db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bernt M. Johnsen" <Bernt.John...@Sun.COM>
Subject Re: All of derby_all fails when environment corresponding derbyLocale_**.jar exists in CLASSPATH
Date Mon, 30 May 2005 13:32:11 GMT
>>>>>>>>>>>> Dag H. Wanvik wrote (2005-05-30 03:24:37):
> I used the setting below setting in the shell (bash) when running
> tests. Worked for me, but then only one test failed due to the no_NO
> setting originally (I forget which one, I can re-run to find out?).
> Otherwise, I agree, ideally tests should be agnostic to
> current language/region, 

The pprpoer solution is of course to stop using
getMessage()/toString() in tests. (Unless for tests of locales) To
identify errors, one would have enumerate them and use the
SQLException's vendorCode and getErrorCode() to check in the tests. I
believe the errorCode is not systematically used in Derby.

Bernt

> but if that isn't achievable, the harness
> could check for correct setting.
> 
> export LC_CTYPE=en_US
> 
> Dag

-- 
Bernt Marius Johnsen, Database Technology Group, 
Sun Microsystems, Trondheim, Norway

Mime
View raw message