Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c++ > #5179 > unrolled thread

'5'-'0' = 5 alway? guaranteed?

Back to article view | Back to comp.lang.c++

Contents

  '5'-'0' = 5 alway? guaranteed? "crea" <no@invalid.com> - 2011-05-19 20:13 +0100
    Re: '5'-'0' = 5 alway? guaranteed? "Alf P. Steinbach /Usenet" <alf.p.steinbach+usenet@gmail.com> - 2011-05-19 21:28 +0200
      Re: '5'-'0' = 5 alway? guaranteed? "crea" <no@invalid.com> - 2011-05-19 20:35 +0100
        Re: '5'-'0' = 5 alway? guaranteed? Noah Roberts <dont@email.me> - 2011-05-19 12:58 -0700
          Re: '5'-'0' = 5 alway? guaranteed? Victor Bazarov <v.bazarov@comcast.invalid> - 2011-05-19 16:10 -0400
          Re: '5'-'0' = 5 alway? guaranteed? Pete Becker <pete@versatilecoding.com> - 2011-05-19 16:21 -0400
            Re: '5'-'0' = 5 alway? guaranteed? "crea" <no@invalid.com> - 2011-05-19 21:24 +0100
          Re: '5'-'0' = 5 alway? guaranteed? "Alf P. Steinbach /Usenet" <alf.p.steinbach+usenet@gmail.com> - 2011-05-20 00:39 +0200

#5179 — '5'-'0' = 5 alway? guaranteed?

I am just programming this kind of thing, and was wondering I according to 
c++ language it is always guaranteed that
'5'-'0' = 5. Character calculations. Because character should be in order 
isnt it? Is it guaranteed that they are in order and the difference is 1?

[toc] | [next] | [standalone]

#5180

* crea, on 19.05.2011 21:13:
> I am just programming this kind of thing, and was wondering I according to
> c++ language it is always guaranteed that
> '5'-'0' = 5. Character calculations. Because character should be in order
> isnt it? Is it guaranteed that they are in order and the difference is 1?

Yes.

But that does not hold for letters (which are discontiguous in EBCDIC, which the 
C++ standard "must" support).


Cheers & hth.,

- Alf

-- 
blog at <url: http://alfps.wordpress.com>

[toc] | [prev] | [next] | [standalone]

#5181

"Alf P. Steinbach /Usenet" <alf.p.steinbach+usenet@gmail.com> wrote in 
message news:ir3qlq$a84$1@dont-email.me...
>* crea, on 19.05.2011 21:13:
>> I am just programming this kind of thing, and was wondering I according 
>> to
>> c++ language it is always guaranteed that
>> '5'-'0' = 5. Character calculations. Because character should be in order
>> isnt it? Is it guaranteed that they are in order and the difference is 1?
>
> Yes.
>
> But that does not hold for letters (which are discontiguous in EBCDIC, 
> which the C++ standard "must" support).
>

ok, good to know also the letter-thing. 

[toc] | [prev] | [next] | [standalone]

#5182

On 5/19/2011 12:35 PM, crea wrote:
> "Alf P. Steinbach /Usenet"<alf.p.steinbach+usenet@gmail.com>  wrote in
> message news:ir3qlq$a84$1@dont-email.me...
>> * crea, on 19.05.2011 21:13:
>>> I am just programming this kind of thing, and was wondering I according
>>> to
>>> c++ language it is always guaranteed that
>>> '5'-'0' = 5. Character calculations. Because character should be in order
>>> isnt it? Is it guaranteed that they are in order and the difference is 1?
>>
>> Yes.
>>
>> But that does not hold for letters (which are discontiguous in EBCDIC,
>> which the C++ standard "must" support).
>>
>
> ok, good to know also the letter-thing.
>
>

He's wrong actually.  It's a really good bet, but by no means guaranteed.

