Thats not new in this release, there were perf improvements committed by
mikemccandless before, too.

-----
Uwe Schindler
H.-H.-Meier-Allee 63, D-28213 Bremen
http://www.thetaphi.de
eMail: [email protected]

> -----Original Message-----
> From: Chris Hostetter [mailto:[email protected]]
> Sent: Friday, June 11, 2010 6:34 PM
> To: [email protected]
> Subject: RE: [VOTE] Apache Lucene Java 2.9.3 and 3.0.2 artifacts to be
> released
> 
> 
> FWIW: I have not had a chance to review the RC, but....
> 
> : Simon and me decided, that we want to have LUCENE-2494 (Modify
> : ParallelMultiSearcher to use a CompletionService instead of slowly
polling
> : for results) inside Lucene Java 3.0.2. This is an significant
performance
> : improvement. It will not go to 2.9.3, as its Java 1.5 only.
> 
> ...this doesn't make senes to me.  even if this is a rock solid patch, if
it's a
> performance improvement then presumably it's not a bug fix -- in which
case
> it seems like it should wait for a feature release.
> 
> maybe i'm just missunederstanding the nature of hte issue -- but the goal
of
> bug fix releases should be to fix known bugs that prevent existing client
code
> from functioning, while minimizing the risk of introducing new bugs in
> existing code that already functions -- making performance iprovements to
> code that is currently functional risks introducing bugs.
> 
> 
> 
> -Hoss


Reply via email to