PERL Beginners - reg ex problem

Author

reg ex problem

Ryan Moszynski

2006-06-27, 6:57 pm

i have this string extracted from a text file i'm writing a program to process:

test_freq = 1.0001;

and i have to extract the "1.0001"

i can't count on the whitspace being where it now is.

I would like to change this line of perl

$getTestFRQ =~ s/\D+//g;

so that instead of killing all non digit characters, it will kill all
non digit characters except for the period.

How do i do this?

thanks, ryan

Wagner, David --- Senior Programmer Analyst --- WG

2006-06-27, 6:57 pm

Ryan Moszynski wrote:
> i have this string extracted from a text file i'm writing a program
> to process:=20
>=20
> test_freq =3D 1.0001;
>=20
> and i have to extract the "1.0001"
>=20
> i can't count on the whitspace being where it now is.
>=20
> I would like to change this line of perl
>=20
> $getTestFRQ =3D~ s/\D+//g;
>=20
> so that instead of killing all non digit characters, it will kill all
> non digit characters except for the period.
>=20
> How do i do this?
>=20
if ( /=3D\s*([\d\.]+)/ ) {
$MyNbr =3D $1;
}
if always with a ; then
if ( /=3D\s*([^;\s]+)/ ) {
$MyNbr =3D $1;
}
The first one would be my choice, but just a thought.
> thanks, ryan

Wags ;)
WGO: x2224

****************************************
******************************
This message contains information that is confidential and proprietary to F=
edEx Freight or its affiliates. It is intended only for the recipient name=
d and for the express purpose(s) described therein. Any other use is proh=
ibited.
****************************************
******************************

Timothy Johnson

2006-06-27, 6:57 pm

Here's one example of how you could go about it:

##########################

use strict;
use warnings;

my $testString =3D " test_freq =3D 1.0001;";

print "Test One: ";
if($testString =3D~ /=3D\s*([0-9.]+)\s*;\s*$/){
print $1;
}else{
print "Failed!";
}
print "\n";

##########################

'[0-9.]' is a character class that represents the numbers zero through
nine and the period character.

I added in a few '\s*' matches in case there were spaces before and
after the semicolon. This should be specific enough to match only a
number that is a n rvalue.

-----Original Message-----
From: Ryan Moszynski [mailto:ryan.m.lists@gmail.com]=20
Sent: Tuesday, June 27, 2006 1:27 PM
To: beginners@perl.org
Subject: reg ex problem

i have this string extracted from a text file i'm writing a program to
process:

test_freq =3D 1.0001;

and i have to extract the "1.0001"

i can't count on the whitspace being where it now is.

I would like to change this line of perl

$getTestFRQ =3D~ s/\D+//g;

so that instead of killing all non digit characters, it will kill all
non digit characters except for the period.

How do i do this?

thanks, ryan

--=20
To unsubscribe, e-mail: beginners-unsubscribe@perl.org
For additional commands, e-mail: beginners-help@perl.org
<http://learn.perl.org/> <http://learn.perl.org/first-response>

Dr.Ruud

2006-06-27, 6:57 pm

"Ryan Moszynski" schreef:
> i have this string extracted from a text file i'm writing a program
> to process:
>
> test_freq = 1.0001;
>
> and i have to extract the "1.0001"
>
> i can't count on the whitspace being where it now is.
>
> I would like to change this line of perl
>
> $getTestFRQ =~ s/\D+//g;
>
> so that instead of killing all non digit characters, it will kill all
> non digit characters except for the period.
>
> How do i do this?

Possible:

s/[^\d.]+//g

but not safe.

Variant:

($getTestFRQ) = /=\s*(\d+\.\d+)/ ;

assuming the string is in $_.

You could also use 'test_freq' as a hash key:

my %h ;
/^\s*(\w+)\s*=\s*(\d+(?:\.\d*)?)/ and $h{lc $1} = $2 ;

(untested)

--
Affijn, Ruud

"Gewoon is een tijger."

John Ackley

2006-06-27, 6:57 pm

Ryan Moszynski wrote:
> i have this string extracted from a text file i'm writing a program to
> process:
>
> test_freq = 1.0001;
>
> and i have to extract the "1.0001"
>
> i can't count on the whitspace being where it now is.
>
> I would like to change this line of perl
>
> $getTestFRQ =~ s/\D+//g;
>
> so that instead of killing all non digit characters, it will kill all
> non digit characters except for the period.
>
> How do i do this?
>
> thanks, ryan
>
try this

use strict;
use warnings;

my $input = 'test_freq = 1.0001; ';

if($input =~ /([\d\.]+)/) {
print $1,$/;
} else {
print "not found\n";
}

print $1,$/ if $input =~ /([\d\.]+)/;

$input =~ s/[^\d\.]//g;
print $input,$/;

Ryan Moszynski

2006-06-27, 6:57 pm

thanks for the help, that did the trick

On 6/27/06, Ryan Moszynski <ryan.m.lists@gmail.com> wrote:
> i have this string extracted from a text file i'm writing a program to process:
>
> test_freq = 1.0001;
>
> and i have to extract the "1.0001"
>
> i can't count on the whitspace being where it now is.
>
> I would like to change this line of perl
>
> $getTestFRQ =~ s/\D+//g;
>
> so that instead of killing all non digit characters, it will kill all
> non digit characters except for the period.
>
> How do i do this?
>
> thanks, ryan
>

Sponsored Links