-- 
http://crazycpp.wordpress.com

[toc] | [prev] | [next] | [standalone]

#5184

On 5/19/2011 3:58 PM, Noah Roberts wrote:
> On 5/19/2011 12:35 PM, crea wrote:
>> "Alf P. Steinbach /Usenet"<alf.p.steinbach+usenet@gmail.com> wrote in
>> message news:ir3qlq$a84$1@dont-email.me...
>>> * crea, on 19.05.2011 21:13:
>>>> I am just programming this kind of thing, and was wondering I according
>>>> to
>>>> c++ language it is always guaranteed that
>>>> '5'-'0' = 5. Character calculations. Because character should be in
>>>> order
>>>> isnt it? Is it guaranteed that they are in order and the difference
>>>> is 1?
>>>
>>> Yes.
>>>
>>> But that does not hold for letters (which are discontiguous in EBCDIC,
>>> which the C++ standard "must" support).
>>>
>>
>> ok, good to know also the letter-thing.
>>
>>
>
> He's wrong actually. It's a really good bet, but by no means guaranteed.

No, he is not.  Read [lex.charset]/3.  Carefully.

V
-- 
I do not respond to top-posted replies, please don't ask

[toc] | [prev] | [next] | [standalone]

#5185

On 2011-05-19 15:58:29 -0400, Noah Roberts said:

> On 5/19/2011 12:35 PM, crea wrote:
>> "Alf P. Steinbach /Usenet"<alf.p.steinbach+usenet@gmail.com>  wrote in
>> message news:ir3qlq$a84$1@dont-email.me...
>>> * crea, on 19.05.2011 21:13:
>>>> I am just programming this kind of thing, and was wondering I according
>>>> to
>>>> c++ language it is always guaranteed that
>>>> '5'-'0' = 5. Character calculations. Because character should be in order
>>>> isnt it? Is it guaranteed that they are in order and the difference is 1?
>>> 
>>> Yes.
>>> 
>>> But that does not hold for letters (which are discontiguous in EBCDIC,
>>> which the C++ standard "must" support).
>>> 
>> 
>> ok, good to know also the letter-thing.
>> 
>> 
> 
> He's wrong actually.  It's a really good bet, but by no means guaranteed.

The character representations of the digits 0-9 are required to be 
contiguous and in ascending order. And there's no similar requirement 
for any other characters.

-- 
  Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com) Author of "The 
Standard C++ Library Extensions: a Tutorial and Reference 
(www.petebecker.com/tr1book)

[toc] | [prev] | [next] | [standalone]

#5186

"Pete Becker" <pete@versatilecoding.com> wrote in message 
news:2011051916213466006-pete@versatilecodingcom...
> On 2011-05-19 15:58:29 -0400, Noah Roberts said:
>>>
>>> ok, good to know also the letter-thing.
>>>
>>>
>>
>> He's wrong actually.  It's a really good bet, but by no means guaranteed.
>
> The character representations of the digits 0-9 are required to be 
> contiguous and in ascending order. And there's no similar requirement for 
> any other characters.
>

Thanks guys, good info which would be difficult to find from books (well, 
maybe google would help but this is not something which is told in schools 
or books) - yes, they use it sometimes, but I never read that its 
"guaranteed to work" which I really need to know. 

[toc] | [prev] | [next] | [standalone]

#5191

* Stefan Ram, on 19.05.2011 22:09:
> Noah Roberts<dont@email.me>  writes:
>> He's wrong actually.  It's a really good bet, but by no means guaranteed.
>
>    Then how for any digit character c, the expression
>    (do_narrow(c,dfault)-'0') evaluates to the digit value of
>    the character? (according to ISO/IEC 14882:2003(E)
>    22.2.1.1.2 char do_narrow(charT c, char dfault) const,
>    Effects)

Noah is in my killfile so I don't see his article. And he knows that. And 
needless to say, Noah's public insinuation that I was guessing, knowing full 
well that I don't see his article, is just what one can expect from him.

If you want to check out the formal it's at the very start of the standard 
somewhere .

You can download the latest draft (of the next standard) from the committee pages.


Cheers & hth.,

- Alf

-- 
blog at <url: http://alfps.wordpress.com>

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.c++

csiph-web