Re: SYNTAX ONLY - Precedence of prefix minus

[Kent Karlsson <kent@se.chalmers.cs>]

Original-Via: uk.ac.ukc; Fri, 15 Mar 91 16:40:42 GMT
From: Kent Karlsson <kent@se.chalmers.cs>
Date: Fri, 15 Mar 91 17:30:33 +0100
To: haskell@cs.glasgow.ac.uk
Subject: Re: SYNTAX ONLY - Precedence of prefix minus
Original-Sender: kent%se.chalmers.cs@mcsun.eu.net
Sender: haskell-request@cs.glasgow.ac.uk


Lex Augusteijn writes:

>	it was suggested that -2^2 should be equivalent to 0-2^2, which
>	would be parsed as (0-2)^2. 

No, parsed as 0-(2^2), '-' has lower precedence than '^'.

>	interpretation' of -x whould alsways be 0-x.
>	
>	May I conclude from this, that the mathematical interpreation of
>	2*-3 is 2*0-3, which will be parsed as (2*0)-3?
>	Hopefully not.

According to my proposal  2 * -3  is a syntax error, you must write 2*(-3).
But 2 >= -3 is ok, and parsed as 2 >= (-3), or if you like  2 >= 0-3, i.e. 2 >= (0-3).
And *- is one infix operator really, initially undefined, but you may define it
if you like:	a *- b   =  a * (-b)   (I'm ignoring the disallowance of the 
character '-' in arbitrary operators, this disallowance is a lexical problem
source...  Why should  *- be read as * -, while *+ is not split up?)

	This was my last(?) word on this subject :-)
			/kent k