#!/usr/bin/perl

# use module
use XML::Simple;
use Data::Dumper;

open (ERR, ">error.txt");
# create object
$xml = new XML::Simple;

# read XML file
$data = $xml->XMLin($ARGV[0]);

# deal with the comments first
$comments = 0;
foreach $key (sort keys %{$data->{'entry'}}) {

	# all comments are "posts"
	next if !($key =~ /\.post-/);

	# but only comments have a "thr:in-reply-to" element
	$hr = $data->{'entry'}->{$key};
	next if !(exists($hr->{'thr:in-reply-to'}));

	$comments += 1;
		
#	print $hr->{'thr:in-reply-to'}->{'ref'}, "\n";

	my $d = fixDate($hr->{'published'});

	$comment{$hr->{'thr:in-reply-to'}->{'ref'}} .= <<EOF;
COMMENT:
AUTHOR: $hr->{'author'}->{'name'}
DATE: $d
EMAIL: $hr->{'author'}->{'email'}
URL: $hr->{'author'}->{'uri'}
$hr->{'content'}->{'content'}
-----
EOF
#	print $comment{$hr->{'thr:in-reply-to'}->{'ref'}}, "\n";
#	last;
}

$entries = 0;
foreach $key (sort keys %{$data->{'entry'}}) {

	next if !($key =~ /\.post-/);				# it's a post or a comment

	$hr = $data->{'entry'}->{$key};
	next if (exists($hr->{'thr:in-reply-to'}));		# already processed the comments

	$entries += 1;

#	print STDERR $key, "\n";;

	# locate the title (and URL) of the blog post

	foreach $l (@{$hr->{'link'}}) {
		if ($l->{'rel'} eq 'alternate') {
			$lr = $l;
			last;
		}
	}
	if (!defined($lr)) {
		print STDERR "ERROR - no title for $key (skipped)\n";
		next;
	}

	# $lr now points to the 'alternate' link for this post (title and URL data)

	print "AUTHOR: ",   getName($hr->{'author'}->{'name'}), "\n";
	print "DATE: ",     fixDate($hr->{'published'}), "\n";
	print "TITLE: ",    $hr->{'title'}->{'content'}, "\n";
	print "BASENAME: ", getBase($lr->{'href'}), "\n";

	# draft or published
	print "STATUS: ",   (exists($hr->{'app:control'}) && $hr->{'app:control'}->{'app:draft'} eq 'yes') ? "Draft" : "Publish", "\n";

	# miscellany
	print "ALLOW COMMENTS: 1\n";
	print "CONVERT BREAKS: 0\n";

	# get the tags, if any
	if (ref($hr->{'category'}) eq 'ARRAY') {
		$tags = "";
		foreach $r (@{$hr->{'category'}}) {
			if ($r->{'scheme'} eq 'http://www.blogger.com/atom/ns#') {
				$tags .= "," if ($tags ne "");
				if ($r->{'term'} =~ / /) {
					$tags .= '"' . $r->{'term'} . '"';
				}
				else {
					$tags .= $r->{'term'};
				}
			}
		}
		if ($tags ne "") {
			print "TAGS: $tags\n";
		}
	}

	print "-----\n";
	print "BODY:\n";

	print uncode($hr->{'content'}->{'content'}), "\n";

	print "-----\n";

	if (exists($comment{$key})) {
#		print STDERR "Comment found for $key.\n";
		print uncode($comment{$key}), "\n";
#		print $comment{$key}, "\n";
	}
	
	print "--------\n";
}

print STDERR "$entries entries, $comments comments.\n";

close(ERR);
exit;

# translate Blogger date format (YYYY-MM-DDThh:mm:ss) into MT date format (MM/DD/YYYY hh:mm:ss) 
sub fixDate {
	my $d = shift;

	# fixing 2008-01-22T16:04:00.000-08:00
	if (($d =~ /^(\d\d\d\d)-(\d\d)-(\d\d)T(\d\d:\d\d:\d\d)/) == 0) {
		print STDERR "Error parsing $d\n";
		return;
	}
	return "$2/$3/$1 $4";
}

# translate the Blogger file name into an MT basename
sub getBase {
	my $href = shift;	# e.g. http://photo.who2.com/blog/yyyy/mm/filename-with-dashes.ext

	if (($href =~ /^http:\/\/.*\/blog\/\d\d\d\d\/\d\d\/(.+)\.(txt|html)$/) == 0) {
		print STDERR "Error parsing $href\n";
		return;
	}
	return $1;
}

# Take care of any strange (Unicode) characters in the Blogger data
sub uncode {
	my $t1 = shift;
	# that pesky Unicode!
	$t1 =~ s/\x{2013}/\&ndash;/g;	# E2-80-93
	$t1 =~ s/\x{2014}/\&mdash;/g;	# E2-80-94
	$t1 =~ s/\x{2018}/\&lsquo;/g;	# E2-80-98
	$t1 =~ s/\x{2019}/\&rsquo;/g;	# E2-80-99
	$t1 =~ s/\x{201C}/\&ldquo;/g;	# E2-80-9C
	$t1 =~ s/\x{201D}/\&rdquo;/g;	# E2-80-9D
	$t1 =~ s/\x{2026}/.../g;	# E2-80-A6
	$t1 =~ s/\x{2028}/<br\/>/g;	# E2-80-A8 (line separator)
	$t1 =~ s/\x{2122}/\&trade;/g;	# E2-82-A2
	$t1 =~ s/\x{FB01}/fi/g;		# EF-AC-81
	$t1 =~ s/\x{FB02}/fl/g;		# EF-AC-82

	$t1 =~ s/\x85/.../g;		#
	$t1 =~ s/\xa3/&pound;/g;	#
	$t1 =~ s/\xa9/\&copy;/g;	#
	$t1 =~ s/\xe0/\&agrave;/g;	#
	$t1 =~ s/\xe7/\&ccedil;/g;	#
	$t1 =~ s/\xe9/\&eacute;/g;	#
	$t1 =~ s/\xea/\&ecirc;/g;	#
	$t1 =~ s/\xf1/\&ntilde;/g;	#
	$t1 =~ s/\xf3/\&oacute;/g;	#
	$t1 =~ s/\xf6/\&ouml;/g;	#
	$t1 =~ s/\xfc/\&uuml;/g;	#

	my $t2 = $t1;
	$t1 =~ s/[^ -~\n\r]//g;
	if ($t1 ne $t2) {
		my $t3 = $t2;
		$t3 =~ s/[ -~\n\r]//g;
		print ERR $lr->{'href'}, " contains ", length($t3), " non-ASCII characters ($t3).\n";
	}

	return $t1;
}

# map Blogger display names to MT usernames
sub getName {
	$n = shift;

	return "fritz" if ($n =~ /Mr. Holznagel/);
	return "paul"  if ($n =~ /Mr. Hehn/);
	return "adam"  if ($n =~ /Adam/);

	print STDERR "ERROR - Unknown author($n)\n";
	return "fritz";
}