[ http://issues.apache.org/jira/browse/DDLUTILS-89?page=comments#action_12374280 ] 

Jun Li commented on DDLUTILS-89:
--------------------------------

I found this issue also, and the fix I made locally was modify the readForeignKeys method in the JdbcModelReader class.

Used Map fks = new TreeMap() instead of Map fks = new ListOrderMap();

> Lack of ordering for foreign keys in WriteSchemaToFile results in spurious diffs
> --------------------------------------------------------------------------------
>
>          Key: DDLUTILS-89
>          URL: http://issues.apache.org/jira/browse/DDLUTILS-89
>      Project: DdlUtils
>         Type: Improvement

>  Environment: Oracle 10g EE
> SQLServer 2005 EE
> Windows XP SP2
> Java 1.5.0_06
>     Reporter: Guy Davis
>     Assignee: Thomas Dudziak
>  Attachments: order_fks.patch
>
> We use the WriteSchemaToFile task to store our applications schema in a file we place in our Subversion CM tool.  Every time we run this task, it reorders the foreign keys on a table.  So, for example, if Table A has FKs X, Y, Z then schema.xml file alternates between:
> <table ...>
>    <fk name='X' .../>
>    <fk name='Y' .../>
>    <fk name='Z' .../>
> </table>
> AND 
> <table ...>
>    <fk name='Z' .../>
>    <fk name='Y' .../>
>    <fk name='X' .../>
> </table>
> Notice, that the keys haven't changed between schema dumps, just the ordering of them in the file.  This means tracking version changes is more complicated as a developer's real changes are mixed in with these spurious key position changes.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira