[
https://issues.apache.org/jira/browse/DERBY-3512?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tiago R. Espinha updated DERBY-3512:
------------------------------------
Issue & fix info: [High Value Fix, Repro attached]
Urgency: Normal
Triaged for 10.5.2.
Checked High Value Fix and Repro attached.
Assigned Normal urgency.
> Using functions, you can create a LONG VARCHAR value which is longer than the
> maximum length of a Derby LONG VARCHAR
> --------------------------------------------------------------------------------------------------------------------
>
> Key: DERBY-3512
> URL: https://issues.apache.org/jira/browse/DERBY-3512
> Project: Derby
> Issue Type: Bug
> Components: SQL
> Affects Versions: 10.4.1.3
> Reporter: Rick Hillegas
> Attachments: BigStringMaker.java, bigStrings.sql
>
>
> I can create a static method which returns a long String and register this as
> a function which returns LONG VARCHAR. I am able to invoke this function and
> create a LONG VARCHAR value which is longer than the maximum length which
> Derby allows for this datatype (32700 characters).
> It is not clear to me what is supposed to happen in this situation according
> to the SQL standard. However, the current behavior seems wrong. Here are two
> possible correct behaviors:
> 1) Raise an exception if the user-coded method violates the declared contract
> of the CREATE FUNCTION statement.
> 2) Silently truncate the value returned by the user-coded method.
> There may be a similar problem with methods which return byte[] and which are
> registered as functions returning (LONG) VARCHAR FOR BIT DATA. However,
> experiments to test this are blocked by DERBY-3511.
> I will attach a test case demonstrating the problem.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
