Perl - Unicode::String sub need to add/convert for Latin-9 support

Question

Part 3 (Part 2 is here) (Part 1 is here)

Here is the perl Mod I'm using: Unicode::String

How I'm calling it:

print "Euro: ";
print unicode_encode("€")."\n";
print "Pound: ";
print unicode_encode("£")."\n";

would like it to return this format:

€ # Euro
£ # Pound

The function is below:

sub unicode_encode {

    shift() if ref( $_[0] );
    my $toencode = shift();
    return undef unless defined($toencode);

    print "Passed: ".$toencode."\n";

    Unicode::String->stringify_as("utf8");
    my $unicode_str = Unicode::String->new();
    my $text_str    = "";
    my $pack_str    = "";

    # encode Perl UTF-8 string into latin1 Unicode::String
    #  - currently only Basic Latin and Latin 1 Supplement
    #    are supported here due to issues with Unicode::String .
    $unicode_str->latin1($toencode);

    print "Latin 1: ".$unicode_str."\n";

    # Convert to hex format ("U+XXXX U+XXXX ")
    $text_str = $unicode_str->hex;

    # Now, the interesting part.
    # We must search for the (now hex-encoded)
    #       Unicode escape sequence.
    my $pattern =
'U\+005[C|c] U\+0058 U\+00([0-9A-Fa-f])([0-9A-Fa-f]) U\+00([0-9A-Fa-f])([0-9A-Fa-f]) U\+00([0-9A-Fa-f])([0-9A-Fa-f]) U\+00([0-9A-Fa-f])([0-9A-Fa-f])';

    # Replace escapes with entities (beginning of string)
    $_ = $text_str;
    if (/^$pattern/) {
        $pack_str = pack "H8", "$1$2$3$4$5$6$7$8";
        $text_str =~ s/^$pattern/\&#x$pack_str/;
    }

    # Replace escapes with entities (middle of string)
    $_ = $text_str;
    while (/ $pattern/) {
        $pack_str = pack "H8", "$1$2$3$4$5$6$7$8";
        $text_str =~ s/ $pattern/\;\&#x$pack_str/;
        $_ = $text_str;
    }

    # Replace "U+"  with "&#x"      (beginning of string)
    $text_str =~ s/^U\+/&#x/;

    # Replace " U+" with ";&#x"     (middle of string)
    $text_str =~ s/ U\+/;&#x/g;

    # Append ";" to end of string to close last entity.
    # This last ";" at the end of the string isn't necessary in most parsers.
    # However, it is included anyways to ensure full compatibility.
    if ( $text_str ne "" ) {
        $text_str .= ';';
    }

    return $text_str;
}

I need to get the same output but need to Support Latin-9 characters as well, but the Unicode::String is limited to latin1. any thoughts on how I can get around this?

I have a couple of other questions and think I have a somewhat understanding of Unicode and Encodings but having time issues as well.

Thanks to anyone who helps me out!

Answer 1

As you have been told already, Unicode::String is not an appropriate choice of module. Perl ships with a module called 'Encode' which can do everything you need.

If you have a character string in Perl like this:

my $euro = "\x{20ac}";

You can convert it to a string of bytes in Latin-9 like this:

my $bytes = encode("iso8859-15", $euro);

The $bytes variable will now contain \xA4.

Or you can have Perl automatically convert it out output to a filehandle like this:

binmode(STDOUT, ":encoding(iso8859-15)");

You can refer to the documentation for the Encode module. And also, PerlIO describes the encoding layer.

I know you are determined to ignore this final piece of advice but I'll offer it one last time. Latin-9 is a legacy encoding. Perl can quite happily read Latin-9 data and convert it to UTF-8 on the fly (using binmode). You should not be writing more software that generates Latin-9 data you should be migrating away from it.