Re: Why AC_C_CHAR_UNSIGNED?

[Stepan Kasal]

Hello,

On Thu, Jun 26, 2008 at 04:52:43AM -0500, Bob Friesenhahn wrote:
> On Thu, 26 Jun 2008, Stepan Kasal wrote:
>> Or would it help to put it all to one message?
>>
>>> AC_C_CHAR_UNSIGNED: The macro is obsolete.  Instead of checking
>>> __CHAR_UNSIGNED__, include limits.h and test for CHAR_MIN == 0.
>>
>> If your autotools maintainer cannot do that, you shall return him for
>> replacement of defective parts.   ;-)
>
> I would be truely aghast if I found that the several people who
> independently maintain autotools for my own package were doing this
> kind of surgury on it.

IMHO, this is not a strong reason.  Autoconfigury is interconnected
with the code anyway, and things like this can happen.  There has to
be an agreement about what is the autotoolser allowed to commit and
what needs your approval, anyway.

> Some of the worlds most valuable packages are not longer actively
> developed yet they are individually maintained by people who care
> to be sufficient (or acceptable according to some standard) for use
> in Cygwin, many Linux distributions, *BSD, etc. These sort of
> "deprecation" scenarios can lead to silent undetected failure due
> to an over-vigilant maintainer who is not prepared to adequately
> test the software.

This is much more important reason, at least in my opinion.
I'll try to keep it in mind for future "deprecations".

I hope that you like the -Wbest-prectices solution proposed by Ralf.

Have a nice day,
        Stepan