[ 
https://issues.apache.org/jira/browse/IMAP-181?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Eric Charles updated IMAP-181:
------------------------------

    Attachment: jpa-migrator.zip

Very first draft uncomplete of jpa migrator project.
Comment welcome on the proposed approach:
- separate project
- main with list of jiras

Solutions such as http://www.liquibase.org/ may be a more solid approach, but 
we may need more flexibility, especially on the data level.
I suppore jpa will in the future better handle the migration and schema changes.

- https://issues.apache.org/jira/browse/IMAP-180 : Need manuall recreate the fk 
one by one (intermediary tables dissapear)
- https://issues.apache.org/jira/browse/IMAP-176 : Need to change #mail with 
#private in mailbox.namespace column
- https://issues.apache.org/jira/browse/IMAP-172 : Need to remove messagecount 
and size columns from mailbox
- https://issues.apache.org/jira/browse/IMAP-168 : Does seem to impact database 
but not sure. Need confirmation ? 
- https://issues.apache.org/jira/browse/IMAP-165 : JPA doesn't incrementally 
create the @Index. So we need to create manually


> Implement database schema and data incremental migration for 3.0 release
> ------------------------------------------------------------------------
>
>                 Key: IMAP-181
>                 URL: https://issues.apache.org/jira/browse/IMAP-181
>             Project: JAMES Imap
>          Issue Type: New Feature
>          Components: JPA Mailbox
>         Environment: all
>            Reporter: Eric Charles
>            Assignee: Eric Charles
>         Attachments: jpa-migrator.zip
>
>
> We need a way to upgrade the database (for JPA store) whenever it is needed.
> In general, a database upgrade is needed further a JIRA that impacts the 
> database.
> This can be:
> 1. On a db schema level (table, columns, fk, index,...)
> 2. On data level (update some data, feed a column with relation keys,...).
> Plain SQL would be enough for the first point (not for the second one). PLSQL 
> (or similar) would be needed for the second point.
> To ease things, a java "software" process will developed based on the 
> existing JPA layer.
> The tool should be invoked with JIRA numbers as parameters.
> Each provided JIRA number would invoke a DB Change Command.
> As POC (Proof of Concept), we will first develop a tool that takes into 
> account.
> - https://issues.apache.org/jira/browse/IMAP-176
> - https://issues.apache.org/jira/browse/IMAP-172
> - https://issues.apache.org/jira/browse/IMAP-168
> - https://issues.apache.org/jira/browse/IMAP-165

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to