strtol const-ness problem

[dstevel]

The signature for strtol is:

strtol( const char*, char**, int)

So.. if we start with a passed "const char*" (pointer to const char),
then we can't create a non-const char pointer pointer to that const
char pointer as in:

void func( const char* a )
{
// ERROR: invalid conversion from `const char**' to `char**'
char** pa = &a;
int i = strtol( a, pa, 10 );
}

And here is another way of saying the same thing without temporary
variables.

void func( const char* a )
{
// ERROR: invalid conversion from `const char**' to `char**'
// ERROR: initializing argument 2 of `long int strtol...
int i = strtol( a, &a, 10 );

}

But you CAN do:

void func( const char* a )
{
int i = strtol( a, (char**)&a, 10 ); // OK
}

AND you can also do this:

void func( const char* a )
{
char* tmp;
int i = strtol( a, &tmp, 10 ); // OK
a = tmp;
}

This indirectly allows us to modify the original const char* a through
the new pointer tmp since tmp will point into the character array a. It
doesn't involve a hard cast, but it seems just as dangerous or even
more so because it's not obvious what just happened. The calling
function could see a change in the string pointed to by a, even though
it's passed as pointer to const.

Can someone help me figure out why this is OK?

[Heinz Ozwirk]

"dstevel" <go****@lakepage.comschrieb im Newsbeitrag
news:11**********************@m73g2000cwd.googlegr oups.com...

The signature for strtol is:

strtol( const char*, char**, int)

So.. if we start with a passed "const char*" (pointer to const char),
then we can't create a non-const char pointer pointer to that const
char pointer ...

In my opinion, this to be a bug in the standard. For many other functions
inherited from C, the standard has defined two functions with const and
non-const arguements, but strtol seems to be missing in that group. There
should better be two different overloads:

long strtol(char*, char**, int)
long strtol(char const*, char const**, long)

Perhaps you should post your message to comp.std.c++.

[...]

Can someone help me figure out why this is OK?

I'm curious to see if anyone could...

Heinz

[wittempj]

not a direct answer, but I would consider what the faq says about
converting a string to a number:
http://www.parashift.com/c++-faq-lit....html#faq-39.3

dstevel wrote:

The signature for strtol is:

strtol( const char*, char**, int)

So.. if we start with a passed "const char*" (pointer to const char),
then we can't create a non-const char pointer pointer to that const
char pointer as in:

void func( const char* a )
{
// ERROR: invalid conversion from `const char**' to `char**'
char** pa = &a;
int i = strtol( a, pa, 10 );
}

And here is another way of saying the same thing without temporary
variables.

void func( const char* a )
{
// ERROR: invalid conversion from `const char**' to `char**'
// ERROR: initializing argument 2 of `long int strtol...
int i = strtol( a, &a, 10 );

}

But you CAN do:

void func( const char* a )
{
int i = strtol( a, (char**)&a, 10 ); // OK
}

AND you can also do this:

void func( const char* a )
{
char* tmp;
int i = strtol( a, &tmp, 10 ); // OK
a = tmp;
}

This indirectly allows us to modify the original const char* a through
the new pointer tmp since tmp will point into the character array a. It
doesn't involve a hard cast, but it seems just as dangerous or even
more so because it's not obvious what just happened. The calling
function could see a change in the string pointed to by a, even though
it's passed as pointer to const.

Can someone help me figure out why this is OK?

[dstevel]

Thanks for the link. The nice part about strtol compared with all of
those convert functions is that strtol only converts the number at the
first portion of the string, then returnsa pointer to the rest of the
string in the char** you provide to it.

I'm using it to pick out the numbers in a long string that contains a
mix of letters and numbers like:

"ABC1234/CDEF44/XYZ/275B"

and my program picks out 1234, 44, 275

It goes through a string 1 character at a time, and if it finds a digit
it called strtol on the current character to get the full number. Then
it continues examining the non-digits at the point where strtol
indicates the number ended. I don't know that any of those convert
functions do that. They seem to expect the full string to be a valid
number like "12345" = 12345.


wi******@hotmail.com wrote:

not a direct answer, but I would consider what the faq says about
converting a string to a number:
http://www.parashift.com/c++-faq-lit....html#faq-39.3

dstevel wrote:

The signature for strtol is:

strtol( const char*, char**, int)

So.. if we start with a passed "const char*" (pointer to const char),
then we can't create a non-const char pointer pointer to that const
char pointer as in:

void func( const char* a )
{
// ERROR: invalid conversion from `const char**' to `char**'
char** pa = &a;
int i = strtol( a, pa, 10 );
}

And here is another way of saying the same thing without temporary
variables.

void func( const char* a )
{
// ERROR: invalid conversion from `const char**' to `char**'
// ERROR: initializing argument 2 of `long int strtol...
int i = strtol( a, &a, 10 );

}

But you CAN do:

void func( const char* a )
{
int i = strtol( a, (char**)&a, 10 ); // OK
}

AND you can also do this:

void func( const char* a )
{
char* tmp;
int i = strtol( a, &tmp, 10 ); // OK
a = tmp;
}

This indirectly allows us to modify the original const char* a through
the new pointer tmp since tmp will point into the character array a. It
doesn't involve a hard cast, but it seems just as dangerous or even
more so because it's not obvious what just happened. The calling
function could see a change in the string pointed to by a, even though
it's passed as pointer to const.

Can someone help me figure out why this is OK?