hi all,

I have a little question about serialization, shown by the following example:
Qt Code: 
  1. class MyClass
  2. {
  3. 	public:
  4. 		MyClass(){}
  5. 		~MyClass(){}
  6. 		void save(QDataStream &out){
  7. 			out << (qint32)_id;
  8. 			out << (qint32)_sizeX;
  9. 			out << (qint32)_sizeY;
  10. 			out << (qint32)_sizeZ;
  11. 		}
  12. 		void load_Version1(QDataStream &in){
  13. 			qint32 id, sizeX, sizeY, sizeZ;
  14. 			in >> id;
  15. 			in >> sizeX;
  16. 			in >> sizeY;
  17. 			in >> sizeZ;
  18. 			_id = id;
  19. 			_sizeX = sizeX;
  20. 			_sizeY = sizeY;
  21. 			_sizeZ = sizeZ;
  22. 		}
  23. 		void load_Version2(QDataStream &in){
  24. 			in >> _id;
  25. 			in >> _sizeX;
  26. 			in >> _sizeY;
  27. 			in >> _sizeZ;
  28. 		}
  29. 	private:
  30. 		int _id;
  31. 		int _sizeX;
  32. 		int _sizeY;
  33. 		int _sizeZ;
  34. };
To copy to clipboard, switch view to plain text mode 

using load_Version1 to load my data should be safe in terms of data integrity, afaik

doing things like in load_Version1 is a little longish, imo, especially when the list of various types of private members grows

so, would it be ok & safe to use load_Version2 to shorten things???
personally I don't think it is safe...

if load_Version2 isn't ok, as i assume, is there another possibility to shorten the amount of code?
would something like
Qt Code: 
  1. in >> (qint32)_id;
To copy to clipboard, switch view to plain text mode 
work?

I hope you get, what I talk about

greetz
darksaga