• From: John Snelson <john.snelson@o...>
• To: Dimitre Novatchev <dnovatchev@g...>
• Date: Mon, 22 Sep 2008 15:39:54 +0100

Dimitre Novatchev wrote:
> So, I am sure this cannot be achieved with C++ generics, which is
> supported entirely at compile time. As for Java, I am not even aware
> if it has any generics support at all.

So ditch generics and use good old void* or Object. Just because it 
can't be written in a type safe way, doesn't mean it can't be written.

> The types    Node<T>,     Node<Node<T>>,     Node<Node<Node<T>>>, ...
> will need to be created at runtime, as it is impossible to know at
> compile time how deep the type nesting will need to be (although in
> the case of a finger tree we know this will be close to log2(N), where
> N is the number of leaf nodes of the tree).

That sounds like a runtime cost that I'd rather not have to pay for the 
sake of type safety.

John

-- 
John Snelson, Oracle Corporation            http://snelson.org.uk/john
Berkeley DB XML:            http://oracle.com/database/berkeley-db/xml
XQilla:                                  http://xqilla.sourceforge.net

• References:
  • Re: Discover data patterns or Create data patterns?
    • From: "Dimitre Novatchev" <dnovatchev@g...>
  • Re: Re: Discover data patterns or Create data patterns?
    • From: Rick Jelliffe <rjelliffe@a...>
  • Re: Re: Discover data patterns or Create data patterns?
    • From: "Dimitre Novatchev" <dnovatchev@g...>