On Thu, Aug 6, 2020, 22:37 Johannes Schaub <schaub.johannes@googlemail.com> wrote:

Am Do., 6. Aug. 2020 um 22:11 Uhr schrieb Gennaro Prota <gennaro.prota@gmail.com>:
On Thu, Aug 6, 2020, 21:59 Johannes Schaub via Std-Discussion <std-discussion@lists.isocpp.org> wrote:
I made an issue report years ago about this, and the discussion unfortunately ended up being about whether 'false' is a valid null pointer constant, rather than more generally whether the set of values of bool actually _is_ {0, 1}. http://www.open-std.org/jtc1/sc22/wg21/docs/cwg_closed.html#1448 .

To me, the text of your report seems to be focused on that. It's not like it "ended up" there.

It's classified as a closed issue report, because "The resolution of issue 903 makes it clear that false is not a null pointer constant." . But my initial report was also focused on [...]

Well, it's not that important, but the text I see is:

Although 6.9.1 [basic.fundamental] paragraph 7 classifies bool as an integral type, the values of true and false are not specified — only that the results of converting them to another integral type are 1 and 0, respectively. This omission leaves unspecified whether false is an integral null pointer constant or not.

And that seems to focus on whether false is a null pointer constant. If the text of your original report was different, I don't know.

P.S.: sorry for any bad formatting... It's a pain from the phone :-(

.:: Gennaro Prota ::.
.:: https://about.me/gennaro.prota ::.