|
|
|
Created an attachment (id=334)
Sorry, last patch was wrong version (missing a line). Use this one instead. I am reluctant to add this to the code base without abstracting it more because
it is Oracle specific. This would definitely take some time and is not a high priority right now. Also, tests need to be written. Martin, how do you feel about the patch as provided?
The patch is quite old and doesn't apply cleanly this days. I will update and
try it when I find some time. However it is really a very special thing, and only needed to fix the Oracle Thin JDBC driver usage. I don't yet know if it's also needed for the Oracle 10 JDBC driver. Well, we will see... Thanks, Martin. Btw, as stated by Bruce, simply applying this patch is not going
to be an option. As there's Oracle-specific code, we'll have to find a way of abstracting some of that functionality (via strategies, decorators or similar artefacts). Let's first check whether the code is usable at all, and solves the problem, and then we talk about refactoring things. Created an attachment (id=751)
updated patch for current code This is the result of my tests:
So I doubt it's worth the trouble to integrate this patch. It would only help Thanks, Martin, for the write-up. Can I please hear votes ? Iow, a) close this
down (incl. adding a FAQ entry) or b) apply the patch. So far, we have a) +2. |
Fix for Castor with Oracle JDBC THIN & inserting >4k LOBs (Heavily tested with both THIN & OCI, and BLOB & CLOB)