Re: [TCLCORE] TIP #320: Improved Variable Handling in the Core Object System

[Donal K. F. <don...@ma...>]

Will Duquette wrote:
> In Donal's example, we have an instance variable, utterance, declared
> for the "animal" class; the variable is visible without declaration in
> subclass "cat"'s methods, e.g., "speak2".

Oops. I misread.

The speak2 method wouldn't have worked as written without either:
   a) the cat class defining [variable utterance], or
   b) an explicit [my variable utterance].

> This strikes me as dangerous; a new instance variable added to an
> ancestor class that has the same name as a local variable used in a
> subclass method can break the subclass method.
> 
> Donal, in your example, "dog" declares utterance explicitly; does that
> make any difference?  Given that method "cat"'s "speak2" can see
> utterance, it would appear not.

I made a mistake in reading Arnulf's extension of my example. :-)

The only variables automatically brought into a method are those that
are declared at the same level (i.e. in the same class/object) as the
method. This should mean that if a superclass increases the set of
variables it declares, there will only be a problem if the parent
declares a variable that a subclass was previously declaring (and
they're used inconsistently). But that case is one I'm explicitly not
providing help for working around; sealing each variable to its class
completely makes for much more difficult introspection, and I'll need
some arm twisting to change my mind on that.

Donal.