I just upgraded our 2.1 branch and trunk to use OpenJPA 1.2.0 instead of the older 1.0.2 level. Can you grab a 2.1 snapshot build tomorrow (20080904 or later) and see if that solves your problem?

http://people.apache.org/builds/geronimo/server/binaries/2.1/


-Donald


ericp56 wrote:
Hello,

I have a large table with a varchar(16) PK field.  Using Geronimo JPA, the
query being generated by JPA is something like:
declare @P0  nvarchar(4000)
select @P0 = '1234567890'
SELECT t0.field1 /*THE PK FIELD which is varchar(16)*/, t0.field2 FROM dbo.history t0 WHERE t0.field1 = @P0

Instead of using the index (seek) , the conversion of varchar(16) to
nvarchar(4000) causes a full index scan, which is painfully slow (2 seconds,
should be in milliseconds).

I tried using the @Column annotation, but that didn't change anything - @P0
is still nvarchar(4000)
I tried using a named query, but it only allows basic SQL.
I tried a native query, but the casting of the result  fails.

Any suggestions?

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to