[
https://issues.apache.org/jira/browse/OPENJPA-505?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Patrick Linskey updated OPENJPA-505:
------------------------------------
Affects Version/s: (was: 1.0.0)
(was: 1.0.1)
1.1.0
Changing affected version information since OPENJPA-455 was not in the 1.0.x
line.
> CLONE -Incorrect Oracle DDL Generation for integer types since OPENJPA-455
> ---------------------------------------------------------------------------
>
> Key: OPENJPA-505
> URL: https://issues.apache.org/jira/browse/OPENJPA-505
> Project: OpenJPA
> Issue Type: Bug
> Affects Versions: 1.1.0
> Reporter: Joe Weinstein
> Assignee: Michael Dick
> Fix For: 1.1.0
>
> Attachments: my-change.patch
>
>
> Since the fix for 455, I've found a problem. Oracle has a NUMBER
> column type, which may be specified with sizes, eg: NUMBER(7),
> or NUMBER(9,3), but may also be unsized, eg: NUMBER. Since the
> fix for 455, the new code is failing to process the "NUMBER{0}" template
> for non-sized columns, so the DDL sent to the DBMS is like:
> CREATE TABLE ABSTRACTMAPPEDAPPIDSUPER ( ..., VERSN NUMBER{0}, ...
> which needless to say, dies.
> A change I made to DBDictionary, that fixed this for me is:
> protected String insertSize(String typeName, String size) {
> if(StringUtils.isEmpty(size)) {
> int idx = typeName.indexOf("{0}"); // remove the size token if not
> needed...
> if (idx != -1) {
> return typeName.substring(0,idx);
> }
> return typeName;
> }
> int idx = typeName.indexOf("{0}");
> ...
> ie:
> Index:
> openjpa-jdbc/src/main/java/org/apache/openjpa/jdbc/sql/DBDictionary.java
> ===================================================================
> --- openjpa-jdbc/src/main/java/org/apache/openjpa/jdbc/sql/DBDictionary.java
> (revision 610999)
> +++ openjpa-jdbc/src/main/java/org/apache/openjpa/jdbc/sql/DBDictionary.java
> (working copy)
> @@ -1648,8 +1648,13 @@
> */
> protected String insertSize(String typeName, String size) {
> if(StringUtils.isEmpty(size)) {
> - return typeName;
> - }
> +
> + int idx = typeName.indexOf("{0}");
> + if (idx != -1) {
> + return typeName.substring(0,idx);
> + }
> + return typeName;
> + }
> int idx = typeName.indexOf("{0}");
> if (idx != -1) {
> Please let me know what you think, and how to absorb this
> change, or it's purpose. thanks,
> Joe Weinstein at BEA Systems
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
