Hi Karen,
I considered that, but using a complex type
would make specifying batch-fetch more complex, and the size would be rarely
used. I will think about it again.
I don't think there is any harm adding it to
the deprecated elements, the mappings they represent support this option, and
they inherit the annotation/xml processing, and it would allow JPA 1.0 users
access.
-----Original
Message-----
From: Karen Butzke
Sent: Monday, April 19, 2010 4:00
PM
To: Dev mailing list for Eclipse
Persistence Services
Subject: Re: [eclipselink-dev] SVN
main commit: Bug#211700 - batch fetching, also 220400 , 309242
James,
Regarding bug 309242
I think a better way to add the batch-fetch-size support to the eclipselink orm
schema would be to add a batch-fetch complex type that includes type and size
attributes. This would align with the BatchFetch annotation and prevents
the batch-fetch-type and batch-fetch-size elements from being duplicated in
multiple mappings in the schema.
Also, what were your thoughts on not adding functionality to the
basic-collection and basic-map elements in 2.1 since those are now deprecated?
Thanks,
Karen
On 4/19/2010 1:05 PM, James Sutherland wrote:
SVN main commit: Bug#211700 - batch fetching, also 220400 ,
309242
https://bugs.eclipse.org/bugs/show_bug.cgi?id=211700
Code
review: Andrei (pending)
Changes:
-
Added batch-fetch-size to orm schema and metadata processing.
-
Added _expression_ support for nested IN Lists for composite key comparisons.
-
Added batch fetching support for IN with composite primary keys.
-
Changed batch fetching query to first check the cache if possible, to avoid
queries for cached objects.
-
Added ManyToOneMapping to complement the JPA ManyToOne annotation and avoid
distinct in batch fetch for OneToOne.
-
Added support for WLS getVendorConnectionSafe to WLS 10 platform, to avoid
connection being discarded.
-
Added JPA test for overriding sequence generators.
-
Added performance comparison test for batch fetching types.
-
Added JPA metadata support for overriding sequence generators.
_______________________________________________
eclipselink-dev mailing list
eclipselink-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/eclipselink-dev