Re: Help with bit-field semantics in C and C++

[Mike Stump <mrs at apple dot com>]

On Tuesday, August 24, 2004, at 02:40  PM, Gabriel Dos Reis wrote:
> | The intent of the standard is not to confuse bool into this...
>
> Huh?  Exactly which part of the above quote do you derive that from?

Read the paper that introduced bool into the standard and the standard 
just before that and review the complete logs of all that was said on 
the reflector for bool and in the meetings where it was discussed.

> If you change the (input) problem, you should expect a different
> output.  The enum
>
>      enum { foo = 3 } e;
>
> out to be represented by an integer type large enough to represent the
> value 3.  Can bool do that? No.  so, the new example you bring does
> not bear much relevance to the original example at discussion.

The _original_ example was:

extern "C" void abort(void);

enum E { e = 0x0f };

void test(int n)
{
  enum E x, y;

  x = (E)0x10;
  y = (E)n;

  if (x != y)
    abort ();
}

int main()
{
  test(0x10);
  return 0;
}

which if you squint, you can see it more closely matches my example...

> | e = 7;

Ok, I blew that one...  I was thinking of a slightly different test 
case.  One that made full use of the full range of the underlying type 
in a portable program.

T<sizeof (e)> x = 7;
memcpy ((char *)&e, (char *)&x, sizeof (e));

The issue remains the same...  Could be good DR material...

> Values of type bool are either true or false.42)

Yeah, and I was thinking of:

  42) Using a bool value in ways described by this  International  Stan-
  dard as ``undefined,'' such as by examining the value of an uninitial-
  ized automatic variable, might cause it to behave  as  if  is  neither
  true nor false.