James Kuyper wrote:
>
> scott douglass wrote:
> >
> > Hello,
> >
> > I was surprised not to find const char* overloadings of strtod, strtol and
> > strtoul to replace the "mixed constness" versions in C, similarly to the way
> > that strstr has been overloaded, e.g.:
> >
> > double strtod(char *, char **);
> > double int strtod(const char *, const char **);
> > long strtol(char *, char **, int);
> > long strtol(const char *, const char **, int);
> > unsigned long strtoul(char *, char **, int);
> > unsigned long strtoul(const char *, const char **, int);
> >
> > Is there a way (that is more official than posting here) to suggest they be
> > added for next time?
> >
> > Or, is there a problem with doing this that I'm overlooking?
>
> It's not that there's a problem; it's that your suggested change isn't
> needed. The functions you list are already const on all of their pointer
> arguments, so they already be used with either const or non-const
> pointer arguements.
>
> The functions that were overloaded are the ones that return a pointer.

I disagree.  In C these functions do "return" a non-const pointer formed from
their
first argument.  They "return" it by assigning through their second parameter.
This
is what I meant by "mixed constness".
The C prototypes (from stdlib.h) are:

   double int strtod(const char *, char **);
   long strtol(const char *, char **, int);
   unsigned long strtoul(const char *, char **, int);
---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

scott douglass wrote:
>
> James Kuyper wrote:
....
> > The functions that were overloaded are the ones that return a pointer.
>
> I disagree.  In C these functions do "return" a non-const pointer formed from
> their
> first argument.  They "return" it by assigning through their second parameter.
> This
> is what I meant by "mixed constness".
> The C prototypes (from stdlib.h) are:
>
>    double int strtod(const char *, char **);
>    long strtol(const char *, char **, int);
>    unsigned long strtoul(const char *, char **, int);

Sorry - you're right. The definitions of those functions aren't in the
C++ standard itself, so I hadn't bothered to look them up. Now that I've
checked their definitions, I agree that overloaded versions with the
second argument also declared 'const char **' would be useful.
---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

Hello,

I was surprised not to find const char* overloadings of strtod, strtol and
strtoul to replace the "mixed constness" versions in C, similarly to the way
that strstr has been overloaded, e.g.:

double strtod(char *, char **);
double int strtod(const char *, const char **);
long strtol(char *, char **, int);
long strtol(const char *, const char **, int);
unsigned long strtoul(char *, char **, int);
unsigned long strtoul(const char *, const char **, int);

Is there a way (that is more official than posting here) to suggest they be
added for next time?

Or, is there a problem with doing this that I'm overlooking?
---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]

scott douglass wrote:
>
> Hello,
>
> I was surprised not to find const char* overloadings of strtod, strtol and
> strtoul to replace the "mixed constness" versions in C, similarly to the way
> that strstr has been overloaded, e.g.:
>
> double strtod(char *, char **);
> double int strtod(const char *, const char **);
> long strtol(char *, char **, int);
> long strtol(const char *, const char **, int);
> unsigned long strtoul(char *, char **, int);
> unsigned long strtoul(const char *, const char **, int);
>
> Is there a way (that is more official than posting here) to suggest they be
> added for next time?
>
> Or, is there a problem with doing this that I'm overlooking?

It's not that there's a problem; it's that your suggested change isn't
needed. The functions you list are already const on all of their pointer
arguments, so they already be used with either const or non-const
pointer arguements.

The functions that were overloaded are the ones that return a pointer.
You sometimes need to get a non-const pointer returned from those
functions. Two versions are provided, one returning const and the other
non-const. You're not allowed to overload on return type, so those two
functions are given different signatures by having the one which returns
a const pointer, be the one taking a const pointer argument. This
usually works out fairly well, because in the case where you need a
non-const pointer returned, that pointer usually points into the same
object as the corresponding non-const pointer arguement.
---
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@ncar.ucar.edu    ]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://reality.sgi.com/austern_mti/std-c++/faq.html              ]