vectorvectorint ::iterator to std::vectorT conversion

If I understand what you're asking, (though I'm not entirely sure that I do) you can dereference the vector > iterator to get a reference to the vector sub element and use that to assign to a vector.

  std::vector<int> target;  std::vector<std::vector<int> > vector_of_vectors;  // populate vector_of_vectors somehow  target = *vector_of_vectors.begin();

Thanks, that's what I was asking about but i need to use a little algorithm to figure what subsquare the squares index would be in not by iterating linearly from the begin to the end of the row and column

Don't use a vector of vectors to represent a rectangular grid. For Sudoku, you already know the size, so you can just use an array[9][9], possibly wrapped in a struct of some sort (such as boost::array); otherwise, you could use boost::multi_array.

operator= of an iterator is for assigning from one variable of that iterator type to another variable of the same iterator type. Dereferencing an iterator gives you a reference (confusing, hmm?) to the element the iterator pointed at (iterated to?).

With a bit of work, you could make an iterator over the grid that iterates in "subsquare order", but it's realistically not worth the effort.

Quote:

With a bit of work, you could make an iterator over the grid that iterates in "subsquare order", but it's realistically not worth the effort.

Well I am writing a sudoku solver and I need to store the squares in many ways using iterators such as how close the square is to being a single canidate ( has one possible canidate) and then have another dimention of the canidates in ascending order; this is in all the means of efficiency... To do processes like this, I need to check the squares row column and subsquare so I cand find canidates as well as check rows colummns and subsquares for single positions

You'll want to abstract more, so that while your storage might well be a boost::array or a std::vector this is inside another class (such as a 'board' class) which does all the logic of calculating colums, rows and sub-squares for you via functions.

Your calling program then uses these functions to ask the board about it's state and cares not what the underlaying storage system is.

Quote:

Original post by Mad_Coder Quote: With a bit of work, you could make an iterator over the grid that iterates in "subsquare order", but it's realistically not worth the effort. Well I am writing a sudoku solver and I need to store the squares in many ways using iterators such as how close the square is to being a single canidate ( has one possible canidate) and then have another dimention of the canidates in ascending order; this is in all the means of efficiency

No you don't; just make sorted containers of references to squares, and/or iterate manually to cover subsquares/rows/columns. Adding complexity a priori is a bad idea unless you can prove it makes big-O improvements to run time; otherwise, follow the usual rules of optimization.