I'm in between projects at work, so as I was wandering around http://www.c2.com, I figured I could write a script to wander for me.

You give the script a web page to start on and it wanders randomly for as many pages as you tell it to. It's fun to watch!

Use the demo and see where you end up.

Enjoy,
--Dave

update: I have refactored the original code.
update: I've added keyword functionality if you care about the source. 
#!/usr/bin/perl -wT

use strict;
use CGI;
use LWP::Simple;

$|++;

my $q = new CGI;
my $v = $q->Vars;
my $action = $q->url;

print $q->header(-type => 'text/plain'), viewSource() and exit if $v->
+{view};

print $q->header, buildTop($v, $action);

exit unless $v->{start} && $v->{depth};

$v->{depth} = 25 if $v->{depth} > 25;

my $links = [$v->{start}];
my @cycleBreaker = ();

print 'Watch!  The AutoSurfer is surfing!<ol>';

my $i = 0;
do {
    my $rand = randomizeArray($links);
    for my $url (@$rand) {
        next if grep { $_ eq $url } @cycleBreaker;

        print "<br>Getting $url..." if $v->{verbose};
        my $html = get $url;

        if ($html) {
            print "done." if $v->{verbose};
        } else {
            print "failed." if $v->{verbose};
            next;
        }

        my $title    = getTitle($html);
        my $allLinks = getLinks($html);
        $links       = validateLinks($allLinks);

        cleanLinks($links, $url);

        if (@$links) {
            print qq|<li><a target="_blank" href="$url">$title</a> - <
+a href="$action?start=$url&depth=$v->{depth}">Surf it</a>  - <font si
+ze="-1">$url</font>|;
            $i++;
            push @cycleBreaker, $url;
            last;
        }
    }
} while ($i < $v->{depth} && @$links);

print '</ol></body></html>';

sub buildMenu {
    my ($depth) = @_;
    my $options = '';
    for (5,10,15,20,25) {
        $options .= qq|<option value="$_"|;
        $options .= ' selected ' if $depth && $_ == $depth;
        $options .= ">$_";
    }
    return $options;
}

sub viewSource {
    open THIS, $0 or die $!;
    my @this = <THIS>;
    close THIS;
    return @this;
}

sub buildTop {
    my ($v, $action) = @_;

    my $options = buildMenu($v->{depth});
    my $start   = $v->{start} || 'http://';

    my $top = qq|
<html><head>
<title>AutoSurfer</title>
</head><body>
<h2>AutoSurfer</h2>
<form action="$action">
<table><tr><td>
Starting Point:
</td><td>
<input name="start" value="$start">
</td></tr><tr><td>
Surf Depth:
</td><td>
<select name="depth">
$options
</select>
</td></tr><tr><td>
Verbose:
</td><td>
<input type="checkbox" name="verbose" value="1"|;

    $top .= ' checked' if $v->{verbose};

    $top .= qq|>
</td></tr></table>
<p>
<input type="submit" value="  Go!  "> &nbsp; &nbsp; &nbsp;
<a href="$action?view=1">view source</a>
</form>|;

    return $top;
}

sub randomizeArray {
    my ($array) = @_;
    my @rand = ();
    push @rand, splice @$array, rand @$array, 1 while @$array;
    return \@rand;
}

sub getTitle {
    my ($html) = @_;
    my ($title) = $html =~ m|<\s*title\s*>([^<]+)</\s*title\s*>|is;
    $title ||= 'Untitled';
    return $title;
}

sub getLinks {
    my ($html) = @_;
    my @links  = $html =~ m|<[^>]+href=['"]?([^'" >]+)|ig;
    my @frames = $html =~ m|<frame[^>]+src=['"]?([^'" >]+)|ig;
    return [@links, @frames];
}

sub validateLinks {
    my ($links) = @_;
    my %uniq;
    @uniq{(@$links)} = undef;
    my @urls = grep {!/(dtd|gif|jpg|pdf|css|gz|mov)$/i && !/mailto:/i 
+&& $_} keys %uniq;
    return \@urls;
}

sub getDomain {
    my ($url) = @_;
    my ($domain) = $url =~ m|(http://[^/]+)|i;
    return $domain;
}

sub getPath {
    my ($path) = @_;
    $path =~ s|(?<!/)/[^/]+$||;
    return $path;
}

sub cleanLinks {
    my ($links, $url) = @_;

    my $domain = getDomain($url);
    my $path   = getPath($url);

    for (@$links) {
        next unless $_;
        unless (/^http/i) {
            if (/^\//) {
                $_ = $domain . $_;
            } elsif (s/^\.\.\///) {
                my ($new_path) = $path =~ m|^(.+)/|;
                $_ = $new_path . '/' . $_;
            } else {
                my $slash = ($path =~ /\/$/) ? '':'/';
                $_ = $path . $slash . $_;
            }
        }
    }
}

[download]

In reply to AutoSurfer cures boredom by redsquirrel

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.