From: Edward Diener (eddielee_at_[hidden])
Date: 2006-01-21 21:32:42
Paul Mensonides wrote:
>> -----Original Message-----
>> From: boost-bounces_at_[hidden]
>> [mailto:boost-bounces_at_[hidden]] On Behalf Of Edward Diener
>
>> I have never understood the definition of a pp-number in the
>> C++ standard. I am using the 1998 version and I read:
>>
>> ppnumber:
>> digit
>> . digit
>> ppnumber digit
>> ppnumber nondigit
>> ppnumber e sign
>> ppnumber E sign
>> ppnumber .
>>
>> Since "ppnumber digit" and "ppnumber nondigit" covers all
>> sequences beginning either with a "digit" or a ".digit"
>> followed by anything else which is not a whitespace character
>> or a comment, why specify the last three items ?
>
> Despite the name, 'nondigit' is not any non-digit character. It is just the
> alphabetic characters, the underscore, and universal character names (subject to
> annex E).
Thanks, I see it now. The definition of non-digit ( and digit )
unfortunately was made not in Preprocessing Numbers section, where it
was first mentioned, but in the Identifiers section. Thanks for pointing
this out to me.