The pushdown sql for version columns in primary and secondary tables has
unpredictable ordering in the select and from clause
-----------------------------------------------------------------------------------------------------------------------------
Key: OPENJPA-720
URL: https://issues.apache.org/jira/browse/OPENJPA-720
Project: OpenJPA
Issue Type: Improvement
Affects Versions: 1.2.0, 1.1.0, 1.0.3, 1.0.2, 1.0.1, 1.0.0, 0.9.7, 0.9.6,
0.9.0, 1.0.4
Reporter: Catalina Wei
We observe that two differenct pushdown sql could be generated for entities
with multi-table mapping and version columns in each table.
Here is an example, where entity is having 1 primary table, 2 secondary tables,
and each table has a version column, an em.find() could result
in following sqls:
SELECT t0.version,
t1.version,
t2.version,
t1.col1, t0.col2, t2.col3
FROM TS2 t0
INNER JOIN TS1 t1 ON t0.ID = t1.id
INNER JOIN TS3 t2 ON t1.id = t2.ID
WHERE t1.id = ? optimize for 1 row
SELECT t0.version,
t1.version,
t2.version,
t2.col1, t1.col2, t0.col3
FROM TS3 t0
INNER JOIN TS1 t2 ON t0.ID = t2.id
INNER JOIN TS2 t1 ON t2.id = t1.ID
WHERE t2.id = ? optimize for 1 row
If the HashMap is changed to LinkedHashMap, then the pushdown sql will have
predictable ordering in select and from clause.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
