Re: func is not a keyword

From	Keith Thompson <Keith.S.Thompson+u@gmail.com>
Newsgroups	comp.lang.c
Subject	Re: __func__ is not a keyword
Date	2025-03-15 14:53 -0700
Organization	None to speak of
Message-ID	<874izuhsfr.fsf@nosuchdomain.example.com> (permalink)
References	<vr4lgu$63fu$1@dont-email.me> <87bju2htxy.fsf@nosuchdomain.example.com> <vr4rh9$b38v$1@dont-email.me>

Show all headers | View raw

Thiago Adams <thiago.adams@gmail.com> writes:
> Em 3/15/2025 6:20 PM, Keith Thompson escreveu:
>> Thiago Adams <thiago.adams@gmail.com> writes:
>>> This program does not compile..in gcc and clang
>>>
>>> int __func__  i =1;
>>>
>>> int main () {}
>>>
>>> error: expected identifier...
>> The "i" makes that a syntax error anyway, even if "__func__" were
>> accepted as an ordinary identifier.  Still, you have found something
>> odd.
>> 
>
> Yes sorry...
>
>>> Standard has..
>>>
>>> "The identifier __func__ shall be implicitly declared by the
>>> translator as if, immediately following
>>> the opening brace of each function definition, the declaration" ...
>>>
>>>
>>> My understand is that __func__   is not a keyword and that is
>>> something defined inside the functions.. so I don’t know why gcc and
>>> clang complains in the file scope.
>> If I change your program to:
>>      int __func__ = 1;
>>      int main(void) {}
>> 
>
>
> yes..now I don't known if I did it right at first time.
> but the problem I was talking about still there. :)
>
>> both gcc and clang complain "error: expected identifier or ‘(’".
>> If I change it from __func__ to __foo__, neither gcc nor clang
>> complains.
>>      All identifiers that begin with a double underscore (__) or
>> begin
>>      with an underscore (_) followed by an uppercase letter are reserved
>>      for any use, except those identifiers which are lexically identical
>>      to keywords.
>>      ...
>>      If the program declares or defines an identifier in a context in
>>      which it is reserved (other than as allowed by 7.1.4), the behavior
>>      is undefined.
>> So the program has undefined behavior, which means that terminating
>> the translation with the issuance of a diagnostic message is
>> permitted.
>> I'm mildly curious how gcc and clang treat "__func__" internally
>> that leads to this odd behavior.  The obvious way to implement it
>> would be to internally create a declaration of __func__ on entry
>> to each function definition, which shouldn't cause the symptom
>> you're seeing.  But it's not a conformance issue
>
> I think __func__ is begin used as keyword inside gcc /clang.

That seems to be consistent with the behavior I see.

> Thanks for clarifying it.
> I think this UB could be removed.

If you mean that the program could be modified so it no longer has
undefined behavior, of course it could.  Just change __func__ to some
non-reserved identifier.

If you mean that the C standard could be changed so that declaring a
reserved identifier doesn't have undefined behavior, that would be
difficult, and I don't see the point.  Implementations are free to use
reserved identifiers for their own purposes.  That's why they're
reserved.

"Doctor, it hurts when I do this."

-- 
Keith Thompson (The_Other_Keith) Keith.S.Thompson+u@gmail.com
void Void(void) { Void(); } /* The recursive call of the void */

Back to comp.lang.c | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

__func__ is not a keyword Thiago Adams <thiago.adams@gmail.com> - 2025-03-15 16:47 -0300
  Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-15 14:20 -0700
    Re: __func__ is not a keyword Thiago Adams <thiago.adams@gmail.com> - 2025-03-15 18:30 -0300
      Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-15 14:53 -0700
        Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-15 15:04 -0700
    Re: __func__ is not a keyword Kaz Kylheku <643-408-1753@kylheku.com> - 2025-03-16 07:18 +0000
      Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-16 02:06 -0700
        Re: __func__ is not a keyword Tim Rentsch <tr.17687@z991.linuxsc.com> - 2025-03-18 10:00 -0700
    Re: __func__ is not a keyword scott@slp53.sl.home (Scott Lurndal) - 2025-03-16 15:51 +0000
      Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-16 11:51 -0700
        Re: __func__ is not a keyword Kaz Kylheku <643-408-1753@kylheku.com> - 2025-03-16 19:05 +0000
          Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-16 13:29 -0700
            Re: __func__ is not a keyword Kaz Kylheku <643-408-1753@kylheku.com> - 2025-03-17 17:03 +0000
              Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2025-03-17 13:09 -0700
                Re: __func__ is not a keyword Kaz Kylheku <643-408-1753@kylheku.com> - 2025-03-17 20:31 +0000
          Re: __func__ is not a keyword James Kuyper <jameskuyper@alumni.caltech.edu> - 2025-03-16 16:35 -0400
  Re: __func__ is not a keyword Mario Rosell <usenet@mariorosell.es> - 2026-04-17 20:19 +0000
    Re: __func__ is not a keyword Keith Thompson <Keith.S.Thompson+u@gmail.com> - 2026-04-17 14:15 -0700

csiph-web