[SCL] Re: SCL

[Pat Hayes]

>In the document:
>
>The lexical analysis section needs some touching up:
>wordchar is not defined, seqvarchar is not used, ...

wordchar should be namechar. Fixed.

>
>In the core expression syntax section, termseq is defined
>  termseq = S, {term, S}, seqvar?
>which requires some whitespace always be included and
>makes (R) syntactically illegal.  It would be defined
>better by
>  termseq = {S, term}, (S, seqvar)?
>
>I notice there are some places where whitespace is
>clearly necessary but not specified in the syntax
>(in equalities,

  (=(f a)(f b)) seems OK.

I guess we ought to forbid names starting with the '=' character, though.

Actually what is required, I now see, is that if one or the other 
term is more than a simple name, then no whitespace is needed; but if 
they are both simple names then whitespace is mandatory between them. 
Which seems to require re-writing the entire syntax with these two 
cases split, just in order to get the whitespace right, so that there 
will have to be 2 syntax cases for all the connectives. Sigh.

>  separating sentences, and separating
>sentences from connective names, etc.

Since even the smallest sentence must be enclosed by parentheses, I 
don't think whitespaces here are needed, though they should obviously 
be optional.

>)
>
>The syntax is still somewhat illiberal by Lisp standards,
>since whitespace is often required where no ambiguity
>would result from its omission.
>Lisp allows (R(f a b)(g c)), but your syntax requires
>(R (f a b) (g c)) (whitespace separating all terms, even
>those with parentheses).  Is this a feature or a bug?

A bug. I thought I had arranged it so that whitespace next to 
parentheses was (almost) always optional.

Clearly I need to go over the syntax (yet) again. If you have any 
suggestions for how to do it better, I'd be grateful.

Pat



-- 
---------------------------------------------------------------------
IHMC	(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32501			(850)291 0667    cell
phayes at ihmc.us       http://www.ihmc.us/users/phayes