[
https://issues.apache.org/jira/browse/DERBY-3940?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12645803#action_12645803
]
Dag H. Wanvik commented on DERBY-3940:
--------------------------------------
It seems to me the code in CreateTriggerNode#bindStatement should include all
column references
in the set it constructs (referencedColInts), that would be your 1)expand
meaning option.
> Dropping a column does not drop triggers which mention that column
> ------------------------------------------------------------------
>
> Key: DERBY-3940
> URL: https://issues.apache.org/jira/browse/DERBY-3940
> Project: Derby
> Issue Type: Bug
> Components: SQL
> Affects Versions: 10.4.2.1, 10.5.0.0
> Reporter: Rick Hillegas
> Attachments: dropColumnWithTrigger.sql, Triggers.java
>
>
> Put an INSERT trigger on a table and mention a column in the trigger. Then
> drop that column from the table. If you drop the column with RESTRICT
> semantics, you don't get an objection. Both CASCADE and RESTRICT drop the
> column. However, the trigger remains in both cases. After that, INSERTs into
> the table fail because the trigger can't find the dropped column. The
> workaround is to manually drop the trigger either before or after dropping
> the column. I will attach a test case.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
