Re: [Chicken-users] Peculiar TinyCLOS specialization bit

[Joel Reymont]

> Andreas Rottmann wrote:

>The only method on the generic function foo is specialized on (<b>
><a>), while you pass (<b> <boolean>), so there are no applicable
>methods. Some error is the expected behaviour. If you want it to pass,
>do
>
>(define-method (foo (self <b>) (a <top>)) #f)
>
>or
>
>(define-method (foo (self <b>) (a <boolean>)) #f)

My goal is to be able to pass an object or an absence of one. 

Above basically means that I have to either specialize a method twice,
once for object and once for #f or not specialize it at all (top) and
check the passed value in my method. Right?

I was expecting it to sort of work like when passing pointers in C++ :-).

Too bad there's no way to create an "Empty object of no value" and pass
that instead of specializing the method for both class and boolean or
writing extra argument checks.

    Thanks, Joel

-- 
OpenPoker: The Linux of poker software
http://wagerlabs.com/forums
* Tenerife * Canary Islands * Spain