Re: classifying data

Well, you make use of Regexp::Common. For instance, by doing:

    use Regexp::Common;
    $has_dollar_sign = $str =~ s/^([-+]?)\$/$1/;
    if ($str =~ /^$RE{num}{decimal}{-sep => ','}{-keep}$/) {
        $is_numeric        = 1;
        $sign              = $2;
        $has_decimal_point = $5 ? 1 : 0;
        $has_commas        = $str =~ /,/
    }
[download]

Or you could take its regex and modify it to have an optional leading dollar sign.

Abigail

Comment on Re: classifying data Download Code

Replies are listed 'Best First'.
Re: Re: classifying data by Art_XIV (Hermit) on Jan 19, 2004 at 16:45 UTC
This could help you get started: `use warnings; use strict; while (<DATA>) { chomp; my $data = $_; print "$data: "; my $result = ($data =~ /^[+\-]?\$?[0-9.]+$/) ? "numeric" : "non-numeric"; print "$result\n"; } __DATA__ 1020 $10.21 -1023 +1.024 beer 10$25 -$102.6 A1027 1028$ $ $-1.029` [download] BTW, watch out for those asterisk opertors in your patters. You used `\d`, and the `''` could legitimately match `''`, since it does match a digit zero or more times! That pesky asterisk operator can lead to 'zero-width' matches, which can drive you nuts when you are starting with regular expresssions. Hanlon's Razor - "Never attribute to malice that which can be adequately explained by stupidity"	[reply] [d/l] [select]
Re: classifying data by Abigail-II (Bishop) on Jan 19, 2004 at 16:50 UTC
Your regexp will not match `900,000`, but it will match `....`. Abigail	[reply] [d/l] [select]
Re: Re: classifying data by Art_XIV (Hermit) on Jan 19, 2004 at 17:40 UTC
Upon reflection, I think that your plan of using RegExp::Common is superior. Hanlon's Razor - "Never attribute to malice that which can be adequately explained by stupidity"	[reply]