I am surprised that DB2 does not allow nulls and yet forces you to specify "not null" since it is totally redundant: PK cannot be null on DB2 by definition yet it forces user to explicitly specify "not null"
:/ The patch looks fine to me in this case. - Steve Ebersole Project Lead http://hibernate.org st...@hibernate.org Principal Software Engineer JBoss, a division of Red Hat http://jboss.com http://redhat.com steve.ebers...@jboss.com steve.ebers...@redhat.com On Fri, 2008-12-12 at 08:47 -0500, Juraci Costa wrote: > Guys, > > As suggested in a previous topic, I'll pay a genuine Czech beer in any pub in > Brno for whoever reviews the patch in HHH-3650[1] (or Moravian wine, if you > prefer) :-) > > The class org.hibernate.id.enhanced.TableGenerator is used to generate tables > which acts as sequences. In the method sqlCreateStrings(), it builds a > create-table statement, but the primary key column specified by the property > "segmentColumnName" is not being marked as "not-null", causing DB2 (and > probably others) to throw an exception. So, the patch adds a "not null" in > the statement, but I'm wondering if I should get the "not null" string from > somewhere else. I suspected that Dialect would have it, but only have for > "nullable" columns. I believe that "not null" is quite standard, and doesn't > requires the abstraction, but maybe you have another idea :-) > > - Juca. > > [1] http://opensource.atlassian.com/projects/hibernate/browse/HHH-3650 > _______________________________________________ > hibernate-dev mailing list > hibernate-dev@lists.jboss.org > https://lists.jboss.org/mailman/listinfo/hibernate-dev _______________________________________________ hibernate-dev mailing list hibernate-dev@lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev
