#!C:\perl\bin\perl.exe -w
use strict ;

use File::Find ;

#my $dirname = 'C:\temp\dotrose\html\etext\sermons' ;
my $dirname = '.' ;

my $head = '<?php 
	$page_title = "" ;
	include( "skin.php" ) ;
	include( $skin . "/header.php" ) ;
?>

' ;

my $foot = '

</div>

<div id="menu"><?php display_menu( "menu.php" ) ; ?></div>

<?php include( $skin . "/footer.php" ) ; ?>
' ;


find(\&process_directory, $dirname) ;


sub process_directory {

		return 1 unless ( $_ =~ /html?$/ ) ;
		print "\tProcessing File: $_\n" ;

		my ( $infile, $outfile ) ;
		$infile = $outfile = $_ ;
		$outfile =~ s/html?$/php/i ;
		
		open (IN,	"<$infile")		or die "Problem opening $infile: $!\n" ;
		open (OUT,	">$outfile")	or die "Problem opening $outfile: $!\n" ;

		my $out = '' ;
		
		# Make the file one long string
		while ( my $line = <IN> ) {
			# Remove the end of line character
			chomp $line ;
			# Replace it with a blank space
			$line .= ' ' ;
			# And append it to the rest of the document
			$out .= $line ;
		}

		# REGEXes go here to change what wants changing
		$out =~ s/<p[^>]*>/<p>/ig ;
		$out =~ s/<h([0-9])[^>]*>/<h$1>/ig ;
		$out =~ s/<\/h([0-9]>)/<\/h$1\n/ig ;
		$out =~ s/<\/?font[^>]*>//ig ;
		$out =~ s/<\/?meta[^>]*>//ig ;
		$out =~ s/<\/?span[^>]*>//ig ;
		$out =~ s/<\/?small[^>]*>//ig ;
		$out =~ s/<\/?large[^>]*>//ig ;
		$out =~ s/<img[^>]*>//ig ;
		$out =~ s/<hr[^>]*>//ig ;
		$out =~ s/\s<\/strong>(\w)/<\/strong> $1/gi ;
		$out =~ s/<(\/?)i>/<$1em>/ig ;
		$out =~ s/\s<\/em>(\w)/<\/em> $1/gi ;
		$out =~ s/<em><\/em>//ig ;
		$out =~ s/<strong><\/strong>//ig ;
		$out =~ s/<u><\/u>//ig ;
		$out =~ s/<(\/?)b>/<$1strong>/ig ;
		$out =~ s/<\/?o:[^>]*>//ig ;
		$out =~ s/<\/?html[^>]*>//ig ;
		$out =~ s/<\/?head[^>]*>//ig ;
		$out =~ s/<\/?body[^>]*>//ig ;
		$out =~ s/<\/?title[^>]*>/\n/ig ;
		$out =~ s/style="[^"]*"//ig ;
		$out =~ s/<\!DOCTYPE[^>]*>// ;
		$out =~ s/\.html?/.php/ig ;
		$out =~ s/<br>\w<\/p>/<\/p>/ig ;
#		$out =~ s/href="http:\/\/.*\/([a-zA-Z0-9%]*\.php)"/href="$1"/ig ;
		$out =~ s/http:\/\/www.jonathanedwards.com\///gi ;

		# Remove multiple spaces throughout document
		$out =~ s/\s\s*/ /g ;
		
		# Finally, add back some "end of line" characters
		$out =~ s/(<p>)/<\/p>\n\n$1/ig ;
		$out =~ s/<p>\s*/<p>/ig ;
		$out =~ s/\s*<\/p>/<\/p>/ig ;
		$out =~ s/<\/p><\/p>/<\/p>/ig ;
		$out =~ s/(<\/div>)/$1\n/ig ;
		$out =~ s/<br>(.)/<br>\n$1/g ;

		# If any lines start with whitespace, remove it
		$out =~ s/^\s// ;
		
		print OUT $head		or die "Printing Head: $!\n" ;
		print OUT $out		or die "Printing: $!\n" ;
		print OUT $foot		or die "Printing Foot: $!\n" ;

		close(OUT)			or die "Closing: $!\n" ;
		close(IN)			or die "Closing: $!\n" ;

		unlink $infile		or die "Could Not Delete: $infile: $!\n" ;

	return 1 ;
		
}