[.net] Question about compile and run time I think

The first is faster (to write) when all constructors initialize the list the same way. The second is faster (to execute) when constructors initialize the list differently. Otherwise, they are equivalent.

Correct me if I'm wrong but neither of the examples will compile. In the first example you are attempting to initialise a class member with operator new. The correct code would declare aList as a static member which you could define in a corresponding implementation file. This would allocate aList in static storage in the executable, so there would be no overhead of your constructor during runtime.

In the second example you are attempting to set aList to a new List template using operator new which returns a pointer to the allocated memory where the List object was created. The correct code would declare aList as a List template pointer, then the code in the constructor would work as expected. The second example would have runtime overhead of the small object allocation, which isn't much.

Which method you use depends on the class object itself. For the given example (a class object called 'Some') the appropriate usage is not at all obvious. When data is placed in static storage it is accessible in the global scope. Perhaps now I have explained it a little more it has become obvious which method is more appropriate and when? If not then I highly recommend reading a few chapters of Thinking in C++ by Bruce Eckel (Google search it, it's a free download!). Mr Eckel explains it a lot better than I ever could ;-)

Hope that helps! ( it probably doesn't ... )

Quote:

Original post by TheGilb

This is Sparta C# (no colon after public, no pointer semantics for new, no semicolon after class definitions, a standard library container called List).