[
https://issues.apache.org/jira/browse/DERBY-4553?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12835698#action_12835698
]
Knut Anders Hatlen commented on DERBY-4553:
-------------------------------------------
Hi Sylvain,
The fix looks fine to me, but I think we need to update the comments if we
start allowing other objects than strings in ijVectorResult. I'm wondering,
though, if we need the new class, or if it would be better to just add an extra
constructor to ijVectorResult that takes a string argument. Not a big issue,
but that would give slightly less code and less files to maintain.
(On a side note, this would not have been an issue if only ijVectorResult had
used java.util.List instead of java.util.Vector. Then we could produce a scalar
result as simple as this, without any new class or new constructor: "new
ijVectorResult(Collections.singletonList(result)), null);" But cleaning up the
ij code and making it use the more modern collection classes is outside the
scope of this issue.)
> In ij GETCURRENTROWNUMBER directly writeits result to output
> -------------------------------------------------------------
>
> Key: DERBY-4553
> URL: https://issues.apache.org/jira/browse/DERBY-4553
> Project: Derby
> Issue Type: Bug
> Components: Tools
> Reporter: Sylvain Leroux
> Assignee: Sylvain Leroux
> Priority: Trivial
> Attachments: DERBY-4553.patch, DERBY-4553_repro.patch
>
>
> In ij, the statement GETCURRENTROWNUMBER directly write its result to output
> instead of returning it:
> Here are the faulty lines in ij.ij, method GetCurrentRowNumber():
> ...
>
> LocalizedResource.OutputWriter().println(utilInstance.getCurrentRowNumber(rs));
> return null;
> This interferes with testing - and possibly with any external tool using the
> ij.ij parser.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
