July 2010 - hibernate-issues - Jboss List Archives

[Hibernate-JIRA] Updated: (HHH-892) HQL parser does not resolve alias in ORDER BY clause

by Gail Badner (JIRA)

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-892?page=co... ] Gail Badner updated HHH-892: ---------------------------- Attachment: HHH-892.patch I'm attaching a patch of my first attempt at a fix. It replaces aliases for HQL select expressions that are used in ORDER BY clauses with the actual SQL column aliases that Hibernate generated. For example: HQL: select c.name as myname FROM org.hibernate.test.jpa.Item c ORDER BY myname results in Hibernate generating: SQL: select item0_.NAME as col_0_0_ from EJB3_ITEM item0_ order by col_0_0_ >From discussions with Steve Ebersole, I understand that this may not work for all dialects, so something will need to be added to Dialect to indicate how aliases for HQL select expressions should be replaced. The other possibilities are to replace the alias for an HQL select expression with: 1) the position(s) of the SQL select expressions that correspond with the HQL select expression alias (e.g., select item0_.NAME as col_0_0_ from EJB3_ITEM item0_ order by 1) 2) the actual SQL columns that correspond to the aliased HQL select expression (this probably will not work for all dialects). The fix in the patch will not work reliably if a position, rather than an alias is used in the HQL ORDER BY clause. That's because the position in the HQL will simply "fall through" to the generated SQL. Since HQL select expressions can result in multiple SQL expressions, the position of an HQL expression could translate to a different (or multiple) positions in the generated SQL. Does anyone know about a dialect that should use positions instead of column aliases in the ORDER BY clause? Any help working out these details would be most appreciated. I'd really like to get this fixed as well. Thanks, Gail > HQL parser does not resolve alias in ORDER BY clause > ----------------------------------------------------- > > Key: HHH-892 > URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-892 > Project: Hibernate Core > Issue Type: Bug > Components: query-hql > Affects Versions: 3.5.4, 3.6.0.Beta1 > Environment: Hibernate 3.0.5, MySQL, Tomcat > Reporter: Guido Laures > Assignee: Gail Badner > Fix For: 3.5.x, 3.6.x > > Attachments: HHH-892.patch, HibernateHavingAliasTest.java > > > When using an alias for an ORDER BY clause this is not always correctly resolved. Example: > SELECT SUM(A.x) AS mySum FROM MyClass AS A GROUP BY A.y ORDER BY mySum > does not work because "mySum" is not resolved in the ORDER BY clause which results in an exception telling that mySum is an unknown column. > Workaround (not to say "hack") is using: > SELECT SUM(A.x) AS mySum FROM MyClass AS A GROUP BY A.y ORDER BY col_0_0_ -- 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

14 years, 5 months

1
0
0 / 0

[Hibernate-JIRA] Updated: (HHH-892) HQL parser does not resolve alias in ORDER BY clause

by Gail Badner (JIRA)

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-892?page=co... ] Gail Badner updated HHH-892: ---------------------------- Priority: Major (was: Minor) > HQL parser does not resolve alias in ORDER BY clause > ----------------------------------------------------- > > Key: HHH-892 > URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-892 > Project: Hibernate Core > Issue Type: Bug > Components: query-hql > Affects Versions: 3.5.4, 3.6.0.Beta1 > Environment: Hibernate 3.0.5, MySQL, Tomcat > Reporter: Guido Laures > Assignee: Gail Badner > Fix For: 3.5.x, 3.6.x > > Attachments: HibernateHavingAliasTest.java > > > When using an alias for an ORDER BY clause this is not always correctly resolved. Example: > SELECT SUM(A.x) AS mySum FROM MyClass AS A GROUP BY A.y ORDER BY mySum > does not work because "mySum" is not resolved in the ORDER BY clause which results in an exception telling that mySum is an unknown column. > Workaround (not to say "hack") is using: > SELECT SUM(A.x) AS mySum FROM MyClass AS A GROUP BY A.y ORDER BY col_0_0_ -- 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

