Re: assignment of bigger type to smaller type
Ivan Novick wrote:
Jack Klein wrote:
On 5 Jan 2007 15:43:18 -0500, "Ivan Novick" <ivan@0x4849.net> wrote in
comp.lang.c++.moderated:
#include <iostream>
int main()
{
int x = LONG_MAX;
std::cout << x << std::endl;
short y = x;
std::cout << y << std::endl;
return 0;
}
Why would the C++ standard allow this code without requiring an error
message? Surely assigning a larger type to a smaller type could be
caught at compile time and allowing assignments like this is inherintly
dangerous?
Why shouldn't the C++ standard allow this code? It is perfectly
legal, and is in fact quite common in certain types of coding.
Of interest, after posting this question, I read in Stroustrup's
"Design and evolution of C++", that in an early version of C++ he
required an explicit cast for this:
long lng;
....
int i1 = (int) lng;
He then goes on to say this failed miserably because every C program he
looked at had large numbers of these.
I'm not sure about this exact example, but things like:
char* dest ;
int ch = fgetc( file ) ;
while ( ch != EOF && ch != '\n' ) {
*p ++ = ch ;
}
is a standard idiom in C. Functions like fgetc (and
istream::get() in C++) return an int, not a char, but the int is
guaranteed to be either EOF or in the range 0...UCHAR_MAX.
While this is definitly a design flaw in C, it's far too well
established to change now. (Note that the above bit of code
actually contains implementation defined behavior which can
cause an implementation defined signal in C. In sum, it isn't
really legal C, or at least not portable legal C. On the other
hand, it's so wide spread that no C compiler would dare break
it.)
--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orient?e objet/
Beratung in objektorientierter Datenverarbeitung
9 place S?mard, 78210 St.-Cyr-l'?cole, France, +33 (0)1 30 23 00 34
--
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]