The limit for IN is 1000, which is relatively low. I don’t
know the exact overall statement limit with/without binding, but it’s
likely high enough, because we never reached it even if we have read over 5000
objects by primary key batch by ORing the many IN. Of course, the framework
should be cleaver enough to call the DB more than one time, instead of choking
on the size a single huge statement.
From:
eclipselink-dev-bounces@xxxxxxxxxxx
[mailto:eclipselink-dev-bounces@xxxxxxxxxxx] On Behalf Of Goerler,
Adrian
Sent: Friday, February 05, 2010 2:56 AM
To: Dev mailing list for Eclipse Persistence Services
Subject: AW: [eclipselink-dev] Updated r2.1 performance design doc
Hi Sebastien,
-
I assume it will handle the Oracle limit of number of
values in IN clause by breaking it down to many IN separated by OR.
You will likely hit a limit of the number of parameters then, I
am afraid.
-Adrian
Von:
eclipselink-dev-bounces@xxxxxxxxxxx [mailto:eclipselink-dev-bounces@xxxxxxxxxxx]
Im Auftrag von Sebastien Tardif
Gesendet: Donnerstag, 4. Februar 2010 22:29
An: Dev mailing list for Eclipse Persistence Services
Betreff: RE: [eclipselink-dev] Updated r2.1 performance design doc
Batch read using IN would be great. I assume it will handle the
Oracle limit of number of values in IN clause by breaking it down to many IN
separated by OR.
We have coded this functionality ourselves because it is so
handy. We check first if we can get the object in the cache, then retrieve the
missing ones in batch.
From:
eclipselink-dev-bounces@xxxxxxxxxxx
[mailto:eclipselink-dev-bounces@xxxxxxxxxxx] On Behalf Of James
Sutherland
Sent: Thursday, February 04, 2010 3:49 PM
To: Eclipselink-Dev@Eclipse. Org (E-mail)
Subject: [eclipselink-dev] Updated r2.1 performance design doc
Updated
r2.1 performance design doc
http://wiki.eclipse.org/EclipseLink/DesignDocs/298985#Batching_reading_using_exist_and_IN