Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: Motion P1788/M0051:IntervalLiteralsText -- discussion period begins



> > A minor change to the text was made by John Pryce:
>
> Those two lines were identical.

I'm sorry. The difference was in the second paragraph of 12.11.2 .
Motion51Oct20.pdf:
The following forms of number literal shall be supported.
Motion51Oct22.pdf:
The following forms of number literal shall be provided.

> There *is* a problem however, which I think I pointed out in the past:
> In 12.11.2 (b), hexadecimal form, it states that the exponent field is
> optional.  I don't know about C99, but in 754-2008 it is *required*.

I'm ready to make hexadecimal exponent field *required*.
Does anybody have objections ?

Baker. May we have a 1-2 day pause before the voting ?

  -Dima

----- Исходное сообщение -----
От: mhack@xxxxxxx
Кому: stds-1788@xxxxxxxxxxxxxxxxx
Отправленные: Вторник, 22 Октябрь 2013 г 23:40:07 GMT +04:00 Абу-Даби, Маскат
Тема: Re: Motion P1788/M0051:IntervalLiteralsText -- discussion period begins

Dmitry Nadezhin wrote:
> A minor change to the text was made by John Pryce:

Those two lines were identical.

Although the two .pdf files were different in binary, they look alike
on paper, at least on quick inspection.

There *is* a problem however, which I think I pointed out in the past:
In 12.11.2 (b), hexadecimal form, it states that the exponent field is
optional.  I don't know about C99, but in 754-2008 it is *required*.

Well, atof() does accept hex literals without an exponent, so perhaps
it is ok -- but be careful when mentioning 754-2008 here.

Michel.

---Sent: 2013-10-22 19:34:29 UTC