Reference to void
in [C++]
|
From: James Kanze on 30 Oct 2006 11:36 Frederick Gotham wrote: > Greg Herlihy: > int i; > const int* p = &i; > Is "p" not a "pointer to const" because i is non-const? Please. The real problem is elsewhere. If p is a const pointer above, then what is p in: int *const p = &i ; ? Regretfully, common usage is that the first p is a const pointer (and that nobody has heard of the second). Most of the time, the context makes what is really meant clear, and we live with it (since neither you nor I have enough influence to change general use). When we do need to distinguish, or if there is any chance of ambiguity, of course, the first is pointer to const, and the second is const pointer. But in my experience, such cases are rare. Logical use of English would of course apply the same distinction to references (except that there would be no non-const references). But "logical use" and "English" (or any other natural language) probably form an antilogy. And while we should certainly strive to do the best we can, the final goal is communication, and not some logical ideal. (Note that I'm not arguing that it should be this way. Only that it is.) -- James Kanze (GABI Software) email:james.kanze(a)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! ]
From: Greg Herlihy on 30 Oct 2006 13:17 Frederick Gotham wrote: > Greg Herlihy: > > > A const reference is no misnomer: > > > > int i; > > const int& iRef = i; > > > What you have there is a "reference to const" -- a reference cannot be > declared as const. The following is a "reference to const": const int i = 3; const int& iRefConst = i; and differs from the original example significantly: namely iRefConst has an immutable value whereas iRef does not. > > iRef is clearly a "const reference" - and not a "reference to const" > > because the object iRef refers to, i, is not const. > > > int i; > > const int* p = &i; > > Is "p" not a "pointer to const" because i is non-const? Please. Does 2/3 equal 3/2 because 2 * 3 equals 3 * 2 ? No, because using multiplication to demonstrate a point about division is just as effective as using pointers to prove a point about references. The qualifier 'const' whether it is applied to a pointer or a reference variable means that that variable cannot be used to assign a new value to its object. So a "const reference" is a reference variable that cannot be assigned a new value (because "const" appeared in its declaration). A "reference to const" is therefore a particular kind of const reference - one whose aliased object has also been declared const. So stated simply: a "reference to const" variable has an immutable value while a "const reference" variable may or may not have an immutable value. The latest draft of the C++ Standard uses "const reference" or "non-const reference" a combined total of nine times, and "reference to const" just twice. So there is scant support in the C++ Standard for the view that "const reference" is some kind of a misnomer. Greg -- [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ]
From: Frederick Gotham on 30 Oct 2006 13:18 James Kanze: >> int i; > >> const int* p = &i; > >> Is "p" not a "pointer to const" because i is non-const? Please. > > The real problem is elsewhere. If p is a const pointer above, > then what is p in: > > int *const p = &i ; > > ? A const pointer: T *const p; A pointer to const: T const *p; A const pointer to const: T const *const p; -- Frederick Gotham [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ]
From: Gerhard Menzl on 31 Oct 2006 08:55 Greg Herlihy wrote: > The following is a "reference to const": > > const int i = 3; > const int& iRefConst = i; > > and differs from the original example significantly: namely iRefConst > has an immutable value whereas iRef does not. Then what is ref in void f(int const& ref); int my_rand(); void g() { int i = 1; int const ci = 2; f(my_rand() % 2 ? i : ci); } - a reference to const or a reference to non-const? You cannot tell, as it depends on runtime conditions. Since this is about terminology for static types, your distinction is meaningless. > Does 2/3 equal 3/2 because 2 * 3 equals 3 * 2 ? No, because using > multiplication to demonstrate a point about division is just as > effective as using pointers to prove a point about references. Just as a point about arithmetics cannot be used to prove a point about indirection. -- Gerhard Menzl Non-spammers may respond to my email address, which is composed of my full name, separated by a dot, followed by at, followed by "fwz", followed by a dot, followed by "aero". [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ]
From: Frederick Gotham on 31 Oct 2006 11:08
Gerhard Menzl: > > Then what is ref in > > void f(int const& ref); That function parameter is a reference to a const int. > int my_rand(); > > void g() > { > int i = 1; > int const ci = 2; > > f(my_rand() % 2 ? i : ci); When you say that something is a "reference to const" or a "pointer to const", you're saying that the pointer cannot be used to alter the target object. You're _not_saying that the target option is const. .....but you already knew that, right? Your example is no different to: int a; int const b; int const *const p = my_rand() % 2 ? &a : &b; > } > > - a reference to const or a reference to non-const? You cannot tell, as > it depends on runtime conditions. No, it's a "reference to const", because there's no such thing as a const reference. There's a such thing as a const pointer: int *const p; , but not a const reference: int &const r; /* Compile ERROR */ -- Frederick Gotham [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] |