Re: [gmp 3.1.1] LONG_MAX and ULONG_MAX redefined

[Kevin Ryde]

> > In such a setup does fputc still write one byte?
> 
> The standard says that it writes a char, which is a byte *in the C
> standard terminology*. In a low-level point of view, it is 16 bits.

Hmm.  Pretty novel.

> > If not it might be a problem for the mpz_out_raw file formatting.
> 
> I don't think there is a problem. The file uses a 16-bit character set.
> I suppose that the user will use conversion tools when he needs to read
> the file on another machine or something like that.
> 
> I'll try to have more details about that...

Oh well, it's not important, unless there's something gmp should do.
I expect there'd be a lot of old code that might not enjoy 16-bit
chars, so two obscure gmp functions are the least of anyone's worries.