Bug #23686 for Scalar-List-Utils: "Nan" is a number!

Tue Nov 28 17:01:16 2006 timothybolshaw [...] gmail.com - Ticket created

Subject:	"Nan" is a number!
Date:	Wed, 29 Nov 2006 05:00:51 +0700
To:	bug-Scalar-List-Utils [...] rt.cpan.org
From:	"Tim Bolshaw" <timothybolshaw [...] gmail.com>

use Scalar::Util qw\looks_like_number\; my $string = 'Nan'; if ( looks_like_number $string ) { print "$string is a number\n"; } else { print "$string is a string\n"; } It is obvious, of course, what is happening. It tickles my funny bone that 'Not-A-Number' looks like a number. Murphy's law applied: I was processing a word array ... one of the words was "Nan"! -- Tim Bolshaw timothybolshaw@gmail.com Mobile: +66-(0)87 072 5009

Tue Nov 28 19:51:47 2006 gbarr [...] pobox.com - Correspondence added

Subject:	Re: [rt.cpan.org #23686] "Nan" is a number!
Date:	Tue, 28 Nov 2006 18:51:20 -0600
To:	bug-Scalar-List-Utils [...] rt.cpan.org
From:	Graham Barr <gbarr [...] pobox.com>

On Nov 28, 2006, at 4:01 PM, Tim Bolshaw via RT wrote: Show quoted text

> use Scalar::Util qw\looks_like_number\; > > my $string = 'Nan'; > if ( looks_like_number $string ) { > print "$string is a number\n"; > } > else { > print "$string is a string\n"; > } > > It is obvious, of course, what is happening. It tickles my funny > bone that > 'Not-A-Number' looks like a number. Murphy's law applied: I was > processing > a word array ... one of the words was "Nan"!

looks_like_number is a direct mapping onto a C function in the perl source. So it returns true for anything Perl would consider using as a number without giving a warning. NaN and Inf are both strings that are considered valid in a numeric context. Graham.

Tue Nov 28 19:51:47 2006 The RT System itself - Status changed from 'new' to 'open'

Sat Dec 09 08:11:26 2006 GBARR [...] CPAN.ORG - Status changed from 'open' to 'resolved'

Sat Dec 09 08:29:50 2006 timothybolshaw [...] gmail.com - Correspondence added

Subject:	Re: [rt.cpan.org #23686] Resolved: "Nan" is a number!
Date:	Sat, 9 Dec 2006 20:29:27 +0700
To:	bug-Scalar-List-Utils [...] rt.cpan.org
From:	"Tim Bolshaw" <timothybolshaw [...] gmail.com>

Well, the response was that a test of 'NaN' against the standard C library routine test for a valid number will state that it is. I understand that (as surely implied by my original report) but if the issue is to be regarded as closed, at least the documentation ought be updated to warn that 'looks like number' does not necessarily mean the value incorporates a single numeric digit : indeed 'Nan' (not a number) and some other word abbreviations look like numbers. The function can be very useful for some (not all) kinds of numeric field validation, but not wider use. In particular, it should not be used to try to disambiguate numeric from non-numeric values. Perhaps there is a need for a 'looks like real-number' function. Tim Bolshaw. On 12/9/06, Graham_Barr via RT <bug-Scalar-List-Utils@rt.cpan.org> wrote: Show quoted text

> > <URL: http://rt.cpan.org/Ticket/Display.html?id=23686 > > > According to our records, your request has been resolved. If you have any > further questions or concerns, please respond to this message. >

-- Tim Bolshaw timothybolshaw@gmail.com Mobile: +66-(0)87 072 5009

Sat Dec 09 08:29:53 2006 The RT System itself - Status changed from 'resolved' to 'open'

Sat Dec 09 11:54:56 2006 GBARR [...] CPAN.ORG - Status changed from 'open' to 'resolved'