[ 
https://issues.apache.org/jira/browse/OPENJPA-3076?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18143808#comment-18143808
 ] 

ASF subversion and git services commented on OPENJPA-3076:
----------------------------------------------------------

Commit de622f1d9bc1f5df3a258693066ef3c1dc74cba9 in openjpa's branch 
refs/heads/OPENJPA-3007 from Richard Zowalla
[ https://gitbox.apache.org/repos/asf?p=openjpa.git;h=de622f1d9 ]

Merge master into OPENJPA-3007

Resolve the conflicts with OPENJPA-3076 in the MySQL/MariaDB dictionary tests 
(JUnit 6 imports, the reset of the
shared identifier rule is no longer needed) and convert the tests that came 
with it (TestNewTableSchemaName,
TestXMLSchemaParserNames, TestNamesPerDictionary and the new tests of 
TestAbstractMySQLDictionary) to JUnit 6.


>  Delimiting of reserved words is a JVM-wide setting, so the dictionary 
> configured last decides it for all persistence units
> ---------------------------------------------------------------------------------------------------------------------------
>
>                 Key: OPENJPA-3076
>                 URL: https://issues.apache.org/jira/browse/OPENJPA-3076
>             Project: OpenJPA
>          Issue Type: Bug
>            Reporter: Richard Zowalla
>            Assignee: Richard Zowalla
>            Priority: Major
>             Fix For: 4.2.0
>
>
> Whether a reserved word is delimited is decided by a shared static 
> IdentifierRule. MySQLDictionary (and MariaDBDictionary since OPENJPA-3074) 
> switches it on, and every other dictionary resets it. With persistence units 
> for different databases in one JVM, the dictionary configured last decides 
> the names for all of them. For example, after Derby is configured, MySQL's 
> Window table becomes WINDOW0. Delimiting should be decided per configuration.



--
This message was sent by Atlassian Jira
(v10.3.26#10030026)

Reply via email to