Re: [jira] Commented: (DERBY-819) Provide JDBC4 SQLException subclasses support in Embedded driver

[Anurag Shekhar <An...@Sun.COM>]

I didn't get it David if Its set to local locale it should fail unless
the expected output was generated for the same locale.
anurag

On Thu, 2006-01-19 at 12:04 -0800, David W. Van Couvering wrote:
> So it seems to me the right thing to do is to set your local locale :), 
> and not hardcode it in the tests.  Anurag, have you tried this?
> 
> Thanks,
> 
> David
> 
> Knut Anders Hatlen wrote:
> > "David W. Van Couvering" <Da...@Sun.COM> writes:
> > 
> > 
> >>Anurag Shekhar wrote:
> >>
> >>>Hi David
> >>>I will made the changes.
> >>>please see inline
> >>>thanks
> >>>anurag
> >>>David Van Couvering (JIRA) wrote:
> >>>
> >>>
> >>>>- Why do you set the territory to en_US?
> >>>>
> >>>>
> >>>
> >>>I am printing the message so in case the default locale is not the
> >>>same as I used to generate the original out file it will fail. So I
> >>>am setting territory to ensure I get the same message in every run.
> >>>
> >>
> >>Why is it that all tests don't do this?  I am curious what makes this
> >>test special so that it needs to have this set.    Or do most tests
> >>have diffs when you run it in a non-en_US locale?
> > 
> > 
> > FWIW, I always set the locale to en_US before running derbyall. I
> > haven't tried running with another locale lately, but I have seen
> > errors at least with the nn_NO and no_NO locales.
> > 

Re: [jira] Commented: (DERBY-819) Provide JDBC4 SQLException subclasses support in Embedded driver

["David W. Van Couvering" <Da...@Sun.COM>]

Just to be clear, I didn't mean set the locale to your "local locale" 
but set your "local locale" to be en_US

I think Dan raises a good point -- if you don't print out the error 
messages then this issue goes away, which was what I am rooting for anyway.

David

Daniel John Debrunner wrote:
> Anurag Shekhar wrote:
> 
> 
>>I didn't get it David if Its set to local locale it should fail unless
>>the expected output was generated for the same locale.
> 
> 
> I think the point is that the test should follow the same pattern as
> other tests, not be an exception. It's either a test harness issue (it
> should set the local to be en_US) or a generic problem that needs to be
> address (tests should pass in any locale).
> 
> Of course if you convert the test to a Junit test then the problem will
> go away.
> 
> Dan.
> 

Re: [jira] Commented: (DERBY-819) Provide JDBC4 SQLException subclasses support in Embedded driver

[Daniel John Debrunner <dj...@apache.org>]

Anurag Shekhar wrote:

> I didn't get it David if Its set to local locale it should fail unless
> the expected output was generated for the same locale.

I think the point is that the test should follow the same pattern as
other tests, not be an exception. It's either a test harness issue (it
should set the local to be en_US) or a generic problem that needs to be
address (tests should pass in any locale).

Of course if you convert the test to a Junit test then the problem will
go away.

Dan.