Oracle DBA to the rescue ;-)... your query is wrong... read the following: http://asktom.oracle.com/pls/ask/f?p=4950:8:::::F4950_P8_DISPLAYID:553222846752
Now for your original problem... If you only need to get 1 ResultSet back I would change your PLSQL to a function instead of a procedure (returning your output1 ResultSet). The rest of the source code can then almost remain the same... and if full support is built in in iBATIS that kind of processing should still work (with minor changes in your xml maps). If you currently need to map multiple ResultSets from 1 stored procedure your only options are to not use iBATIS or use Michael Fagans patch on the WIKI (which may become obsolete or break sometime in the future). Regards, Sven >----- Oorspronkelijk bericht ----- >Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] >Verzonden: vrijdag, februari 17, 2006 12:38 AM >Aan: [EMAIL PROTECTED] >CC: ibatis-dev@incubator.apache.org, user-java@ibatis.apache.org >Onderwerp: Re: Reg. Problem with Cursors still open. > >Dear Sven, > >Attaching you the query for getting the cursor count. > >select count(1) from v$open_cursor where user_name = 'DBTXUSER'; > >regards, >Clement Justin > > > > >Sven Boden <[EMAIL PROTECTED]> >02/16/2006 06:17 PM >Please respond to list123 > > To: [EMAIL PROTECTED] > cc: ibatis-dev@incubator.apache.org, >user-java@ibatis.apache.org > Subject: Re: Reg. Problem with Cursors still open. > > > >Available options... but none that really fits you... >1) the way you use it now is not officially supported (IBATIS-53 returns >a ResultSet but does not map it to a ResultMap), you don't really use >output1, it just gets mapped to the returning ResultSet... and I'm still >waiting for your Oracle query to count open cursors ;-) >2) Michael Fagan made a patch to iBATIS which you can find in the iBATIS >WIKI (under use of Oracle cursors), but it's not in the baseline of >iBATIS. And if you use that you may end up in a dead track in the end >(unless you patch yourself every new version of iBATIS). > >There are some ideas on supporting what you want, but none of it's >implemented yet (discussed last week on the dev mailing list). And will >depend on whether we find a nice clean solution to it all and when we >find the time to build it in.... this will not be short short term (at >least not by me). > >Regards, >Sven >
