C++Talk.NET C++ language newsgroups   Archives   FAQ   Search   Memberlist   Usergroups   Register   Profile   Log in to check your private messages   Log in  Reverse Iterator does not compile          C++Talk.NET Forum Index -> C++ Language (Moderated) View previous topic :: View next topic   Author Message Pramod Guest Posted: Sat Aug 14, 2004 6:30 pm    Post subject: Reverse Iterator does not compile I was writing my own reverse iterator. Here's the code. I have provided the test cases in main funtion. Strangely the "copy" operation does not compile if I use vector. For list it compiles. Any hints on why? #include <iostream> #include <iterator> #include <list> #include <vector> using namespace std; template <class Iter> class my_reverse_iterator:public iterator < typename iterator_traits typename iterator_traits<Iter>::value_type, typename iterator_traits<Iter>::difference_type, typename iterator_traits<Iter>::pointer, typename iterator_traits<Iter>::reference Quote: { Iter current; public: my_reverse_iterator(Iter x):current(x) {} Iter base() { return current; } reference operator*() { Iter tmp = current; --tmp; return *tmp; } pointer operator->() { return current; } my_reverse_iterator& operator++() { --current; return *this; } my_reverse_iterator operator++(int) { Iter tmp = current; --current; return tmp; } my_reverse_iterator& operator--() { ++current; return *this; } my_reverse_iterator operator--(int) { Iter tmp = current; ++current; return tmp;} bool operator== (my_reverse_iterator& x) { return current==x.current; } bool operator!= (my_reverse_iterator& x) { return current!=x.current; } bool operator<= (my_reverse_iterator& x) { return current<=x.current; } bool operator>= (my_reverse_iterator& x) { return current>=x.current; } bool operator< (my_reverse_iterator& x) { return current bool operator> (my_reverse_iterator& x) { return current>x.current; } reference operator[] (difference_type n) { return advance(current,-n); } my_reverse_iterator operator+ (difference_type n) { Iter tmp = current-n; return tmp; } my_reverse_iterator operator- (difference_type n) { Iter tmp = current+n; return tmp; } my_reverse_iterator& operator+=(difference_type n) { current-=n; return *this; } my_reverse_iterator& operator-=(difference_type n) { current+=n; return *this; } }; int main() { list<int> li; copy( istream_iterator<int>(cin),istream_iterator<int>(),back_inserter(li)); my_reverse_iterator<list rlb(li.end()); my_reverse_iterator<list rle(li.begin()); copy( rlb, rle, ostream_iterator<int>(cout," ")); cout << endl; vector copy( li.begin(), li.end(), back_inserter(vi)); my_reverse_iterator<vector rvb(vi.end()); my_reverse_iterator<vector rve(vi.begin()); copy( rvb, rve, ostream_iterator<int>(cout," ")); //does not compile for( my_reverse_iterator<vector tmp = rvb; tmp != rve; ++tmp ) cout << *tmp << " "; //compiles cout << endl; } Is it the way the iterators are defined for vector and list making any difference here ? Thanks -Pramod [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top John Potter Guest Posted: Sun Aug 15, 2004 10:58 am    Post subject: Re: Reverse Iterator does not compile On 14 Aug 2004 14:30:33 -0400, [email]pramod_iitg (AT) yahoo (DOT) com[/email] (Pramod) wrote: Quote: I was writing my own reverse iterator. Here's the code. I have provided the test cases in main funtion. Strangely the "copy" operation does not compile if I use vector. For list it compiles. Any hints on why? Vector has a random_access_iterator which supports operator- for two iterators and you did not supply one. List has a bidirectional_iterator which does not support the missing operation. The copy algorithm detects the type of iterator and uses a "more efficient" version for random_access_iterators. Quote: reference operator[] (difference_type n) { return advance(current,-n); } This will not compile. Advanace has a void return. If it did compile, you would modify the iterator. There is no need for advance since this operation is only defined for random access. return *(*this + n); Quote: my_reverse_iterator operator+ (difference_type n) { Iter tmp = current-n; return tmp; } my_reverse_iterator operator- (difference_type n) { Iter tmp = current+n; return tmp; } difference_type operator- (my_reverse_iterator rhs) { return rhs.current - current; } There