🚀 go-pugleaf

"JD" <nospam@nospam.com> wrote:

>"Boguslaw Fries" <boguslaw.fries@polivision.it> wrote:
>>
>>Now in my program:
>>Myclass M( 5 );
>>int X;
>>
>>X = M; // X should obtain value 5
>
>That is not the case (that X == 5) and i doubt that that will
>even compile. 

I think you misunderstand the gentleman. He's asking how to ensure that
after the statement, X will contain the value 5.

And no, it won't compile as is. He's asking what changes to make to the
class so that a) it will compile, and b) that after that statement, X
contains the value 5. He's wanting something that works like a VARIANT.

>              The line:
>
>    Myclass M( 5 );
>
>creates an instance of Myclass named 'M', passing 5 to it's
>constructor (and does nothing else with it). The value of 'M'
>is in reality a 32bit pointer that must be correctly dereferenced
>to access class members or methods.

In reality, nothing of the sort. (Perhaps you're thinking of a different
programming language. Java or C#, perhaps?)

M is on the stack. It is a class, containing one int (currently set to
5) and an indeterminate float value. No pointer is involved, any more
than if M was an int rather than a MyClass.

>>How to implement extra things to do it?
>
>Here's a sample class that I use to get version info from my
>application. All you have to do is #include the header and add
>the lines (note that it support multiple languages as well):

[snip example of a TVersionInformation class]

To be honest, I haven't a clue what your example code has to do with
Boguslaw's question, and I doubt he has either. Are you sure it's not
the answer to a different question?

Alan Bellingham
-- 
ACCU Conference 2005 - 20-23 April, Randolph Hotel, Oxford, UK