Unable to match Indian Rupee currency symbol using regex in Perl - regex

Question

Following is my text:
Total: ₹ 131.84
Thanks for choosing Uber, Pradeep
I would like to match the amount part, using the following code:
if ( $mail_body =~ /Total: \x{20B9} (\d+)/ ) {
$amount = $1;
}
But, it does not match, tried using regex debugging, here's the output:
Compiling REx "Total: \x{20B9} (\d+)"
Final program:
1: EXACT <Total: \x{20b9} > (5)
5: OPEN1 (7)
7: PLUS (9)
8: DIGIT (0)
9: CLOSE1 (11)
11: END (0)
anchored utf8 "Total: %x{20b9} " at 0 (checking anchored) minlen 10
Matching REx "Total: \x{20B9} (\d+)" against "Total: %342%202%271%302%240131.84%n%nThanks for choosing Ube"...
UTF-8 pattern...
Match failed
Freeing REx: "Total: \x{20B9} (\d+)"
The full code is at http://pastebin.com/TGdFX7hg.

Answer 1

Disclaimer: This feels more like a comment than an answer, but I need more space.
I've never used MIME::Parser and friends before, but from what I've read in the documentation, the following might work:
use Encode qw(decode);
# according to your code, $text_mail is a MIME::Entity object
my $charset = $text_mail->head->mime_attr('content-type.charset');
my $mail_body_raw = $text_mail->bodyhandle->as_string;
my $mail_body = decode $charset, $mail_body_raw;
The idea is to get the charset from the MIME::Head object, then use Encode to decode the body accordingly.
Of course, if you know that it's always going to be UTF-8 text, you could also hardcode that:
my $mail_body = decode 'UTF-8', $mail_body_raw;
After that, your regex may still fail to work because according to the debugging output in your question the character between ₹ and the number is actually not a simple space (ASCII 32, U+0020), but a non-breaking space (U+00A0). You should be able to match that with \s:
if ( $mail_body =~ /Total: \x{20B9}\s(\d+)/ ) {

Answer 2

This is a bit of searching for the explanation, not an outright answer. Please bear with me.
I believe your $mail_body does not contain what you think it does. You posted the input data as plain text. Was that copied from a mail client?
If I take the code and the input data from the question and run it with use re 'debug' I get a different output.
use utf8;
use strict;
use warnings;
use re 'debug';
my $mail_body = qq{Total: ₹ 131.84
Thanks for choosing Uber, Pradeep};
if ( $mail_body =~ /Total: \x{20B9} (\d+)/ ) {
my $amount = $1;
}
It will produce this:
Compiling REx "Total: \x{20B9} (\d+)"
Final program:
1: EXACT <Total: \x{20b9} > (5)
5: OPEN1 (7)
7: PLUS (9)
8: POSIXU[\d] (0)
9: CLOSE1 (11)
11: END (0)
anchored utf8 "Total: %x{20b9} " at 0 (checking anchored) minlen 10
Matching REx "Total: \x{20B9} (\d+)" against "Total: %x{20b9} 131.84%n%nThanks for choosing Uber, Pradeep"
UTF-8 pattern and string...
Intuit: trying to determine minimum start position...
Found anchored substr "Total: %x{20b9} " at offset 0...
(multiline anchor test skipped)
Intuit: Successfully guessed: match at offset 0
0 <> <Total: > | 1:EXACT <Total: \x{20b9} >(5)
11 < %x{20b9} > <131.84%n%n>| 5:OPEN1(7)
11 < %x{20b9} > <131.84%n%n>| 7:PLUS(9)
POSIXU[\d] can match 3 times out of 2147483647...
14 <%x{20b9} 131> <.84%n%nTha>| 9: CLOSE1(11)
14 <%x{20b9} 131> <.84%n%nTha>| 11: END(0)
Match successful!
Freeing REx: "Total: \x{20B9} (\d+)"
Let's compare the line with the Matching REx to your output:
against against "Total: %x{20b9} 131.84%n%nThanks for choosing Uber, Pradeep"
against "Total: %342%202%271%302%240131.84%n%nThanks for choosing Ube"...
As we can see, there my output has %x{e2} and so on, while yours has %342.
When I started trying this code I forgot to put use utf8 in my code, so I got a bunch of single characters when the regex engine tried to match:
%x{e2}%x{82}%x{b9}
It then rejected the match.
So my conclusion is: Perl doesn't know your input data is utf8.