When implementing operator[] how should I include bounds checking?

Question

First of all I apologize for the long lead up to such a simplistic question.

I am implementing a class which serves as a very long 1 dimensional index on a space filling curve or the n-tuple representing the Cartesian coordinate that index corresponds to.

class curvePoint
{
public:
    friend class curveCalculate;

    //Construction and Destruction
    curvePoint(): point(NULL), dimensions(0) {}
    virtual ~curvePoint(){if(point!=NULL) delete[] point;}

    //Mutators
    void convertToIndex(){ if(isTuple()) calc(this); }
    void convertToTuple(){ if(isIndex()) calc(this); }
    void setTuple(quint16 *tuple, int size);
    void setIndex(quint16 *index, int size);
    void setAlgorithm(curveType alg){algorithm = alg;}

    //Inspectors
    bool isIndex(){return current==Index;}
    bool isTuple(){return current==Tuple;}
    size_t size(){return dimensions;}
    quint16 operator[](size_t index);

    enum curveType{HilbertCurve, ZCurve, GrayCodeCurve};
    enum status{Index, Tuple};

private:
    curveCalculate calc;
    curveType algorithm;
    quint16 *point;
    size_t dimensions;
    status current;
};

(The length of the array pointed to by point is dimensions)

Anyways in the implementation of operator[] I was wondering what the best method to achieve bounds checking is. I want to avoid throwing exceptions if at all possible, and the full range of values is usable for each number in the array so a special value to return in case of an out of bounds error is not possible either;

I was thinking of something like this though implemented in the class definition:

quint16 curvePoint::operator[](size_t index)
{
    return point[ index % dimensions ];
}

This makes it so that we never leave the bounds of the array and if well documented I think it would be fine; nevertheless, I am leary of this particular implementation.

Does this look acceptable to others? Is there any other way of doing bounds checking while still satisfying my constraints?

Edit: Calculation of things like Hilbert curves etc are highly messy, messy enough that I do not not want the additional interface for the stl libraries in the way.

Additionally because I will have to convert many thousands of these every time the multidimensional database is queried I do not want the additional cost of the stl function calls in the mix, if at all possible.

I rather like the idea of the assert; but, if I remember correctly that breaks in release builds does it not?

I suppose I can use exceptions, that seems to be what everyone is rooting for, but I am using the Qt libraries and those avoid exceptions for both performance and portability and I was hoping to do the same.

Answer 1

The easiest solution is to do as C++ itself does. This limits the amount of surprises that your users will experience.

C++ itself is fairly consistent. Both the built-in [] on pointers and std::vector::operator[] have undefined behavior if you use an out-of-bound array index. If you want bounds checking, be explicit and use std::vector::at

Hence, if you do the same for your class, you can document the out-of-bound behavior as "standard".

Answer 2

Anyways in the implementation of operator[] I was wondering what the best method to achieve bounds checking is. I want to avoid throwing exceptions if at all possible, and the full range of values is usable for each number in the array so a special value to return in case of an out of bounds error is not possible either;

Then the remaining options are:

• Flexible design. What you did. "Fix" the invalid input so that it tries to do something which makes sense. Advantage: Function won't crash. Disadvantage: Clueless callers who access an out of bounds element will get a lie as a result. Imagine a 10-floor building with floors 1 to 10:

You: "Who lives in the 3rd floor?"

Me: "Mary".

You: "Who lives in the 9th floor?"

Me: "Joe".

You: "Who lives in the 1,203rd floor?"

Me: (Wait... 1,203 % 10 = 3...) > "Mary".

You: "Wow, Mary must enjoy great views from up there. So she owns two apartments then?"

• A bool output parameter indicates success or failure. This option usually ends up in not very usable code. Many users will ignore the return code. You are still left with what you return in the other return value.

• Design by Contract. Assert that the caller is within bounds. (For a practical approach in C++, see An exception or a bug? by Miro Samek or Simple Support for Design by Contract in C++ by Pedro Guerreiro.)

• Return a System.Nullable<quint16>. Oops, wait, this is not C#. Well, you could return a pointer to a quint16. This of course has lots of implications which I shall not discuss here and which probably make this option not usable.

My favorite choices are:

• For the public interface of a publicly released library: Input will be checked and an exception will be thrown. You ruled out this option, so it is not an option for you. It is still my choice for the interface of a publicly released library.
• For internal code: Design by contract.