may be many other undetected problems. John [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Pramod Guest Posted: Tue Aug 17, 2004 7:07 pm    Post subject: Re: Reverse Iterator does not compile Talking about vectors and other containers. Is it a good idea to return containers from functions. Are containers allowed to have reference counts? won't returning them cause unnecessary copies to and from temporaries? In this case creating a dynamic array and returning that would be better right? IN what cases should dynamic array be preferred over arrays or vice-versa? [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top kanze@gabi-soft.fr Guest Posted: Wed Aug 18, 2004 6:18 pm    Post subject: Re: Reverse Iterator does not compile [email]pramod_iitg (AT) yahoo (DOT) com[/email] (Pramod) wrote in message news:<581b1e22.0408162349.161761dd (AT) posting (DOT) google.com>... Quote: Talking about vectors and other containers. Is it a good idea to return containers from functions. It depends. Quote: Are containers allowed to have reference counts? I presume you mean that they are implemented by some sort of COW. I don't think so: accessing an element in a vector (operator[]) is guaranteed to take place in constant time, and I don't think you can do this with COW. Quote: won't returning them cause unnecessary copies to and from temporaries? Returning a local variable always requires at least one copy. More, sometimes. Quote: In this case creating a dynamic array and returning that would be better right? I'm not sure what you mean by dynamci arrays. I certainly wouldn't use new[], and return the pointer returned by that. Depending on the context, you can often get by with returning smallish vectors. If performance is a consideration, the usual solution is to have the caller create the container, and to pass a non-const reference to it; if the called function is really small, and you can afford to have its implementation in a header file, an even better solution might simply be to template it on an output iterator, and pass it a back_inserter -- this leaves a maximum of flexibility to the caller. Quote: IN what cases should dynamic array be preferred over arrays or vice-versa? If by dynamic array, you mean something like new[], then I can't ever think of a case where it would be prefered over std::vector. -- James Kanze GABI Software http://www.gabi-soft.fr Conseils en informatique orientée objet/ Beratung in objektorientierter Datenverarbeitung 9 place Sémard, 78210 St.-Cyr-l'École, 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! ] Back to top David Abrahams Guest Posted: Fri Aug 20, 2004 1:29 am    Post subject: Re: Reverse Iterator does not compile [email]kanze (AT) gabi-soft (DOT) fr[/email] writes: Quote: pramod_iitg (AT) yahoo (DOT) com (Pramod) wrote in message news:<581b1e22.0408162349.161761dd (AT) posting (DOT) google.com>... Talking about vectors and other containers. Is it a good idea to return containers from functions. It depends. Are containers allowed to have reference counts? I presume you mean that they are implemented by some sort of COW. I don't think so: accessing an element in a vector (operator[]) is guaranteed to take place in constant time, and I don't think you can do this with COW. won't returning them cause unnecessary copies to and from temporaries? Returning a local variable always requires at least one copy. More, sometimes. Not true. The (Named) Return Value Optimization can construct the result directly into its location in the caller. It's also possible, through elaborate trickery, to force the compiler to automatically use move semantics for returns... but that's another story. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Display posts from previous: All Posts1 Day7 Days2 Weeks1 Month3 Months6 Months1 Year Oldest FirstNewest First         C++Talk.NET Forum Index -> C++ Language (Moderated) All times are GMT Page 1 of 1   Jump to: Select a forum C/C++----------------C LanguageC Language (Moderated)C++ Language (Moderated)C++ language (comp.lang.c++)C++ language, library and standards Non-English C/C++ Newsgroups----------------C++ (German)C++ (French)  You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum Powered by phpBB © 2001, 2006 phpBB Group SEO toolkit © 2004-2006 webmedic.