Update EJBQL query generating bad SQL ------------------------------------- Key: HHH-2724 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-2724 Project: Hibernate3 Issue Type: Bug Components: query-hql Affects Versions: 3.2.4.sp1 Environment: Hibernate version: 3.2.4SP1 with 3.3.1GA EntityManager and 3.3.0GA Annotations, running on Postgresql 8.1 Reporter: Jonathan O'Connor Attachments: test.jar The generated SQL (show_sql=true): update Person set name=? where (id in (select customer1_.id from Customer customer1_ ) ) and person0_.name<>? My EJBQL query is: update Person set name = ?1 where id in (select c.id from Customer c) and name <> ?2 The generated SQL refers to an unknown table alias "person0_". A workaround is to change around the conditions in the where clause. That is: update Person set name = ?1 where name <> ?2 and id in (select c.id from Customer c) does not generate bad code. If I add a where clause to the inner select, then the problem also goes away. Finally, correct SQL is generated in the matching select. Doing a delete instead of update also causes bad code, and the workarounds are also the same: delete Person where id in (select id from Customer) and name <> 'Y' The attached jar file has two classes Person and Customer. There is also a persistence.xml in the META-INF directory. ts configured for a local postgres system. You may need to alter this file. I tested the SQL generation using Hibernate Tools plugin for Eclipse. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira