[Cfp-interest 2389] Re: WG14 IEEE 754-C binding meeting minutes 2022/02/10

[Jim Thomas]

> On Feb 10, 2022, at 5:48 PM, Fred J. Tydeman <tydeman at tybor.com> wrote:
> 
> On Fri, 11 Feb 2022 00:48:11 +0100 Vincent Lefevre wrote:
>> 
>> If the 'i' suffix is added, what about the imaginary infinities?
>> Will a new macro be added or is multiplying by 'I' sufficient?

Right. The i suffix couldn’t replace the I macro, which would still be needed for INFINITY*I and NAN*I. I suspect this is the reason for not proposing the suffix initially.
> 
> A problem is:  I is not required to be imaginary, it can be complex.
> If it is complex, then INFINITY * I ends up being 
> CMPLXF(NAN, INFINITY) 
> 
> If imaginary types are not supported, but the 'i' suffix is supported,
> I do not know what  INFINITY * 1.fi  would end up being.

I suppose the type domain determined by the suffix could depend on whether imaginary times are supported. Then, if imaginary types are not supported, 1.fi <http://1.fi/> would be the float complex number (0, 1).

I don’t see a WG14 proposal for this.

- Jim Thomas

> 
> ---
> Fred J. Tydeman        Tydeman Consulting
> tydeman at tybor.com      Testing, numerics, programming
> +1 (702) 608-6093      Vice-chair of PL22.11 (ANSI "C")
> Sample C99+FPCE tests: http://www.tybor.com
> Savers sleep well, investors eat well, spenders work forever.
> 
> _______________________________________________
> Cfp-interest mailing list
> Cfp-interest at oakapple.net
> http://mailman.oakapple.net/mailman/listinfo/cfp-interest

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.oakapple.net/pipermail/cfp-interest/attachments/20220303/61b6ab2a/attachment-0001.htm>