14 years, 5 months

1
0
0 / 0

[Hibernate-JIRA] Updated: (HHH-892) HQL parser does not resolve alias in ORDER BY clause

by Gail Badner (JIRA)

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-892?page=co... ] Gail Badner updated HHH-892: ---------------------------- Assignee: Gail Badner Affects Version/s: (was: 3.0.5) 3.5.4 3.6.0.Beta1 Fix Version/s: 3.6.x 3.5.x > HQL parser does not resolve alias in ORDER BY clause > ----------------------------------------------------- > > Key: HHH-892 > URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-892 > Project: Hibernate Core > Issue Type: Bug > Components: query-hql > Affects Versions: 3.5.4, 3.6.0.Beta1 > Environment: Hibernate 3.0.5, MySQL, Tomcat > Reporter: Guido Laures > Assignee: Gail Badner > Priority: Minor > Fix For: 3.5.x, 3.6.x > > Attachments: HibernateHavingAliasTest.java > > > When using an alias for an ORDER BY clause this is not always correctly resolved. Example: > SELECT SUM(A.x) AS mySum FROM MyClass AS A GROUP BY A.y ORDER BY mySum > does not work because "mySum" is not resolved in the ORDER BY clause which results in an exception telling that mySum is an unknown column. > Workaround (not to say "hack") is using: > SELECT SUM(A.x) AS mySum FROM MyClass AS A GROUP BY A.y ORDER BY col_0_0_ -- 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

14 years, 5 months

1
0
0 / 0

[Hibernate-JIRA] Updated: (HHH-1570) criteria-api: filtering by key-many-to-one causes invalid sql

by Gail Badner (JIRA)

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-1570?page=c... ] Gail Badner updated HHH-1570: ----------------------------- Affects Version/s: (was: 3.1.2) 3.5.3 3.6.0.Beta1 > criteria-api: filtering by key-many-to-one causes invalid sql > ------------------------------------------------------------- > > Key: HHH-1570 > URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-1570 > Project: Hibernate Core > Issue Type: Bug > Components: query-criteria > Affects Versions: 3.5.3, 3.6.0.Beta1 > Reporter: Joris Verschoor > Attachments: HHH-1570-fix.patch, HHH-1570-it.patch, hibernate.patch > > > We have a class Price that has a composite-id using a couple of key-properties and some key-many-to-ones) > One of those is "transportation", which has a many-to-one to transportationType > When we query using: criteria.add(Expression.eq("price.transportation.transportationType.id", transId);, we get an invalid SQL statement: The table of transportation was not selected. > I will try to make a testcase today or tomorrow, depending on my schedule.. Things will be more clear by then... > We have created a work-around, by mapping the transportationID twice: once in key-property, and once as a many-to-one. (instead of one key-many-to-one) -- 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

14 years, 5 months

1
0
0 / 0

[Hibernate-JIRA] Created: (HHH-5399) Unidirectional one-to-many relationship fails when creating referenced entity object (testcase provided)

by Jürgen Zimmermann (JIRA)

Unidirectional one-to-many relationship fails when creating referenced entity object (testcase provided) -------------------------------------------------------------------------------------------------------- Key: HHH-5399 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5399 Project: Hibernate Core Issue Type: Bug Components: entity-manager Affects Versions: 3.5.3 Reporter: Jürgen Zimmermann Attachments: testcase.zip The attached "Maven-ized" testcase works as follows: * Create an H2 database with tables t_customer and t_order, and insert one customer with two orders. * Compile and run the test class: both test methods fail. 1st method: * lookup the customer and verify the 2 referenced orders * create a new order for the customer * EntityManager.merge(customer) 2nd method: * lookup the customer and verify the 2 referenced orders * create a new order for the customer * EntityManager.persist(newOrder) -- 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

14 years, 5 months

2
2
0 / 0

[Hibernate-JIRA] Created: (HHH-5246) Addition of withClause for DetachedCriteria

by Shawn Clowater (JIRA)

