C++Talk.NET C++ language newsgroups   Archives   FAQ   Search   Memberlist   Usergroups   Register   Profile   Log in to check your private messages   Log in  iterator points to entry of vector          C++Talk.NET Forum Index -> C++ Language (Moderated) View previous topic :: View next topic   Author Message Stephen Vavasis Guest Posted: Fri Mar 25, 2005 11:01 am    Post subject: iterator points to entry of vector In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. Can someone confirm that the above statement is nonconforming? What is the preferred way to get the desired outcome? -- Steve Vavasis [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Francis Glassborow Guest Posted: Sat Mar 26, 2005 4:33 am    Post subject: Re: iterator points to entry of vector In article <d1uurt$kfp$1 (AT) news01 (DOT) cit.cornell.edu>, Stephen Vavasis <vavasis (AT) cs (DOT) cornell.edu> writes Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. I assume that u is a vector<T> then vector<T>::iterator i(u.begin() + j); would seem to do what you want. -- Francis Glassborow ACCU Author of 'You Can Do It!' see http://www.spellen.org/youcandoit For project ideas and contributions: http://www.spellen.org/youcandoit/projects [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Sascha Müller Guest Posted: Sat Mar 26, 2005 4:34 am    Post subject: Re: iterator points to entry of vector Stephen Vavasis wrote: Quote: vector<T>::iterator i = &u[j]; Can someone confirm that the above statement is nonconforming? If u is a vector<T>, then u[j] returns a T and has no information about the vector it came from. Quote: What is the preferred way to get the desired outcome? What about vector<int>::iterator i = u.begin() + j; Sascha Müller [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Torsten Robitzki Guest Posted: Sat Mar 26, 2005 4:35 am    Post subject: Re: iterator points to entry of vector Stephen Vavasis wrote: Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. Can someone confirm that the above statement is nonconforming? What is the preferred way to get the desired outcome? You've assumed that vector::iterator have to be a pointer or to be constrctable from a pointer. Just use u.begin() + j; for a vector or the more generale form std::advance(u.begin(), j) for any container type. regards Torsten [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Bo Persson Guest Posted: Sat Mar 26, 2005 4:48 am    Post subject: Re: iterator points to entry of vector "Stephen Vavasis" <vavasis (AT) cs (DOT) cornell.edu> skrev i meddelandet news:d1uurt$kfp$1 (AT) news01 (DOT) cit.cornell.edu... Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. Can someone confirm that the above statement is nonconforming? What is the preferred way to get the desired outcome? The code is not conforming, but just happens to work in implementations where the iterator is a typedef for T*. That is *one* possible way to implement vector, but not the only one. To get an iterator to an element j positions from the start of the vector, you can use v.begin() + j That is guaranteed to work, as long as j is convertible to vector<T>::size_type. :-) Bo Persson [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Antoun Kanawati Guest Posted: Sat Mar 26, 2005 9:31 am    Post subject: Re: iterator points to entry of vector Stephen Vavasis wrote: Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. &u[j] has the type T*, which may, or may not, be the same as vector<T>::iterator. Quote: At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. Can someone confirm that the above statement is nonconforming? What is the preferred way to get the desired outcome? vector<T>::iterator i = u.begin() + j; -- A. Kanawati [email]NO.antounk.SPAM (AT) comcast (DOT) net[/email] [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Randy Guest Posted: Sat Mar 26, 2005 9:32 am    Post subject: Re: iterator points to entry of vector Stephen Vavasis wrote: Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. Can someone confirm that the above statement is nonconforming? What is the preferred way to get the desired outcome? -- Steve Vavasis A vector iterator may, but is not required to, be a pointer, but &u[j] is by definition a pointer. Use iterators instead: vector<T>::iterator i = u.begin() + j; Now everybody's happy. :-) Randy. [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Nemra® Guest Posted: Sat Mar 26, 2005 9:42 am    Post subject: Re: iterator points to entry of vector Hello Stephen, This is a common problem. Usually library developers just typedef-ed the T* to be an iterator type, but in the latest realization it is a new practice to provide a random-access-type iteration classes, which in some cases (for example in release build) can still be replaced by pointers. The purpose is to have more consistency checking in debug mode. Thus it is not supposed now and never was in standard that the return type of the &u[j] can be an iterator of the same type. It was just an implementation characteristics which made possible your example. You have to replace it