[Cfp-interest] D/d double type floating point suffix

[Jim Thomas]

I wasn't on the C committee at the time, but did review the TR and argued, unsuccessfully, against the FLOAT_CONST_DECIMAL64 pragma. I don't see the d/D suffix as a mistake/oversight. It's a solution to the problem of how to write double constants that are protected against reinterpretation under the pragma. For example, the suffix lets you write a <float.h> header that doesn't break if the pragma happens to be in effect where the header is included. The suffix is used in this way in shipping headers.

For the CFP TS Part 2, it might be reasonable to 

1. deprecate both the pragma and the suffix, or
2. remove both the pragma and the suffix.

With 2, I think an implementation could continue to support the pragma and/or suffix as extensions and still be in compliance with TS Part 2. Is this correct?

-Jim

On Jul 26, 2012, at 4:35 AM, Rajan Bhakta wrote:

> Hi, 
> 
> I talked to Edison regarding the d/D suffix for the double generic type and he said it went into the Decimal TR by mistake/oversight as a blind acceptance of Fred's WG14 comment 10904 (the suffix is point 6) without review. He did say he didn't intend for it to happen after seeing it. 
> In essence he said we (IBM) screwed up and would want to rectify the issue by removing the suffix and wants to have this as a defect report. So I would like to keep my comments in the TS part 2 regarding removing the d/D suffix. 
> 
> Regards,
> 
> Rajan Bhakta
> z/OS XL C/C++ Compiler Technical Architect
> ISO C Standards Representative for Canada
> C Compiler Development
> Contact: rbhakta at ca.ibm.com, Rajan Bhakta/Toronto/IBM
> Telephone: (905) 413-3995_______________________________________________
> 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/20120726/907462bd/attachment.html