Re: not(1) && 0 is true!

[p5pRT]

Migrated from rt.perl.org#1750 (status was 'resolved')

Searchable as RT1750$

[p5pRT]

From ilya@math.ohio-state.edu

I very much doubt this description. You are missing "if it looks like
a function call, it is a function call" rule (see C<print (1)>
clause). lc has a very low precedence "on the right" as well.

But somehow C<not> got through the cracks of this rule. This is most
probably a bug (so I CC perlbug).

it may be ugly but how would you get this to work​:

not ($foo && $bar ) || ($aaa && $bbb)

not can't grab the first parens according to boolean algebra.

Boolean algebra has no relationship to parens. You are trying to
pretend that Perl is parsed basing on precedences. It is not.

Ilya

[p5pRT]

From @TimToady

-------------- Forwarded message -----------------
Return-Path​: <larry@​kiev.wall.org>
Received​: (qmail 1213 invoked from network); 9 Nov 1999 00​:40​:30 -0000
Received​: from kiev.wall.org (205.178.11.135)
  by tmtowtdi.perl.org with SMTP; 9 Nov 1999 00​:40​:30 -0000
Received​: by kiev.wall.org (8.9.3/8.9.3) id KAA09914;
  Mon, 8 Nov 1999 10​:00​:36 -0800 (PST)
Date​: Mon, 8 Nov 1999 10​:00​:36 -0800 (PST)
From​: Larry Wall <larry@​wall.org>
Message-Id​: <199911081800.KAA09914@​kiev.wall.org>
To​: ilya@​math.ohio-state.edu (Ilya Zakharevich)
Cc​: perl5-porters@​perl.org
Subject​: Re​: [ID 19991106.003] Re​: not(1) && 0 is true!
In-Reply-To​: <199911062047.PAA09734@​monk.mps.ohio-state.edu>
(from Ilya Zakharevich on Sat, 6 Nov 1999 15​:47​:30 -0500 (EST))

Ilya Zakharevich writes​:
: [A complimentary Cc of this posting was sent to Uri Guttman
: <uri@​sysarch.com>],
: who wrote in article <x7ln8b34dk.fsf@​home.sysarch.com>​:
: > different precedence covers that. not ($x) is an low precedence op
: > followed by a term. lc ($x) is a higher precedence unary op with an
: > arg.
:
: I very much doubt this description. You are missing "if it looks like
: a function call, it is a function call" rule (see C<print (1)>
: clause). lc has a very low precedence "on the right" as well.
:
: But somehow C<not> got through the cracks of this rule. This is most
: probably a bug (so I CC perlbug).

Sarathy put in a fix for this a month or so ago. Try it in _62.

Larry
---------- End of forwarded message --------------