 
#the rest (header and footer) are Zwiki specific 
#that is why we just extract this div and just this one
	    cleanup( $tmp, $src ) ;

	    if (! -e $dst."txt" ) {
		if (! copy($tmp, $dst)) {
		    warn("Can't copy file $tmp to $dst: $!\n");
		}
		else {

		    print "Processing ",$dst,".txt\n";			
		    open FILE, ">$dst.txt";
		    print FILE $wc->html2wiki( file => $dst );
		    close FILE;
		    unlink $dst;
		    unlink $tmp;
		}
	    }
	}

    }
} 


sub cleanup {
# this sub will read a html file
# extract the div class=content 
# and create a clean html page with UTF 8 header

# get the parm given to  cleanup ( $tmp , $src) ;

    my ($filetmp,$filesrc) = @_;

#read the source file
    my $tree = HTML::TreeBuilder->new();
    $tree->parse_file($filesrc);


    my $root = $tree->find_by_tag_name('html');

#make the temp file
    open SRC, ">$filetmp";
    print SRC  "\n";

    print SRC "\t\n";
    print SRC "\t\t\n";
    print SRC "\t\n";

#read the body content
    foreach my $body ($root->find_by_tag_name('body')) {
	print SRC "\t\n";

# search the div
	foreach my $divs ($body->find_by_tag_name('div')) {

#isolate the

 
	    foreach my $div ($divs->find_by_attribute('class','content')) {

#if we find a form in the div ....
		foreach my $forms ($div->find_by_tag_name('form')) {
#... we remove it
		    $forms->detach;
		}

		foreach my $uls ($div->find_by_tag_name('ul') ) {
		    foreach my $lis ($uls->find_by_tag_name('li') ) {
			foreach my $hrefs ($lis->find_by_tag_name('a') ) {

			    my $links = $hrefs->attr_get_i('href');
			    my ($content) = $hrefs->content_list();

			    if ( $links !~ /^(http|ftp|file)/) {
			    	# if the link contain things like 
				# AdminWiki:FrontPage
				# we replace the / (slash) by : (semi column) 
				# thus Dokuwiki will "understand" the content of the href.
				$links =~ s/\//:/g;

# ~literal is the only way to 'rebuld' a clean link from scratch
# because as_HTML encode < and > to > $lt;
				my $new_content = HTML::Element->new('~literal',
					'text' => ''.$content.''
					);
				#we replace the content no well formed for dokuwiki
				#by this new one
				$hrefs->replace_with($new_content);
			    }
			}
		    }
		}

# we add the rest of the file to the temp file
		print SRC  $div->as_HTML();
	    }
	}
# end of body
	print SRC  "\t\n";
    }
# end of the html page and close the tmp file
    print SRC "";
    close SRC;
}

=====Installation in Dokuwiki ===== copy of the working directory to a Dokuwiki installation (which can be done in step 2 if you give the final directory on the command lines