[
https://issues.apache.org/jira/browse/OPENJPA-1449?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12801367#action_12801367
]
Milosz Tylenda commented on OPENJPA-1449:
-----------------------------------------
I guess the behaviour exists because OpenJPA modifies column name if it is an
SQL 92 reserved word even if a particular database allows it to be a column
name.
I can see these potential workarounds:
1. Extend SQLServerDictionary and override endConfiguration(). Call super and
modify reservedWordSet variable as you wish.
2. Modify org/apache/openjpa/jdbc/sql/sql-keywords.rsrc which is included in
OpenJPA jar or provide your own and somehow make the classloader load it first.
> TableGenerator pkColumnName / valueColumnName cannot be "name" / "value"
> respectively
> -------------------------------------------------------------------------------------
>
> Key: OPENJPA-1449
> URL: https://issues.apache.org/jira/browse/OPENJPA-1449
> Project: OpenJPA
> Issue Type: Bug
> Affects Versions: 1.2.1
> Environment: IBM 1.5 JDK
> databasePlatform=org.apache.openjpa.jdbc.sql.SQLServerDictionary
> Reporter: John Selby
> Priority: Minor
>
> @Id
> @GeneratedValue(strategy=GenerationType.TABLE,generator="KeyGen")
> @TableGenerator(name="KeyGen", table="rtt_sequence_number",
> pkColumnName="name",
> valueColumnName="value", pkColumnValue="como")
> Results in a query:
> SELECT VALUE0 FROM rtt_sequence_number WITH (UPDLOCK) WHERE NAME0 = ?
> [params=(String) como]
> The extra "zero" appended to the end of the string causes the query to fail.
> Using any other values for pkColumnName and valueColumnName results in valid
> SQL (where the values for pkColumnName and valueColumnName match the
> annotation)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
