In article <1995Jul13.111446.22483@friend.kastle.com>,
Richard Krehbiel <rich@kastle.com> wrote:
> If bool++ means bool=true then what about bool+=i?  Where i is zero?

It would leave bool unchanged.  A positive i would set bool true and a
negative i would set it false.
--
Shields.

>>>>> "MS" == Michael Shields <shields@tembel.org> writes:

 MS> In article <1995Jul13.111446.22483@friend.kastle.com>,
 MS> Richard Krehbiel <rich@kastle.com> wrote:
 >> If bool++ means bool=true then what about bool+=i?  Where i is zero?

 MS> It would leave bool unchanged.  A positive i would set bool true and a
 MS> negative i would set it false.

If you implement `bool+=i' in terms of the existing implicit conversions, then
`b+=i' is `b=b+i' is `b=bool(int(b)+i)' is one of

   b=bool(0+i)
or
   b=bool(1+i)

which has different truth values depending on the value of i and the previous
value of b.

Michael.

In article <MCOOK.95Jul16141925@erawan.cognex.com>,
Michael Cook  <mcook@cognex.com> wrote:
> If you implement `bool+=i' in terms of the existing implicit conversions, then
> `b+=i' is `b=b+i' is `b=bool(int(b)+i)' is one of
>
>    b=bool(0+i)
> or
>    b=bool(1+i)
>
> which has different truth values depending on the value of i and the previous
> value of b.

If you don't add new semantics to bool, why make it a type at all,
when you could just treat an int as a boolean, as people do now?
--
Shields.

>>>>> "MS" == Michael Shields <shields@tembel.org> writes:

 MS> If you don't add new semantics to bool, why make it a type at all,
 MS> when you could just treat an int as a boolean, as people do now?

Overloading.

BTW, has someone pointed out yet that `E1 op= E2' is ill formed if E1 is of
type `bool'?

Michael.

shankar@sgi.com (Shankar Unni) wrote:

>bool-- is an undefined operation, and is not guaranteed to set the boolean
>to zero. bool++ will, however, set it to true.

>I guess the intent was that implementations can quietly implement bool as
>"unsigned char" and do nothing else, but that's clearly silly, since they
>would still have to guard against 256 ++'s in a row.

>In practice, I guess most implementations would actually implement bool as
>an unsigned char, with bool++ being changed to bool=true, and bool-- to
>bool=false;

If bool++ means bool=true then what about bool+=i?  Where i is zero?
(MHO: Should be illegal.  Also IMHO bool++ should be illegal - if you
mean bool=true, damn it, spend the three extra keystrokes and say it
that way.)

--
Richard Krehbiel, Kastle Systems, Arlington VA USA
rich@kastle.com (work) or richk@cais.com (personal)