db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Knut Anders Hatlen (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DERBY-4806) SysinfoCPCheckTest fail on different Windows platforms on 10.6.2.0 release candidate
Date Mon, 27 Sep 2010 12:30:33 GMT

    [ https://issues.apache.org/jira/browse/DERBY-4806?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12915288#action_12915288
] 

Knut Anders Hatlen commented on DERBY-4806:
-------------------------------------------

+                try {
+                    // DERBY-4806 Should use UTF-8 according to
+                    // http://www.w3.org/TR/html40/appendix/notes.html#non-ascii-chars
+                    // to get the string of the file name
+                    return URLDecoder.decode(result.toString(), "UTF-8");
+                } catch (UnsupportedEncodingException e) {
+                    // All JVMs are required to support UTF-8.
+                    return Main.getTextMessage("SIF01.W", e.getMessage());
+                }

The comment correctly states that all JVMs are required to support UTF-8, so the UnsupportedEncodingException
is never supposed to happen. Most other places we just ignore the UnsupportedEncodingException
if we know it can't happen, or have a THROWASSERT in a debug block for extra sanity checking.
Not a big issue, but I'd prefer that we didn't add a new message (SIF01.W) for something that
should never happen, so that we save the translators some unnecessary work. What about just
returning e.getMessage() in the case of UnsupportedEncodingException here, similar to what
you did with IOException in formatURL()?

> SysinfoCPCheckTest fail on different Windows platforms on 10.6.2.0 release candidate

> -------------------------------------------------------------------------------------
>
>                 Key: DERBY-4806
>                 URL: https://issues.apache.org/jira/browse/DERBY-4806
>             Project: Derby
>          Issue Type: Improvement
>          Components: Test
>    Affects Versions: 10.6.1.0
>         Environment: Windows
>            Reporter: Lily Wei
>            Priority: Minor
>         Attachments: DERBY-4806-1.diff, DERBY-4806-2.diff, rjall.out
>
>
> For the preliminary platform testing [1] Kathey, Dag and Lily
> have seen SysinfoCPCheckTest fail on different Windows platforms(XP and Windows 7) when
> run as part of suitesAll.
> [1] http://wiki.apache.org/db-derby/TenSixTwoPlatformTesting
> Thanks to Myrna points out that it could relate to DERBY-3771.

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


Mime
View raw message