Re: [Suggestion] GNUstep-test for quality control (WAS: Re: deferred deallocation)

[Chris B. Vetter]

On Tue, 14 Oct 2003 21:59:31 -0400
"Yen-Ju Chen" <yjchenx@hotmail.com> wrote:
[...]
>   How well the gnustep-test can detect the bugs depends on how well it
>   is written.
>   We can just start some and add more when new bug is discovered.
>   Over the time, it will cover over every details.

Yes, but even a very well written test-suite will not be able to find
bugs, as in "the current implementation is broken by design because the
implementor has no f*cking idea how to implement it correctly in the
first place" bugs.

As a simple example (without pointing any fingers):
There is a method that Apple's documentation states as returning an
NSData, while GNUstep's implementation returns an NSString.

No big deal, mind you, but in the end it's a bug.

>   Typo is another issue and I prefer not to touch this one. :)

Sissy ;-)

-- 
Chris