nil or #nil?

David N. Smith dnsmith at watson.ibm.com
Sun Aug 23 20:20:38 UTC 1998


At 14:07 -0400 8/23/98, Michael Donegan wrote:
>
>OK. Tim is right about breaking existing code. That could be fixed. The
>fact that you
>couldn't put a nil in a constant array was the reason I thought that
>
>#( nil )
>
> and
>
>#( #nil )
>
>should be different. I'm happy to work on it either way. Let's get some
>more feedback when everyone is actually working. I don't know what we're
>doing.

For what it's worth, IBM Smalltalk 4.5 does this:

   #( nil #nil true #true false #false self super )

produces:

    #( nil #nil true #true false #false #self #super )

where nil, true, and false are the special objects.

I'd vote for this result, not because IBM Smalltalk does it, but because it
makes sense, is much more often the 'right thing', is in the proposed
standard, and makes Squeak have one less gotcha. It's these little things
that can drive you crazy when switching between systems.

Dave
_______________________________
David N. Smith
IBM T J Watson Research Center
Hawthorne, NY
_______________________________
Any opinions or recommendations
herein are those of the author
and not of his employer.





More information about the Squeak-dev mailing list