[
https://issues.apache.org/jira/browse/TORQUE-128?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12907916#action_12907916
]
Thomas Vandahl commented on TORQUE-128:
---------------------------------------
I tend to disagree. If I'm not mislead, I can have columns which are primary
keys but have no auto-increment flag on (or no associated sequence, for that
matter). N:M-tables are a typical example or tables with compound primary keys.
How would you handle those? As I see it, this modification would impose a
serious limitation on the way you describe database schemas with Torque.
Besides, Torque handles sequences just fine with database systems that support
them.
> remove autoIncrement attribute of column element
> ------------------------------------------------
>
> Key: TORQUE-128
> URL: https://issues.apache.org/jira/browse/TORQUE-128
> Project: Torque
> Issue Type: Sub-task
> Affects Versions: 4.0
> Reporter: Thomas Fischer
> Fix For: 4.0
>
>
> In the 3.3 schema, the id method for a column is determined by the idMethod
> attribute of the table, the defaultIdMethod attribute of the database and the
> primaryKey attribute of the column. This is sufficient for determining the
> idMethod.
> However, in the 3.3 schema, there is still the autoIncrement attribute which
> kind of overrides the settings outlined above. There is no such equivalent
> for sequence generation, thus the attribute is bad for database
> interoperability.
> Therefore I'd suggest to remove this attribute with no replacement. The
> consequence would be that it would no longer be possible to have
> auto_increment on columns which are no primary key (but again, torque
> supports no such feature for sequence generation).
--
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]
