test_nonascii: two unicode a's

[Alexander Belchenko]

John Arbash Meinel пишет:
> Aaron Bentley wrote:
>> Alexander Belchenko wrote:
>>>> We discuss that test_contents_merge2 should be skipped on win32 because
>>>> diff3 on win32 change line-endings. Attached patch fix this.
>> Well, if typical implementations of diff3 on win32 are broken, I think
>> we should not permit merge-type diff3 on that platform at all.  The test
>> failure indicates that, on win32, --merge-type diff3 really is broken.
> 
> I agree with you. And I can also say that my cygwin implementation of
> diff3 actually is doing the right thing. (The tests pass on my system).
> 
> I'm not 100% comfortable with skipping tests just because they might
> fail, especially because of external factors (whether OS or an external
> program).
> 
> Using 'bzr merge --merge-type=diff3' will potentially do bad things on
> Alexander's system. It seems like he should be made aware of that, by
> something like a test failure.

Well, this test might be split into two part: one part use text file 
with native line-endings, and other use binary for (i.e. LF line 
endings). After both part execute their job test can detect is current 
user's platform + diff3 can handle with diff3 properly. And based on 
this information make decision about limitations of current user. But 
this will true only for current user. I don't think that much users 
(especially on Windows) will run overall selftest to detect what 
limitations they have.

If we already know weak points we could provide text with attention 
(during installation or in the documentation) or provide special command 
a-la "bzr limitation" to check only known limitation and print to user 
what he have.

--
Alexander