttributeImpl. Because of the wrong calculation in the filter, the
larger default size could break this filter. When using termBufferLength
instead of termBuffer.length it should be fixed.
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-dev-h...@lucene.apache.org
--
- Mark
http://www.lucidimagination.com
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-dev-h...@lucene.apache.org
