See HTML::Escape - most likely you don't want to escape stuff yourself.

For your second question, you will need to think about the encoding that the text is in (as submitted to your program, see the Content-Encoding header), and the encoding your source code is in (did you place use utf8; at the top of your script).

G'day htmanning,

Some of your regex modifiers aren't making any sense to me; for instance, /i in s/&//gi, /e in s/\n//eg, and so on. Perhaps with sample input and output that might have been clearer.

It's also unclear what you mean by "the binary code for these bullets" and why you'd need them. Does [PDF] "Unicode code chart: General Punctuation -- Range: 2000–206F" help you at all?

I put together this sample input:

$ cat pm_11148011_char_del.txt
A&
BÆ
Cæ
D
E
FìÌ
GíÍ
H•I∙J◦
K L
21°C 0.5µl 4,000Å

It doesn't show up there, but the fourth line is "D<carriage return><newline>" and the fifth is just "E<newline>".

I would probably write something like the following (pm_11148011_char_del.pl) to do what I think you want:

#!/usr/bin/env perl

use 5.014;
use warnings;
use autodie;
use utf8;
use open IO => qw{:encoding(UTF-8) :std};

my $in_file = 'pm_11148011_char_del.txt';
my %replace = (
    'í' => "'",
    'Í' => "'",
);

my $text = do {
    local $/; open my $fh, '<', $in_file; <$fh>;
};

my $new_text = $text =~ y/&ÆæìÌ\r\n•∙◦//dr
    =~ s<([^ -~])><$replace{$1} // $1>egr;

say 'Original text:';
say $text;
say 'New text:';
say $new_text;

If your Perl version is less than v5.14, you'll need to make a few adjustments. I'll assume you'll know what's required; ask if that's not the case.

Output from test run:

$ ./pm_11148011_char_del.pl
Original text:
A&
BÆ
Cæ
D
E
FìÌ
GíÍ
H•I∙J◦
K L
21°C 0.5µl 4,000Å


New text:
ABCDEFG''HIJK L21°C 0.5µl 4,000Å

I've used <pre> tags throughout so that you can see the characters instead of entities (that look like &#nnnn;).

Edit: Within <pre> tags, I needed some changes; e.g. < to &lt;, [ to &#91;, and so on. I hope I've caught them all; do let me know if I missed any and I'll fix them up also.

$text =~ s/Æ//gi;
$text =~ s/ì//gi;
[download]

This hints to me that you might be having encoding issues rather than people actually entering these characters. Since you haven't shown any of your code of how you actually get $text, I can't make any suggestions as to how you can fix it, but you should be fixing this at the source.

$text =~ s/&//gi;
...
$text =~ s/\x95/\<li\>/gi;
[download]

Are you processing the raw RSS (XML) directly? I already said it two weeks ago: Do not use regular expressions to process XML/HTML. In this case the correct solution would be to either use an XML module and only apply the regexes to the content of text nodes, or to process the text before putting it in the XML.

$text =~ s/’/'/gi;
[download]

I might suggest Text::Unidecode for these kind of replacements.

$text =~ s/\x9s/-/gi;
[download]

Use strict and warnings.

These are the bullets when copied out of the text, but using this code doesn't work:

Did you save your Perl file as UTF-8 and declare use utf8;? SSCCE.