[Hibernate-JIRA] Created: (HHH-3818) Hibernate/JBC integration doesn't property handle Entity/CollectionRegionAccessStrategy evict
by Brian Stansberry (JIRA)
Hibernate/JBC integration doesn't property handle Entity/CollectionRegionAccessStrategy evict
---------------------------------------------------------------------------------------------
Key: HHH-3818
URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-3818
Project: Hibernate Core
Issue Type: Bug
Components: caching (L2)
Affects Versions: 3.3.1
Reporter: Brian Stansberry
Assignee: Brian Stansberry
Fix For: 3.3.x
EntityRegionAccessStrategy and CollectionRegionAccessStrategy have evict(Object key) and evictAll() that say they must cause removal of items from the cache "immediately without regard for transaction isolation." The Hibernate/JBC integration is not properly handling this as the JBC removeNode calls it makes are not dealing with transactional issues. The integration needs to:
1) Perhaps suspend any ongoing tx (particularly since these calls are typically made from a tx Synchronization's afterCompletion() callback, when the tx is Status.COMMITTED and it isn't appropriate to call into the cache expecting the cache to incorporate the call into the transaction. (Alhthough I believe JBC handles this correctly by ignoring the tx, since it isn't ACTIVE).
2) Deal with the fact that the tx that is being committed is likely holding locks in JBC. This is the big issue. I believe dealing with this will a) require keeping state in the Hib/JBC integration layer's Region to track that an eviction has occurred but may not be reflected in JBC b) using JBC as a notification bus to propagate the fact of the eviction to other nodes c) attempting to evict the data from JBC locally, failing promptly in the face of lock conflicts d) including in any get() or putFromLoad() calls logic to check the state from a), again attempting to evict locally (with a very short timeout) if not yet evicted and not allowing calls to proceed into JBC until successful.
--
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
11 years, 4 months
[Hibernate-JIRA] Created: (HHH-4857) Sharing some of the config info from Ejb3Configuration and SessionFactoryImpl
by Chris Baxter (JIRA)
Sharing some of the config info from Ejb3Configuration and SessionFactoryImpl
-----------------------------------------------------------------------------
Key: HHH-4857
URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-4857
Project: Hibernate Core
Issue Type: New Feature
Components: core, entity-manager
Reporter: Chris Baxter
Priority: Minor
In our system, we have each clients data in a separate DB instance. We're using JPA (via hibernate) on top of that to get at the data. The problem here is that for each client DB supported by one of our jboss servers, we have to have an instance of EntityManagerFactory compete with an Ejb3Configuration instance. Each config instance takes up around 20 MBs of permanent space, so if we have have 1000 client DBs supported by a jboss instance, it takes up a hugh amount of memory for information that will always be the same (the DB schemas are always the exact same in each client DB) across all EntityManagerFactory instances. So what I want to know is if there is a way to only have one instance of Ejb3Configuration that can be shared across all of the EntityManagerFactory instances that are spawning my EntityManagers. This would be huge as it would help me save many gigs of memory that never can be collected (old gen in jboss).
I prototyped some changes locally to Ejb3Configuration (to allow an instance of AnnotationConfiguration to be supplied in the constructor, allowing sharing across all instanced of this class) and to SessionFactoryImpl (to share some of the properties like entityPersisters, classMetadata and such) and I was able to reduce the footprint to 20 MBs for the first DB and then under a MB for each additional DB. The changes I made were a bit quick and dirty though, just to see if it was possible. So I'm taking out this feature request to see if this is a piece of work that you guys can do (or I can contribute after refactoring to a more suitable approach) and integrate into the main code for hibernate-core and hibernate-entitymanager. Let me know if this is something you guys think to be a valuable addition.
--
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
11 years, 4 months
[Hibernate-JIRA] Created: (HHH-3465) HiRDB Support
by Tomoto Shimizu Washio (JIRA)
HiRDB Support
-------------
Key: HHH-3465
URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-3465
Project: Hibernate3
Issue Type: New Feature
Components: core
Affects Versions: 3.2.5
Environment: 3.2.5 + HiRDB V8 + Type4 JDBC Driver
Reporter: Tomoto Shimizu Washio
Attachments: hirdbdialect.patch
I would like to contribute with a dialect for HiRDB, Hitachi's RDBMS.
