In fact i'am able to persist my data using an array of bytes like that :

/**
* @ojb.field
*/
private byte[] primaryContent;

But my problem is that when i retrieve a lot of objects having this type of
attribute, i get an OutOfMemory error, and the blob , has i understand it
allow to
have a "reference" to the data, but looking at the ojb test suite i see
that the blob tests are done the way i do it and not using the Blob
interface,
am i wrong on that ? Do you think i have to declare my field to type Blob ?
And if so  is there  a way to construct a blob  to be  able to persit it ?

Jean-Yves


2006/1/25, Thomas Dudziak <[EMAIL PROTECTED]>:
>
> On 1/25/06, Jean-Yves Sironneau <[EMAIL PROTECTED]> wrote:
>
> > I have a question regarding the handling of large data with ojb on the
> > Postgres platform.
> >
> > - I think Blobs are not supported on this platform, is that true ?
> > - What is the best way to tell ojb to fetch the data from a field on
> each
> > access to the field ?
> > - Is that better to manage that issue "by hand"
> >
> > Right now i have an array of bytes as the field corresponding to the big
> > content, it's
> > working pertty well except that i get Out of memory errors because the
> data
> > stays on the heap.
>
> I think, the Postgres driver (at least the new 8 ones) have no
> problems handling BYTEA columns with the JDBC Blob interface, so you
> should be able to tell OJB its a Blob and use BYTEA in the database.
>
> Tom
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>

Reply via email to