Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:52514
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: error (pb1.pair.com: domain macvicar.net from 74.125.83.170 cause and error)
References: <BA175FCA-75BB-43E6-BE2A-03D1997F6C15@roshambo.org>
In-Reply-To: <BA175FCA-75BB-43E6-BE2A-03D1997F6C15@roshambo.org>
Mime-Version: 1.0 (iPhone Mail 8J2)
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii
Message-ID: <2EFE9027-13EA-4570-B6D2-2C56D85C7DAE@macvicar.net>
Cc: PHP Internals <internals@lists.php.net>
Date: Thu, 26 May 2011 21:10:30 -0700
To: Philip Olson <philip@roshambo.org>
Subject: Re: [PHP-DEV] INF behavior
From: scott@macvicar.net (Scott MacVicar)

On 26 May 2011, at 20:03, Philip Olson <philip@roshambo.org> wrote:

> Hello geeks,
>=20
> A geek is needed to clarify PHP bug #45712. This is an edge case but the t=
est (bug45712.phpt) contains code similar to the following:
>=20
> <?php
> $inf =3D pow(0, -2);
>=20
> var_dump($inf);          // float(INF)
> var_dump($inf =3D=3D  $inf); // bool(false)
> var_dump($inf =3D=3D=3D $inf); // bool(true)
> ?>
>=20
> That's how it's behaved since ~forever (AFAICT) and remains in 5.3.7-dev, b=
ut PHP 5.4.0-dev changes behavior so both now return true.
>=20
> Is this is how we want it? And how should this be documented/explained?

I think I changes this :-)

It didn't make sense that =3D=3D and =3D=3D=3D produce different results.

Though if someone has a better understanding of maths then we can fix it.

S=