(For more information, please see http://www.hitachi.co.jp/Prod/comp/soft1/global/prod/hirdb/)
I added 4 classes and slightly modified 2 existing classes as shown below. Please look at the documentation at the top of the patch for explanations.
org.hibernate.dialect.HiRDBDialect (added)
org.hibernate.dialect.UserFuncDeclParser (added)
org.hibernate.dialect.DialectFactory (modified)
org.hibernate.dialect.function.AnsiTrimEmulationFunction (modified)
org.hibernate.dialect.function.AnsiTrimEmulationFunctionWithTrimstrs (added)
org.hibernate.dialect.function.TypeQualifiedSQLFunction (added)
You may think this code is relatively large as a dialect. It is because I also implemented user-defined function support. As HiRDB requires '?' parameters to be qualified by 'as <type>' in user-defined function invocations, I made a parser for the type declarations that the user specified in the properties file, and a renderer for the type qualifier. See javadoc comments of HiRDBDialect and UserFuncDeclParser for more details.
This patch is created on Hibernate 3.2.5.ga and tested on HiRDB V8 with Type4 JDBC Driver. The example configuration for HiRDB is shown below:
hibernate.dialect org.hibernate.dialect.HiRDBDialect
hibernate.connection.driver_class JP.co.Hitachi.soft.HiRDB.JDBC.HiRDBDriver
hibernate.connection.url jdbc:hitachi:hirdb://DBID=@HIRDBENVGRP=C:/hirdb.ini
hibernate.connection.username hitachi
hibernate.connection.password hitachi
--
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
11 years, 4 months
[Hibernate-JIRA] Created: (HHH-3733) Introduce an annotation to disable ad-hoc null semantic of Hibernate components
by Vladimir Kovalyuk (JIRA)
Introduce an annotation to disable ad-hoc null semantic of Hibernate components
-------------------------------------------------------------------------------
Key: HHH-3733
URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-3733
Project: Hibernate Core
Issue Type: Improvement
Components: core
Reporter: Vladimir Kovalyuk
see http://forums.hibernate.org/viewtopic.php?t=993972
For the model
@Entity
public class Person {
private Address name = new Address();
...
@Embedded Address getAddress() {}
}
@Embeddable
public class Address {
public String getCity() {}
public String getStreet() {}
}
the following xhtml excerpt causes NPE constantly when editing a person which address properties hasn't been filled in yet
<h:inputText value="#{person.address.city}" />
I suggest introducing an annotation @Static which could be applied to Address class and lead to disabling null ad-hoc semantic on its instances.
For some environments such as user interface this may be considered as the default behavior. I suggest introducing an Hibernate configuration parameter which would switch off the semantic.
I would propose to consider embeddable classes as static mixins from OO design, not as just limited entities without identity. This point of view explains that embedded instance is a part of entity so it's lifecycle is the same as the entity's lifecycle. Thus constant presence of embeddable instance (property is never null) is the default behavior. Thus configuration parameter is "must have", not the thing which is "nice to have".
>From the other hand null ad-hoc semantic is like dynamic mixins. So it would be nice to have a @Dynamic annotation to force this semantic for some components.
--
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
11 years, 4 months
[Hibernate-JIRA] Created: (HHH-2308) Adjusting the Outer Join Predicate using Criteria Query
by Ben Grant (JIRA)
Adjusting the Outer Join Predicate using Criteria Query
-------------------------------------------------------
Key: HHH-2308
URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-2308
Project: Hibernate3
Type: New Feature
Components: query-criteria
Versions: 3.2.1
Environment: Linux Using MS SQLServer
Reporter: Ben Grant
I have two tables
Table A
||Col_1||Col_2||
|London| UK |
|Liverpool| UK |
| New York | USA |
Table B
||Col_1||Col_2|| Col_3||
| UK | Europe | 0
| USA | Americas | 1
Using the Criteria class, Restriction Class and FetchMode, Hibernate manages to create a query that looks like this
select distinct top 2000
this_.Col_1 as y0_, TableB3_.Col2 as y1_
from TableA this_
left outer join TableB TableB3_ on this_.Col_2= TableB3_.Col_1
where TableB3_.Col_3=1
When really i need the query to be like this
select distinct top 2000
this_.Col_1 as y0_, TableB3_.Col2 as y1_
from TableA this_
left outer join TableB TableB3_ on this_.Col_2= TableB3_.Col_1 AND TableB3_.Col_3=1
currently their isn't any know way for hibernate to adjust or apply filters within the join clause.
--
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
11 years, 4 months