Re: "A diagram of C23 basic types"

On 08/04/2025 00:01, Keith Thompson wrote:

bart <bc@freeuk.com> writes:
[...]

Since numbers using exponents without also using decimal points are
rare in my code base, I've decided to experiment with numbers like 1e6
being integer constants rather that floats. (This is in my language.)

 You might want to look at Ada for existing practice.
 In C, a constant with either a decimal point or an exponent is
floating-point.  In Ada, 1.0e6 is floating-point and 1e6 is an integer.
Of course this isn't very helpful if you want to represent numbers with
a lot of non-zero digits; for that, you need digit separators.

The context here /is/ lots of zeros. In the real world, usage such as such as 'N million' or 'N billion' typically scales N by a million or billion.
But it's good to know I'm copying existing practice from Ada.
(However it turns out that trying to add it to my C compiler is pointless; '1e9' needs to be an integer billion across multiple compilers.)