openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "bernard (JIRA)" <j...@apache.org>
Subject [jira] [Created] (OPENJPA-2041) Unnecessary Table Join in Native Many-to-Many Query generated from JPQL
Date Fri, 19 Aug 2011 09:38:27 GMT
Unnecessary Table Join in Native Many-to-Many Query generated from JPQL
-----------------------------------------------------------------------

                 Key: OPENJPA-2041
                 URL: https://issues.apache.org/jira/browse/OPENJPA-2041
             Project: OpenJPA
          Issue Type: Bug
          Components: query
    Affects Versions: 2.1.1
         Environment: java version "1.6.0_25"
Java(TM) SE Runtime Environment (build 1.6.0_25-b06)
Java HotSpot(TM) Client VM (build 20.0-b11, mixed mode, sharing)
            Reporter: bernard


A rather simple JPQL query produces an unnecessary table join.
Please refer to the attached testcase (NetBeans project).

JPQL:

select
    object(emp)
from
    Employee emp
    , in(emp.projects) proj
where
    proj.id = :projectId


The generated native SQL is:

SELECT
    t0.id
    , t0.name
FROM
    Employee t0
    INNER JOIN Employee_Project t1 ON t0.id = t1.EMPLOYEE_ID
    INNER JOIN Project t2 ON t1.PROJECTS_ID = t2.id
WHERE (
    t1.PROJECTS_ID = ?)

The generated SQL should be like the following, in other words the project table should not
be included.

SELECT
    t1.ID
    , t1.NAME
FROM
    , EMPLOYEE_PROJECT t2
    , EMPLOYEE t1
WHERE (
    (t2.projects_ID = ?)
    AND ((t2.Employee_ID = t1.ID)
)


--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message