Path: csiph.com!news.mixmin.net!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <Keith.S.Thompson+u@gmail.com>
Newsgroups: comp.lang.c
Subject: Re: Is C ready to become a safer language?
Date: Sun, 11 Feb 2024 13:47:12 -0800
Organization: None to speak of
Lines: 51
Message-ID: <87o7cmad27.fsf@nosuchdomain.example.com>
References: <uq1jnk$1qebo$2@dont-email.me> <uq2vjp$21qev$1@dont-email.me> <86eddl5bag.fsf@linuxsc.com> <uq8lus$3dceu$1@dont-email.me> <8734tzdepl.fsf@nosuchdomain.example.com> <uqb5bq$136n4$1@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain
Injection-Info: dont-email.me; posting-host="a3ea5551d1916ca86f611a52355c117b"; logging-data="1205929"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18wOEK7KwXht1hyO0ht6PuZ"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.2 (gnu/linux)
Cancel-Lock: sha1:ffneEqoldwYcnj3+7HBLrVRgJHk= sha1:/VtW+Kf7o4VzIAjyJE6IOu7ibGk=
Xref: csiph.com comp.lang.c:382350

Thiago Adams <thiago.adams@gmail.com> writes:
> Em 2/10/2024 9:31 PM, Keith Thompson escreveu:
>> bart <bc@freeuk.com> writes:
>>> On 10/02/2024 01:59, Tim Rentsch wrote:
>>>> bart <bc@freeuk.com> writes:
>>>> [...]
>>>>
>>>>> This is something which has long been of fascination to me:  how
>>>>> exactly do you get a C compiler to actually fail a program with a
>>>>> hard error when there is obviously something wrong, while not also
>>>>> failing on completely harmless matters.
>> The only thing that *requires* a compiler to reject a translation
>> unit
>> is the #error directive.  For any violation of a syntax rule or
>> constraint, the standard only requires a *diagnostic message*, which can
>> be a non-fatal warning.
>
> I haven't checked the standard but
>
> #if 1/0
> #endif
>
>
> <source>:4:6: error: division by zero in preprocessor expression
>     4 | #if 1/0
>       |     ~^~
>
>
> stops compilation.

Then I suggest you check the standard.  It's perfectly valid for an
implementation to choose to stop compilation if it encounters 1/0 in a
preprocessor expression, but it's not required.  "The resulting tokens
compose the controlling constant expression which is evaluated according
to the rules of 6.6."  6.6, Constraints: "Each constant expression shall
evaluate to a constant that is in the range of representable values for
its type."  So that's simply a constraint violation, requiring a
diagnostic but not requiring the translation unit to be rejected.

> With constexpr in C23 I guess division by 0 will stop code generation
> as well.

No, it merely requires a constant expression.
    constexpr int n = 1/0;
is a constraint violation, requiring a diagnostic but not requiring
rejection.

-- 
Keith Thompson (The_Other_Keith) Keith.S.Thompson+u@gmail.com
Working, but not speaking, for Medtronic
void Void(void) { Void(); } /* The recursive call of the void */