[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Questions about reflective operations on functions and methods
- To: info-dylan@cambridge.apple.com
- Subject: Re: Questions about reflective operations on functions and methods
- From: lynch@ils.nwu.edu
- Date: Mon, 5 Oct 92 12:17:16 CDT
>> 1) Are the methods of APPLICABLE-METHOD?, SORTED-APPLICABLE-METHODS, etc.,
>> for <generic>'s frozen? If not, what is supposed to happen if I substitute
>> other methods in their place.
>
>In general, the current manual doesn't sufficiently specify which
>branches of which generic functions are frozen. We are going to
>specify this as part of the next round of language design, which
>we're about to start.
>
>> 2) What is supposed to happen if I create new methods for these generics.
>> For instance, can I add methods to these generics in such a way that
>>
>> (<object> <keyword> ...)
>>
>> is a function application a la Smalltalk?
>
>The Dylan manual does not say that these functions are used by the Dylan
>runtime in performing function calls. So, if you added methods to them,
>it wouldn't necessarily have any effect on the system at all.
Just a quick interjection. PLEASE make this explicit as well, if this is
the case.
>> 3) If the answer to question #1 is Yes, and the answer to #2 is No, what is
>> the motivation for including these generics in the language specification?
>
>That's a good question. It may be that, lacking any intercessory capability,
>these functions aren't terribly useful. I can imagine APPLICABLE-METHOD?
>being useful as a sort of safety check, even without a full meta-object
>protocol. It's harder for me to see how SORTED-APPLICABLE-METHODS would
>be useful, except perhaps by the programming environment.
>
>Opinions?
If they're really not useful, perhaps they should be part of an
implementation suggestions, rather than language spec.