FetchSpecificationBatchIterator

Hi,
	So I finally got around to writing the batch iterator that was  
discussed a while back on the lists. I opted for the simpler solution  
of first fetching all of the primary keys for the given qualifier of  
the fetch specification and then iterating through chunks of those  
primary keys fetching the actual eos with an IN qualifier (Thanks  
Anjo). This has the effect of on the request for the first batch  
causing two fetches to the database. All subsequent batch requests just  
require one fetch. I also allow the editing context used to be swapped  
out between batch requests. So a typical use case for iterating through  
1,000,000 eos would be to create a new editing context and set that on  
the iterator before calling the nextBatch method. This way all of the  
eos created within a batch could be garbage collected.
	The batch iterator works great except for one draw back. Currently if  
the batch size is set over around 300 then the JDBC query actually runs  
out of memory (something like more than 64 KB has been exceeded). I  
remember this wasn't the case in the 4.5 world, Oracle actually chokes  
for IN queries greater than 1,000 if I recall correctly. For my  
purposes batching groups of 250 eos is fine, but for other applications  
a larger batch size would be nice. Any suggestions?

Regards,
	Max

ps - Source for ERXFetchSpecificationBatchIterator is in ERExtensions  
also here:  
(http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/wonder/Wonder/Common/ 
Frameworks/ERExtensions/Sources/er/extensions/ 
ERXFetchSpecificationBatchIterator.java?rev=HEAD&content-type=text/ 
vnd.viewcvs-markup).



-------------------------------------------------------
This sf.net email is sponsored by: Influence the future 
of Java(TM) technology. Join the Java Community 
Process(SM) (JCP(SM)) program now. 
http://ads.sourceforge.net/cgi-bin/redirect.pl?sunm0004en