1. Extensibility vs... ?
I'm creating a program and want to allow some easy extensibility in the
future... Please let me know your opinion on the matter:
I like the idea of
result = MyFuncs( { "FunctionName" , { Parameters } } )
and the brevity (read: "less typing" or "fewer keystrokes") of
result = FunctionName( Parameters )
but the former is more extensible/flexible.
Does anybody have a problem with the MyFuncs() method? A preference for one
or the other?
2. Re: Extensibility vs... ?
----- Original Message -----
From: "C. K. Lester" <euphoric at cklester.com>
To: "EUforum" <EUforum at topica.com>
Subject: Extensibility vs... ?
>
>
> I'm creating a program and want to allow some easy extensibility in the
> future... Please let me know your opinion on the matter:
>
> I like the idea of
>
> result = MyFuncs( { "FunctionName" , { Parameters } } )
>
> and the brevity (read: "less typing" or "fewer keystrokes") of
>
> result = FunctionName( Parameters )
>
> but the former is more extensible/flexible.
>
> Does anybody have a problem with the MyFuncs() method? A preference for one
> or the other?
Welcome to my world 
The first method does not preclude the second method. You define a new routine
(eg. FunctionName) and use it internally by the MyFuncs method and whoever wants
to can define a traditional-style interface to it, thus...
function FunctionName(object x)
return MyFuncs( {"FunctionName", {x}})
end function
I would go with the MyFuncs method because it is extensible.
In win32lib, the event handlers get passed a 'params' parameter. By using this
method, I've been able to add new data values in 'parms' without breaking
existing code.
--
Derek
3. Re: Extensibility vs... ?
> > I'm creating a program and want to allow some easy extensibility in the
> > future... Please let me know your opinion on the matter:
> >
> > I like the idea of
> >
> > result = MyFuncs( { "FunctionName" , { Parameters } } )
> >
> > and the brevity (read: "less typing" or "fewer keystrokes") of
> >
> > result = FunctionName( Parameters )
> >
> > but the former is more extensible/flexible.
> >
> > Does anybody have a problem with the MyFuncs() method? A preference for
one
> > or the other?
>
> The first method does not preclude the second method. You define a new
routine (eg.
> FunctionName) and use it internally by the MyFuncs method and whoever
wants to
> can define a traditional-style interface to it, thus...
>
> function FunctionName(object x)
> return MyFuncs( {"FunctionName", {x}})
> end function
True! This is basically a wrapper, and I could even provide a wrapper to the
base functions.
> I would go with the MyFuncs method because it is extensible.
>
> In win32lib, the event handlers get passed a 'params' parameter. By
> using this method, I've been able to add new data values in 'parms'
> without breaking existing code.
Yeah, I can appreciate that.
Thanks, Derek.
