A simplistic view of what constitutes a legitimate identifier is any name that matches the following regexp:

m/\A\[_\p{ID_Start}]\p{XID_Continue}*\z/
[download]

But this has many shortcomings, some of which are described in a StackOverflow article from tchrist (When I read his articles dealing with Unicode I mostly just feel small). It doesn't allow for most punctuation variables described in perlvar, for example... but even discounting those, it's just a start.

I mention this because the symbols you are using in your identifier will not match that regular expression, despite ID_Start and XID_Continue matching well over a hundred thousand characters. And these aren't punctuation characters described in perlvar either. So for typical use you can probably disqualify them outright.

One thing of note is that when wielding symbolic refs, nearly any set of characters is legal, including your currency symbols. Case in point:

use utf8;
binmode STDOUT, ':utf8';

no strict 'refs';
my $symbol = '£1%^@"₯';

$$symbol = 42;

print "\$$symbol = $$symbol\n";
print "$symbol is found in package ", *{__PACKAGE__ . '::£1%^@"₯'}{PAC
+KAGE}, "\n";
print "\$main::${symbol} has a value of ", ${'main::£1%^@"₯'}, "\n";
[download]

Spoiler alert, it works. And notice we're using the £ and ₯ characters.

Symbol manipulation is something Perl is good at, but that we try to avoid for our own sanity. A language that more fully embraces it is Scheme. And since symbol manipulation is the name of the game for Scheme, naturally that language is fine with ₯ as a function identifier:

(define ₯ (lambda (x y) (+ x y)))

;Value: |β₯|

1 ]=> (₯ 3 5)

;Value: 8
[download]

Thanks davido.

This is the kind of information what I was looking for. Let me add this link http://www.unicode.org/reports/tr31/tr31-31.html

Hi, not all characters may be used in variable names, even under UTF-8. See perldata.

Try this:

#!/usr/bin/perl
use strict;
use warnings;
use diagnostics;
use utf8;
use feature qw/say/;
binmode STDOUT, ":utf8";

my $ιlan;
say '------------------------';
[download]

Hope this helps!

Hi 1nickt,

It seems to be that unicode characters can be in variable names if they are letters and currency symbols are not considered as letters.

my $愛 = 4; # correct

my $s = '★'; # correct

my $★ = 'star'; # incorrect

However, it makes sense. What could we do with this?

sub $_to_₯ { ... }
[download]

Update: This one works ...

my $元 = 3;

Hi again, I'm not sure whether What could we do with this? refers to the sub after the question, or the way Perl disallows certain characters in variable names. If the latter, I would suggest that independent of your variable naming issue, you consider extensibility of your program. There doesn't seem to be much advantage to a sub named $_to_₯ as compared to a more generic one that takes arguments:

use strict;
use warnings;
use Test::More;
use Test::More::UTF8; # loads utf8.pm

sub convert_currency {
    my %map = (
        '$₯' => sub { $_[0] * 105.4 },
        '₯$' => sub { $_[0] * 0.0095 },
        '$€' => sub { $_[0] * 0.89 },
        '€$' => sub { $_[0] * 1.12 },
    );

    my ($from, $to, $amount) = @_;

    my $converter = $map{"$from$to"};

    return sprintf( '%.3f', $converter->($amount) );
}

is( convert_currency('$', '₯', 5),     '527.000', '$ -> ₯' );
is( convert_currency('€', '$', 42.02),  '47.062', '€ -> $' );
is( convert_currency('$', '€', 42.02),  '37.398', '$ -> €' );

done_testing;
[download]

$ prove -v monks/11104981.pl

monks/11104981.pl ..
ok 1 - $ -> ₯
ok 2 - € -> $
ok 3 - $ -> €
1..3
ok
All tests successful.
Files=1, Tests=3,  0 wallclock secs ( 0.02 usr  0.00 sys +  0.04 cusr 
+ 0.00 csys =  0.06 CPU)
Result: PASS
[download]

Hope this helps!



The way forward always starts with a minimal test.

While other monks have mentioned the restrictions on variable names and recommended a single conversion function that takes arguments to specify a currency pair, I would suggest using standard currency codes instead of Unicode symbols. The standard codes are all in ASCII, and that has the advantage that a person who does not recognize a particular code can still read it, while an unknown symbol is just "what is that supposed to mean?".

To write this code, your text editor must understand UTF-8 and you must save the file with the appropriate encoding. Again, any two programs which communicate with Unicode data must agree on the encoding of that data.

hi holli,

I think this note is about how to keep the encoding in the saved file.