Re: strlen man-page misinformation

Re: strlen man-page misinformation

[Michael Kerrisk (man-pages)]

Hello Alan,

On 02/03/2016 06:25 PM, Alan Aversa wrote:
> Hello,
> 
> The 2015-08-08 strlen man-page is incorrect. Here's a diff:
> 
>     --- a/man3/strlen.3
>     +++ b/man3/strlen.3
>     @@ -45,7 +45,7 @@ excluding the terminating null byte (\(aq\\0\(aq).
>      .SH RETURN VALUE
>      The
>      .BR strlen ()
>     -function returns the number of bytes in the string
>     +function returns the number of *characters* in the string that
>     precede the terminating null character

I went for a simpler change: s/bytes/characters/

>      .IR s .
>      .SH ATTRIBUTES
>      For an explanation of the terms used in this section, see
>     @@ -60,7 +60,7 @@ T{
>      T}     Thread safety   MT-Safe
>      .TE
>      .SH CONFORMING TO
>     -POSIX.1-2001, POSIX.1-2008, C89, C99, SVr4, 4.3BSD.
>     +POSIX.1-2001, POSIX.1-2008, C89, C99, C11, SVr4, 4.3BSD.

Fixed.

>      .SH SEE ALSO
>      .BR string (3),
>      .BR strnlen (3),
> 
> Page 392 (PDF p. 390, §7.24.6.3) of the C11 standard
> <http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1570.pdf> says:
> 
>     The *strlen* function returns the number of characters that precede
>     the terminating null character.

Thanks for the report. Interesting, POSIX.1 still uses the term "bytes"
the spec.

Cheers,

Michael


-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: strlen man-page misinformation

[walter harms]

Am 18.02.2016 14:12, schrieb Michael Kerrisk (man-pages):
> Hello Alan,
> 
> On 02/03/2016 06:25 PM, Alan Aversa wrote:
>> Hello,
>>
>> The 2015-08-08 strlen man-page is incorrect. Here's a diff:
>>
>>     --- a/man3/strlen.3
>>     +++ b/man3/strlen.3
>>     @@ -45,7 +45,7 @@ excluding the terminating null byte (\(aq\\0\(aq).
>>      .SH RETURN VALUE
>>      The
>>      .BR strlen ()
>>     -function returns the number of bytes in the string
>>     +function returns the number of *characters* in the string that
>>     precede the terminating null character
> 
> I went for a simpler change: s/bytes/characters/


For my understanding this is wrong. 1 character may be represented by 2 or more bytes (utf8).
see this example, the string (test) is 3 characters long and takes 6 bytes space.

did i miss something ? did the specification of character change ?

re,
 wh

#include <stdio.h>
#include <string.h>

int main()
{
        char *test="ÖÄÜ";
        int i;
        int len=strlen(test);
        printf("strlen=%d\n",len);

        for(i=0;i<len;i++)
                printf("%02x\n",(unsigned char)*(test+i));

        return 0;
}



output:
strlen=6
c3
96
c3
84
c3
9c



> 
>>      .IR s .
>>      .SH ATTRIBUTES
>>      For an explanation of the terms used in this section, see
>>     @@ -60,7 +60,7 @@ T{
>>      T}     Thread safety   MT-Safe
>>      .TE
>>      .SH CONFORMING TO
>>     -POSIX.1-2001, POSIX.1-2008, C89, C99, SVr4, 4.3BSD.
>>     +POSIX.1-2001, POSIX.1-2008, C89, C99, C11, SVr4, 4.3BSD.
> 
> Fixed.
> 
>>      .SH SEE ALSO
>>      .BR string (3),
>>      .BR strnlen (3),
>>
>> Page 392 (PDF p. 390, §7.24.6.3) of the C11 standard
>> <http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1570.pdf> says:
>>
>>     The *strlen* function returns the number of characters that precede
>>     the terminating null character.
> 
> Thanks for the report. Interesting, POSIX.1 still uses the term "bytes"
> the spec.
> 
> Cheers,
> 
> Michael
> 
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: strlen man-page misinformation

[Keith Thompson]

On Thu, Feb 18, 2016 at 7:42 AM, walter harms <wharms-fPG8STNUNVg@public.gmane.org> wrote:
>
>
> Am 18.02.2016 14:12, schrieb Michael Kerrisk (man-pages):
>> Hello Alan,
>>
>> On 02/03/2016 06:25 PM, Alan Aversa wrote:
>>> Hello,
>>>
>>> The 2015-08-08 strlen man-page is incorrect. Here's a diff:
>>>
>>>     --- a/man3/strlen.3
>>>     +++ b/man3/strlen.3
>>>     @@ -45,7 +45,7 @@ excluding the terminating null byte (\(aq\\0\(aq).
>>>      .SH RETURN VALUE
>>>      The
>>>      .BR strlen ()
>>>     -function returns the number of bytes in the string
>>>     +function returns the number of *characters* in the string that
>>>     precede the terminating null character
>>
>> I went for a simpler change: s/bytes/characters/
>
>
> For my understanding this is wrong. 1 character may be represented by 2 or more bytes (utf8).
> see this example, the string (test) is 3 characters long and takes 6 bytes space.
>
> did i miss something ? did the specification of character change ?

[...]

Either "bytes" or "characters" would be correct. POSIX says "bytes";
ISO C says "characters".

See the definition of "character" in C11 3.7.1:

>  bit representation that fits in a byte

On the other hand, 3.7 defines an (abstract) "character" as:

> member of a set of elements used for the organization, control,
> or representation of data

It also defines the terms "multibyte character" (a sequence of one
or more bytes representing a member of the extended character set)
and "wide character" (a value of type wchar_t).

"Bytes" is less ambiguous, but "characters" matches the wording of the
ISO C standard (and, in that context, refers to single-byte characters).

-- 
Keith Thompson <Keith.S.Thompson-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html