C++ problem with std::pair and forward declarations

Question

Unfortunately I still got a problem with my templated code from here:

on line 49 in the file 'utility':

error C2440: 'Initializing': cannot convert from 'const int' to 'IntersectionData *' error C2439: 'std::pair<_Ty1,_Ty2>::second': member could not be initialized

how could i figure out where the problem is? the only place i use a pair with 'IntersectionData*' is here:

#include "MRMaterialMatth.h" #include "IntersectionData.h" using namespace std; struct IShaderMatth { virtual ~IShaderMatth() {} vector<pair<MaterialMatth,IntersectionData*> > traceCols; };

and there are not any other compiler errors

how can I track down this?

//edit: utility is not my code. it must be from std.. the code of line 49 looks like this:

template<class _Other1, class _Other2> pair(const pair<_Other1, _Other2>& _Right) : first(_Right.first), second(_Right.second) { // construct from compatible pair }

line 49 is the line of the comment

edit2: the only places where i change something about the content of tracecols look like this:

 IntersectionData* iDataOut = NULL; if(newIData_out!=NULL) { iDataOut = new IntersectionData(*iData); } traceCols->push_back(make_pair(MaterialMatth(),iDataOut));

and

 if(traceCols){ traceCols->push_back(make_pair(MaterialMatth(), NULL)); }

and

 if(traceCols) { (*traceCols)[traceCols->size()].second = new IntersectionData(*newIData); }

is NULL the problem? it's a pointer, so i should be allowed to create a pair with NULL, no??

What is line 49 of utility, and where else do you use traceCols? — Smashery
– Smashery, Commented Dec 16, 2009 at 1:56

dave4420 · Accepted Answer · 2009-12-16 02:08:11Z

2

Try explicitly casting the NULL to IntersectionData * in your call to make_pair().

if(traceCols){ traceCols->push_back(make_pair(MaterialMatth(), (IntersectionData *)NULL)); }

answered Dec 16, 2009 at 2:08

dave4420

47.2k6 gold badges121 silver badges153 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Mat Over a year ago

hey yeah - that solved the Problem! but i'd be really glad if someone could explain me why this is necessary?

Mike Seymour Over a year ago

In Visual Studio (which I assume you're using from the error messages), NULL is defined as a literal 0, so your use of make_pair becomes make_pair(MaterialMatth(),0), of type pair<MaterialMatth,int>.

Drew Dormann Over a year ago

In fact, the C++ standard demands that NULL be defined as 0, regardless of which compiler you're using.

Zan Lynx · Accepted Answer · 2009-12-16 01:55:26Z

There is a problem initializing one of those pairs.

Ask yourself, "What initializes that?"

The answer is the vector traceCols.

Now ask, "Where am I creating elements in traceCols?"

Once you answer that, you should know what is going wrong.

could be the problem that i create a pair with NULL? NULL probably is a const int..

Smashery · Accepted Answer · 2009-12-16 02:08:42Z

Watch out for the line (*traceCols)[traceCols->size()].second = new IntersectionData(*newIData) - it seems like that would go out of the vector's bounds (since the largest index of a vector is size() - 1).

I'm not sure if the NULL is causing it - so comment out that line, and see for yourself (or try Dave's suggestion)! If it doesn't work, comment out another. Eventually, you'll either find what line, and be able to ask a more specific question, or it'll be none of those things, and you'll know you have to search somewhere else. That's what I do when I see all these silly compiler error messages.

Tim Sylvester · Accepted Answer · 2009-12-16 01:56:35Z

It looks like you have an assignment somewhere from a pair<MaterialMatth,int>. The compiler is trying to convert from that to the declaration you listed, but it can't convert from an int to a pointer without an explicit cast.

Collectives™ on Stack Overflow

C++ problem with std::pair and forward declarations

4 Answers 4

3 Comments

1 Comment

Comments

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

3 Comments

1 Comment

Comments

Comments

Linked

Related