#!/usr/bin/perl
# IBM(c) 2007 EPL license http://www.eclipse.org/legal/epl-v10.html

# Downloads/converts the xCAT docs on the sourceforge wiki to local HTML and PDF.
# This script is not dependent on other xCAT code, so you can copy it to a machine
# that has internet access to run it.  Before running this command, you must have
# wget, python, and pisa installed.  See: https://sourceforge.net/apps/mediawiki/xcat/index.php?title=Editing_xCAT_Documentation_Pages#Converting_Wiki_Pages_to_HTML_and_PDFs .

#BEGIN
#{
#    $::XCATROOT = $ENV{'XCATROOT'} ? $ENV{'XCATROOT'} : -d '/opt/xcat' ? '/opt/xcat' : '/usr';
#}
use strict;
#use lib "$::XCATROOT/lib/perl";
#use xCAT::Utils;
use Getopt::Long;
#use Data::Dumper;

#my $VERSION;
my $HELP;
 
my $usage = sub {
   	my $exitcode = shift @_;
   	print "Usage: getxcatdocs [-?|-h|--help] [<destination-dir>]\n";
	exit $exitcode;
};

# Process the cmd line args
Getopt::Long::Configure("bundling");
#Getopt::Long::Configure("pass_through");
Getopt::Long::Configure("no_pass_through");
if (!GetOptions('h|?|help'  => \$HELP )) { $usage->(1); }

if ($HELP) { $usage->(0); }

#if ($VERSION) {
    #print xCAT::Utils->Version(), "\n";
#    exit;
#}

#if (xCAT::Utils->isAIX()) { die "Error: this command is not yet supported on AIX.\n"; }
if ($^O =~ /^aix/i) { die "Error: this command is not yet supported on AIX.\n"; }

my $destdir = scalar(@ARGV) ? $ARGV[0] : '.';
chdir($destdir);

# Download the HTML
mkdir('html');
chdir('html');
print "Downloading the xCAT wiki documentation...\n";
my $wgetcmd = q(wget -rk -nv --progress=bar -l 1 --page-requisites -np -nH --cut-dirs=3 -e robots=off --post-data='printable=yes' --reject '*title=Special:*,*title=Talk:*,*title=-&*,*title=HowTos,*title=Main_Page,*title=Release_Notes,*title=Wish_List_for_xCAT_2,*&action=edit*,*&action=history*,*&printable=yes*,*&oldid=*,index.html,opensearch_desc.php,xcat' 'https://sourceforge.net/apps/mediawiki/xcat/index.php?title=XCAT_Documentation');
runwget($wgetcmd);

# Get the list of files and convert to pdf
opendir(DIR, '.') or die "Error: could not read the just created html directory.\n";
my @dir = grep /^index.php\?title=/, readdir(DIR);		# /
close(DIR);
mkdir('../pdf');
chdir('../pdf');
if (system('which xhtml2pdf >/dev/null 2>&1')) { die "xhtml2pdf is not installed.  See http://sourceforge.net/apps/mediawiki/xcat/index.php?title=Editing_xCAT_Documentation_Pages#Converting_Wiki_Pages_to_HTML_and_PDFs .\n"; }

foreach my $file (@dir) {
	if ($file =~ /^index.php\?title=MediaWiki:/ || $file eq 'index.php?title=XCAT_Documentation') { next; }
	my ($docname) = $file =~ /^index.php\?title=(.+)$/;
	print "Converting $docname to PDF format...\n";
	my $url = "https://sourceforge.net/apps/mediawiki/xcat/$file&printable=yes";
	my $destfile = "$docname.pdf";
	my $cmd = "xhtml2pdf '$url' '$destfile' ";
	#print "$cmd\n";
	runh2p($cmd);
}

# Run the wget cmd and filter out some of the silly output
sub runwget {
	my $cmd = shift;
	#print "$cmd\n";
	open(OUT, "$cmd 2>&1 |") || die "can't fork $cmd: $!\n";
	while (<OUT>) {
		if (/URL:https*:\/\/sourceforge\.net.+\s+->\s+\"(\S+)\"\s+\[/) { print "Downloaded $1.\n"; }
		else { print; }
	}
	close OUT || die "Error running $cmd: $! $?";
}

# Run the xhtml2pdf cmd and filter out some of the silly output
sub runh2p {
	my $cmd = shift;
	#print "$cmd\n";
	open(OUT, "$cmd 2>&1 |") || die "can't fork $cmd: $!\n";
	while (<OUT>) {
		next if /DeprecationWarning:\sthe sets module is deprecated/;
		next if /from sets import ImmutableSet/;
		next if /^Converting\ssourceforge.net/;
		print;
	}
	close OUT || die "Error running $cmd: $! $?";
}