MOO-cows Mailing List Archive


Re: Use for $call_verb

At 18:46 -0800 8/4/96, Kipp the Kidd wrote:
>For the mentally blind, ...

Not shit here! God is speaking. I almost see the light.

>I have a better idea.  In xxx:yyy(zzz), if xxx is not an object, return
>$frobs_utils:yyy(xxx,zzz);  I doubt this will ever make it into a final
>release, but I would REALLY REALLY (please please?) like someone to make a
>patch for this.

Do you really think it is a better idea? May I question the quality of your
Have you ever thought about frob implementation for a minute?

What do you expect? all frobs to define the same verbs or to stick a * verb
your $frobs_utils?

I'm not a frob guru (Jay, Andy, Erik?) but I feel if you want frobs, they
would be best served by a #0:handle_frob_call(frob, verb, args) call that
would dispatch the verb call to the appropriate frob handler with the
proper arguments... Of course that's only a cheap way to do it... if a
generic frob type exist that hold information about the frob handler and
the frob data, then you can have the server dispatch the verb call  to the
frob handler and maybe do some trick like setting the value of 'this' to
the frob data... <shrug>

But as number of people expressed it, custom datatypes is on the big wish
list... and if I remember well some of the tales of Uncle Jay, Erik
implemented frobs in-server ages ago...

So, I guess (and hope) like for the profiler we will see implementation
description of the frobs before they make their way in a next version...

>I think we're just getting too lazy.  We want the MOO to figure out what
>$util has what verb and which uses the correct type.

And you expect a database to have utilities. Which is not OO at all.

I would like to point to your not-mentally-blind-but-almost-ness that with
frobs and non-overridable you can COMPLETELY get ride of $..._utils kind of

I hope that will help to cure your confusion.

Yours in darkness,

Dark Janus

Follow-Ups: References:

Home | Subject Index | Thread Index