Dear Monks, Is there a nice clean way to do a replace after match, sort of like this nice clean but non-working code:

#!/usr/bin/perl -w
use strict;

my %fixers=(
    'amazon.com' => 'danube.com',
    'ibm.com'    => 'bm.com',
    );

my $text = qq(
Blah <a href="http://amazon.com/foo">one</a>
     <a href="http://ebay.com/foo">two</a>
);

sub tweak_links($)
{
    my $text_ref = shift;
    while( $$text_ref =~ m|"http://(.*?)["/]|g ) {
        if( $fixers{$1} ) {
            $1 = $main::fixers{$1};
            }
        }
}

tweak_links(\$text);
[download]

I'm looking for something that's good when most of the matches never happen, but I can zero in on the right section with a match. In my case I even know which of the replacements is most likely on a particular text block (but all of them could happen in each block.

Is there a usual way this is done?

Keywords: perl regex replacement regular expression table-driven match multiple replacements