Re: [castor-dev] Castor JDO & CALL SQL

2002-01-30 Thread Brad Clow
castor can't cope with brackets in a CALL SQL. we have hacked our castor source to cope with it based on a message about this issue in the archives. brad -Original Message- From: Sylvie RAMON [mailto:[EMAIL PROTECTED]] Sent: Wednesday, 30 January 2002 9:57 PM To: [EMAIL PROTECTED] Subje

Re: [castor-dev] java.util.ConcurrentModificationException when change tomcat3.2 to tomcat4.0

2002-01-03 Thread Brad Clow
we have had the same problem with a combination of castor and poolman (we don't use tomcat, but i can c that u r using poolman from the stack trace).   i am not sure if it is a poolman or a castor problem yet, haven't had time to investigate it thoroughly.  to get around it, u can comment o

Re: [castor-dev] inserting objects with many-to-many-relation [RE-POST]

2001-12-05 Thread Brad Clow
i would have expected your mapping of the skills field in the Worker class to be like: i would also expect the following process in your java code (u might be already doing this): begin transaction load worker load skill (the existing one to add to this worker) add loaded skill to worke

Re: [castor-dev] JDO class design constrained by RDBMS relational paradigm

2001-12-02 Thread Brad Clow
nal Message- From: Bruce Snyder [mailto:[EMAIL PROTECTED]] Sent: Friday, November 30, 2001 12:17 AM To: [EMAIL PROTECTED] Subject: Re: [castor-dev] JDO class design constrained by RDBMS relational paradigm This one time, at band camp, Brad Clow said: BC >u don't have to have both si

Re: [castor-dev] JDO class design constrained by RDBMS relational paradigm

2001-11-29 Thread Brad Clow
u don't have to have both sides of a relationship represented in your java objects. for example we have a many-to-many defined in the database. however only one of the two java classes involved contains a collection of the other. we also have one-to-many relationships where we have done the oppo

[castor-dev] generated aliases of tablenames including a schema

2001-11-27 Thread Brad Clow
;."USER_PERMISSIONS"."JOB_GROUP_ID" = 5 and "TEST"."USER_ACCOUNT_0"."CODE" = "fred") Changing the mapping file so that table names are not prefixed by the schema is an adequate workaround (for us) at the moment. This occurs with today's CVS snapshot. brad ___ brad clow chief technical officer workingmouse email: [EMAIL PROTECTED] web: <http://www.workingmouse.com> <>

Re: [castor-dev] Oracle 8i - Poolman

2001-11-25 Thread Brad Clow
we started with the thin driver (there are no blobs involved), but found that castor would intermittently pause for 20-50 seconds.   i poked around in the castor source to make sure it was cleaning up JDBC objects properly - which it could do better.  i inserted some very rough hacks to cle

[castor-dev] Selecting dependant objects

2001-08-15 Thread Brad Clow
s OQL: select p from myapp.Publisher p i get: org.exolab.castor.jdo.QueryException: Could not find mapping for class myapp.Publisher i can access the publishers from their respective organisations, so my mapping.xml appears ok. much thanx brad ___ brad clow

[castor-dev] Lazy Loading

2001-08-15 Thread Brad Clow
much thanx brad _______ brad clow chief technical officer workingmouse email: [EMAIL PROTECTED] web: http://www.workingmouse.com <>