Re: Design Problem Aggregation
From: Universe (universe_at_tAkEcovadOut.net)
Date: 05/12/04
- Next message: Neo: "Re: Transitive Closure with XDb1"
- Previous message: Mark S. Hathaway: "Re: Why is OO popular?"
- In reply to: Robert C. Martin: "Re: Design Problem Aggregation"
- Next in thread: Merlin: "Re: Design Problem Aggregation"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 12 May 2004 16:04:08 -0400
"Robert C. Martin" <unclebob@objectmentor.com> wrote in message
> Thus, both A and B derive from a separate interface named Gpart. G
> contains a list of Gparts. When you pull a Gpart out of the list
you
> then cast it to a Base and use it naturally.
>
> This is called the Interface Segregation Principle.
By RCM, and some XP'ers, but *not* by others associated with OO
software engineering.
Further, with C++, one need only, in turn, wield (use) the
individual pointers of the base classes multiply inherited by a
hierarchy of classes to isolate each of the individual interfaces
contributed to the classes by each those base classes--*no*
casting need be foisted off as necessary.
And given that interfaces form "Aggregate" class abstractions, which
all carry all of the interfaces together within them, there is no
ugly "Segregation" involved [as the malevolent like minded whom
coin such terms for broad use would have it] but on the contrary
such "Aggregations" shout *Integration*. Yippee!!!
Elliott
-- He whom supports abstraction in the lead of development, like BMeyer, is far ahead of the nickel and dime, baby steppers. <g>
- Next message: Neo: "Re: Transitive Closure with XDb1"
- Previous message: Mark S. Hathaway: "Re: Why is OO popular?"
- In reply to: Robert C. Martin: "Re: Design Problem Aggregation"
- Next in thread: Merlin: "Re: Design Problem Aggregation"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|