CGLIB
by Steve Ebersole
I think the time has come to drop CGLIB support :( It is just becoming
near impossible to get them to respond to issues. We even recently sent
them 2 patches of major issues and still cannot get them to apply the
changes upstream and cut releases.
So that leaves us the choice of deprecating CGLIB support and eventually
dropping it altogether or forking it. I personally do not want to fork
it, mainly because I do not want to have to maintain it. Anyway lets
discuss it this week if need be and we'll vote at the meeting monday.
--
Steve Ebersole <steve(a)hibernate.org>
http://hibernate.org
14 years, 7 months
IRC meeting (5/10)
by Steve Ebersole
I did not save off the transcript of todays meeting. If anyone has it and wants to share feel free. The minutes as I remember were:
1) 3.5.2 a go for wednesday
2) Hudson rounding into shape
3) Emmanuel will initiate discussions about including spatial library and special geo query support
-- Sent from my Palm Pre
steve(a)hibernate.org
http://hibernate.org
14 years, 7 months
Composite ID suggestion
by Ω Alisson
Couldn't we simplify the process of using compound keys? Actually the
official JPA implementation sux a lot in my opinion(extra classes with
@Embeddable and @EmbeddedId on the parent side).
Something like:
@Entity
class Item {
@CompositeId
int itemId;
@CompositeId
int orderId;
}
14 years, 7 months
Move to H2 for tests
by Steve Ebersole
Here are the last in testsuite:
1)
testScrollingJoinFetchesEmptyResultSet(org.hibernate.test.hql.ScrollableCollectionFetchingTest)
This seems top be an error in the H2 JDBC driver. I have asked on their
forum. The issue is that on an empty result set, ResultSet.next is
returning true.
2)
testMappedAliasStrategy(org.hibernate.test.sql.hand.query.NativeSQLQueriesTest)
testFailOnNoAddEntityOrScalar(org.hibernate.test.sql.hand.query.NativeSQLQueriesTest)
Not really sure. What is really perplexing is that afaict this should
fail on any dialect.
3)
testCastInSelect(org.hibernate.test.hql.ASTParserLoadingTest)
testCustomColumnReadAndWrite(org.hibernate.test.compositeelement.CompositeElementTest)
Both seem to be caused by imprecise handling of double/float values. I
need to look at these more closely (unless someone gets to it before
me).
4)
testWithClauseFailsWithFetch(org.hibernate.test.hql.WithClauseTest)
testWithClause(org.hibernate.test.hql.WithClauseTest)
These both pass when I run them individually or when I run the class by
itself. Need more investigation.
5)
testInsert(org.hibernate.test.legacy.CustomSQLTest)
testCollectionCUD(org.hibernate.test.legacy.CustomSQLTest)
testCRUD(org.hibernate.test.legacy.CustomSQLTest)
Again need to look more, but I think these tests are just hosed. They
are supplying custom insert SQL but are specifying the native id
generator while the insert string is expecting a (what appears to be
VARCHAR) value.
Anyway, looks like we are getting close...
--
Steve Ebersole <steve(a)hibernate.org>
http://hibernate.org
14 years, 7 months
Hibernate Core (trunk and branch35) hudson jobs
by Strong Liu
Hi team,
now we have set up our hudson jobs for Hibernate Core (http://hudson.jboss.org/hudson/view/hibernate/),
from the URL above, you can see there are four jobs that relate to Hibernate Core.
1. hibernate-core-trunk
2. hibernate-core-testsuite
3. hibernate-core-branch35
4. hibernate-core-branch35-testsuite
for 1,3 jobs, they run everyday to fully build Core project and run testsuite on hsqldb (will move to H2 soon), and if no failure nor error, then it will deploy snapshot to our maven repo and it will trigger another downstream job (1 --> 2, 3 --> 4), the downstream job will run the testsuite on the DB matrix.
every job page is RSS enabled, 2 RSS feeds for each job, one is for all results, and the other is for failures only, feel free to subscribe them.
Okay, guys, time to make testsuite jobs blue!
and questions are always welcome :D
-------------------------
Best Regards,
Strong Liu <stliu at redhat.com>
14 years, 7 months
