[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: sections



Original-Via: uk.ac.ukc; Wed, 31 Oct 90 16:28:42 GMT
From: Kent Karlsson <kent>
Date: Wed, 31 Oct 90 13:50:33 +0100
To: haskell@cs.glasgow.ac.uk
Subject: Re:  sections
Original-Sender: kent%se.chalmers.cs@sunic.sunet.se
Sender: haskell-request@cs.glasgow.ac.uk


> a ``semantic'' check that the "exp" in the second production is really
> an "aexp", so giving the exact syntax that you suggest -- "exp" is used
...
> There's no problem with *parsing* this that I can see (just allow "_"
> as a valid "aexp"!), but the compiler must be a bit more clever if it is
> to turn one of these into a lambda later on (and also check for illegal
> uses of "_").

There is nothing context sensitive about "extended" sections, as this
fragment of context free grammar shows:

	aexp -> ( exp )
	      | var
	      | ...

	exp  -> application
	      | aexp
	      | exp1 op exp2

	      | sapplication
	      | _    op exp2
	      | exp1 op _

	      | ...

	application  -> application  aexp
	              | aexp aexp

	sapplication -> sapplication aexp
	              | application  _
	              | aexp _


				/kent k