Grouper Developers Forum

[Chris Hyzer <>]

Hey,

 

I wrapped this enhancement up.  Here are some advantages:

 

- Don’t want to have to xml export / import on upgrades that require ddl changes

- Will do ddl diffs easily (e.g. if you accidentally drop a column, it will be detected)

- Some things hibernate schemaexport wont do (e.g. multiple indexes on the same columns, foreign keys)

- Can mix ddl with sql (e.g. to remove the id cols)

- Can have custom (not ddlutils generated) ddl/sql for particular databases easily

 

https://wiki.internet2.edu/confluence/display/GrouperWG/Ddl+management

https://bugs.internet2.edu/jira/browse/GRP-132

 

It is unit tested and all unit tests work, though I wouldn’t be surprised to see issues arise, please let me know asap.  In oracle (and maybe other non-mysql dbs), ddlutils has issues with unique constraints (not indexes), so some foreign keys don’t work.  When I remove the ID cols (next step), the UUIds will be primary keys, so most of those issues will go away, and the others can be solved with custom sql I think…

 

The next time you upgrade, follow the change log step:

 

https://wiki.internet2.edu/confluence/display/GrouperWG/Grouper+change+log+v1.3+-+v1.4

 

2008/07/27: v1.4 HEAD: DDL management changed.

   - Copy the DDL section from the grouper.example.properties into your grouper.properties, configure them

   - Add the jar to lib dir: ant-1.7.1.jar

   - run: ant schemaexport to sync everything up (or generate the script to sync everything up, depending on grouper.properties)

   - Remove any schema-export.sql files

 

Regards,

Chris