Addition of withClause for DetachedCriteria ------------------------------------------- Key: HHH-5246 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5246 Project: Hibernate Core Issue Type: Patch Components: core Affects Versions: 3.5.2 Reporter: Shawn Clowater Priority: Minor Attachments: hhh-5246.patch HHH-2308 added the ability to specify a withClause to the createCriteria and createAlias methods but were never added to the DetachedCriteria API. I can see that being handy for us in the future so voila, patch. -- 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

14 years, 5 months

2
5
0 / 0

[Hibernate-JIRA] Created: (HHH-5296) AbstractFromImpl::getJoin() shall return empty set, not null

by Sergey Vladimirov (JIRA)

AbstractFromImpl::getJoin() shall return empty set, not null ------------------------------------------------------------ Key: HHH-5296 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5296 Project: Hibernate Core Issue Type: Bug Components: entity-manager Affects Versions: 3.5.0-Final Reporter: Sergey Vladimirov According to JavaDoc AbstractFromImpl::getJoin() shall return empty set if no joins were made, not null reference. Either javadoc or AbstractFromImpl behavior shall be fixed. -- 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

14 years, 5 months

3
11
0 / 0

[Hibernate-JIRA] Created: (HHH-5295) Rendered JPAQL query shall be the same all the times, aliases shall not have random indexes

by Sergey Vladimirov (JIRA)

Rendered JPAQL query shall be the same all the times, aliases shall not have random indexes ------------------------------------------------------------------------------------------- Key: HHH-5295 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5295 Project: Hibernate Core Issue Type: Improvement Components: entity-manager Affects Versions: 3.5.0-Final Reporter: Sergey Vladimirov Priority: Trivial Each time JPAQL query is rendered from CriteriaQuery it has new aliases for table aliases. It should be the same for same CriteriaQuery (or CriteriaQueries that builded in the same way) all the times. Reason: QueryStructure uses HashSet to store roots. Solution: replace HashSet with LinkedHashSet: from private Set<Root<?>> roots = new HashSet<Root<?>>(); to private Set<Root<?>> roots = new LinkedHashSet<Root<?>>(); -- 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

14 years, 5 months

3
8
0 / 0

[Hibernate-JIRA] Created: (HHH-5283) Add BasicType handling of java.net.URL

by Steve Ebersole (JIRA)

Add BasicType handling of java.net.URL -------------------------------------- Key: HHH-5283 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5283 Project: Hibernate Core Issue Type: Improvement Components: core Reporter: Steve Ebersole Assignee: Steve Ebersole Fix For: 3.6 Create a BasicType for handling java.net.URL. See org.hibernate.type.BasicTypeRegistryTest.UrlType in trunk (3.6) -- 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

14 years, 5 months

1
3
0 / 0

[Hibernate-JIRA] Created: (HHH-5401) Update to HHH-5381 HSQLDB new dialect (Fred Toussi)

by Fred Toussi (JIRA)

Update to HHH-5381 HSQLDB new dialect (Fred Toussi) ---------------------------------------------------- Key: HHH-5401 URL: http://opensource.atlassian.com/projects/hibernate/browse/HHH-5401 Project: Hibernate Core Issue Type: Patch Environment: HSQLDB Hibernate 3.5.x Reporter: Fred Toussi Attachments: HSQLDialect.diff Issue opened as advised by Steve. An updated version of HHH-581 that reflects some user feedback with existing databases upgraded to 2.0 and new databases. In this version BOOLEAN is included in the list of types and the literal values for BOOLEAN are defined as "true" or "false" instead of "1" or "0". Both v. 1.8.x and 2.0 support BOOLEAN (with true and false). In 1.8.x BIT is a synonym for BOOLEAN. In 2.0, BIT is a separate type representing a bit map which can have 0 or 1. As BOOLEAN is the prefered type in terms of storage requirements and usage in queries, it is better if this is used directly. I did not make this change initially as I was concerned about possible effects with existing databases. It now seems the change has no side effects and actually results in better SQL conformance. -- 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

14 years, 5 months

3
4
0 / 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

hibernate-issues July 2010