You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-user@lucene.apache.org by karl wettin <ka...@gmail.com> on 2007/03/28 15:26:40 UTC

Unison index handling (was: Reverse search)

28 mar 2007 kl. 13.22 skrev mark harwood:

> Odd. I'm sure it used to have a getReader method somewhere.
>
> Still, you can use MemoryIndex.createSearcher().getIndexReader()

I've wrapped MemoryIndex in the unison index facade of LUCENE-550,  
just as I did with all the other index implemementations I could find  
(RAMDirectory, FSDirectory, MemoryIndex and InstantiatedIndex). It's  
very simple: an interface containing factory methods for creating  
IndexReader and IndexWriter so that there is no change to how one  
communicate with the index, no matter what implementation.

Of course it is completly backwards compatible. I think it would make  
a lot of sense to push it down to the trunk. It needs to be reviewed  
though. I could cut that out and place it in a new patch if that  
would help.

And it makes it really easy to decorate it with layers upon layers  
with bonus features. I did that too. They are in the same patch.  
Notifications, auto freshed readers and searchers, search cache, yada  
yada.

If you ask me, the static methods used when creating readers, opening  
an index, et c. is pretty darn bad design that makes life really hard  
unless you implement a formula 1-A solution. "But that's just like  
your opinion, man."


-- 
karl

---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org