with: vector<T>::iterator i = u.begin() + j; Regards [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Thomas Maeder Guest Posted: Sat Mar 26, 2005 9:46 am    Post subject: Re: iterator points to entry of vector "Stephen Vavasis" <vavasis (AT) cs (DOT) cornell.edu> writes: Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. Can someone confirm that the above statement is nonconforming? Yes. I can. &u[i] is of type T *, which isn't necessarily implicitly convertible into vector<T>::iterator Quote: What is the preferred way to get the desired outcome? vector<T>::iterator i(u.begin()+j); [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Carl Barron Guest Posted: Sat Mar 26, 2005 5:24 pm    Post subject: Re: iterator points to entry of vector In article <d1uurt$kfp$1 (AT) news01 (DOT) cit.cornell.edu>, Stephen Vavasis <vavasis (AT) cs (DOT) cornell.edu> wrote: Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. nowhere did/does the standard requre or state tbat a vector<T>::iterator and T * were the same type. In many early and some 'production' versions it is/was just a typedef. A way to do the above that should work regardless if vector<...>::iterator is a typedefed pointer or a special class. is: vector<T>::iterator i = u.begin()+j; [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Ali Çehreli Guest Posted: Sat Mar 26, 2005 5:32 pm    Post subject: Re: iterator points to entry of vector Stephen Vavasis wrote: Quote: In some C++ code that I wrote around 2000, I needed an iterator that points to an entry of a vector. I used the following statement: vector<T>::iterator i = &u[j]; where u is of type vector<T> and j is an int. Type T is a struct. That code used to work by an unlucky coincidence. In the library(ies) that was compiled with, vector<T>::iterator was highly likely a typedef of a pointer to the stored type (T*). But it never had to be that way. Quote: At that time, all the compilers I tried accepted this statement and produced a program that ran correctly, so I didn't realize that the above statement is perhaps nonconforming. You've been unlucky not to discover the problem before so long. Quote: But gcc-3.2.2 is now rejecting that statement (giving an error message that the type of the right-hand-side cannot be converted to the type of the left-hand side), and I suspect that it may be nonconforming. vector::iterator can be obtained through the vector interface; anywhere else may or may not work. Quote: Can someone confirm that the above statement is nonconforming? I don't think that the standard defines the behavior of that statement. It may work as a result of implementation choices (as it did in the past for you.) Quote: What is the preferred way to get the desired outcome? For a random access iterator like vector<T>::iterator: vector<T>::iterator i = u.begin() + j; or more generally for any type of iterator, still at constant time for random access iterators: vector<T>::iterator i = u.begin(); advance(i, j); Ali [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top JK Guest Posted: Sat Mar 26, 2005 5:39 pm    Post subject: stl - containers question regarding stl containers... is there a container which would allow me to search the container using the value field, not just the key, e.g., i have a list of people names and phone numbers which i wanna search using both the phone number and the person name? With maps, you can only search using the key and not the value? Thanks for any suggestions. [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Stefan Strasser Guest Posted: Sun Mar 27, 2005 10:20 am    Post subject: Re: stl - containers JK schrieb: Quote: question regarding stl containers... is there a container which would allow me to search the container using the value field, not just the key, e.g., i have a list of people names and phone numbers which i wanna search using both the phone number and the person name? With maps, you can only search using the key and not the value? Thanks for any suggestions. have a look at boost multiindex containers: http://www.boost.org/libs/multi_index/doc/index.html I've never used them so I don't know if it's possible to access the key of one index by the key of another index but you could still also store the keys inside the value. [ See http://www.gotw.ca/resources/clcm.htm for info about ] [ comp.lang.c++.moderated. First time posters: Do this! ] Back to top Carl Barron Guest Posted: Sun Mar 27, 2005 10:24 am    Post subject: Re: stl - containers In article <pha1e.9339$e_7.7638 (AT) fe10 (DOT) lga>, JK <fakeemail (AT) yahoo (DOT) com> wrote: Quote: question regarding stl containers... is there a container which would allow me to search the container using the value field, not just the key, e.g., i have a list of people names and phone numbers which i wanna search using both the phone number and the person name? With maps, you can only search using the key and not the value? Thanks for any suggestions. boost has a multi_index container. [www.boost.org], and other goodies. [ 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.