it seems to me that escaping is the default behaviour of print methods and also parse and sprint ones:

print ($optional_filehandle, $optional_pretty_print_style)

    Prints an entire element, including the tags, optionally to a $opt
+ional_filehandle, optionally with a $pretty_print_style.

    The print outputs XML data so base entities are escaped.
print_to_file ($filename, %options)

    Prints the element to file $filename.

    options: see flush. =item sprint ($elt, $optional_no_enclosing_tag
+)

    Return the xml string for an entire element, including the tags. I
+f the optional second argument is true then only the string inside th
+e element is returned (the start and end tag for $elt are not). The t
+ext is XML-escaped: base entities (& and < in text, & < and " in attr
+ibute values) are turned into entities.
[download]

In contrast text method specify that entities are not escaped.

Without any code and example data i cannot tell you more.

See also XML::Twig modify data, and I don't want that L*

G'day Dimitar,

You can write a very simple function, to convert those (and other) entities back to their original characters, if that's what you need:

#!/usr/bin/env perl

use strict;
use warnings;

{
    my %char_for_ent = qw{&quot; " &apos; ' &lt; < &gt; >};
    my $re = qr/(?x: ( @{[ join '|', keys %char_for_ent ]} ) )/;
    sub ent2char { $_[0] =~ s/$re/$char_for_ent{$1}/g; $_[0] }
}

print "IN:  ${_}OUT: ", ent2char($_) while <DATA>;

__DATA__
I said, &quot;My name&apos;s Ken&quot;.
<pre>Here&apos;s some &lt;em&gt;emphasis&lt;/em&gt;.</pre>
[download]

Output:

IN:  I said, "My name's Ken".
OUT: I said, "My name's Ken".
IN:  <pre>Here&apos;s some &lt;em&gt;emphasis&lt;/em&gt;.</pre>
OUT: <pre>Here's some <em>emphasis</em>.</pre>
[download]

I suspect there may be a CPAN module with this functionality. I don't know for certain: perhaps another monk does.

Only when shown in your browser are these entities replaced on the screen by the usual ' and ".

Thank you for sharing you opinion and about this issue and giving ideas how to tackle this issue. The problem is that, printing what is outside the roots I have no control of (or at least I haven't found a good method). What I mean is that for the roots I can write a handler that will modify the output in whatever way I want (e.g., what @kcott suggest). This I cannot do for the outside of the root.

Furthermore, if I implement a handler that reads the style (which is what is actually outside of the the roots and I don't what to deal with) and simply prints it, it will be just fine.

Here is a sample input:

<style type="text/css">
  @font-face {
    font-family: 'MyFont';
    src: url('http://mywebsite/fonts/MyFont.otf');
  }
</style>
<div>
    This is a test. Let's go. This is a brand name: 'D&G'. And this is
+ "a test".
</div>
[download]

Here are the two example scripts and output:

my $t= XML::Twig->new(     twig_print_outside_roots => 1,
                              twig_roots      =>  {'div' => sub {
                                               my ( $t, $e ) = @_;
                                               print STDOUT $e->text;
                                               $t->purge;
                                               },
                                               'style' => sub {
                                               my ( $t, $e ) = @_;
                                               $e->print();
                                               $t->purge;
                                               } 
                           },
                           keep_atts_order => 1,
                      );
[download]

and

my $t= XML::Twig->new(     twig_print_outside_roots => 1,
                              twig_roots      =>  {'div' => sub {
                                               my ( $t, $e ) = @_;
                                               print STDOUT $e->text;
                                               $t->purge;
                                               },
                           },
                           keep_atts_order => 1,
                      );
[download]

And this is the output from the first and then from the second version of the script.

<html><head></head><body>´&#9559;&#9488;<style type="text/css">

  @font-face {
    font-family: 'MyFont';
    src: url('http://mywebsite/fonts/MyFont.otf');
  }


</style>


        This is a test. Let's go. This is a brand name: 'D&G'. And thi
+s is "a test".
</body></html>
[download]

And

<html><head></head><body>´&#9559;&#9488;<style type="text/css">

  @font-face {
    font-family: &apos;MyFont&apos;;
    src: url(&apos;http://mywebsite/fonts/MyFont.otf&apos;);
  }


</style>


        This is a test. Let's go. This is a brand name: 'D&G'. And thi
+s is "a test".
</body></html>
[download]

Than's a lot. Greetings,
Dimitar

What is wrong with the first version output, it didn't convert any & or '" to entities , so its working, right?

Have you tried output_filter of html?

If you can post a 20 line program which demonstrates the problem, I'll take a stab at fixing it :)

Hello,

I tried output_filter as well as input_filter - nothing. They apply on the roots, but not on the outside content.

Cheers, Dimitar.