#!/usr/bin/perl -w
# $Id: whoisnet,v 1.3 2003/02/14 15:39:27 candy Exp candy $
use strict;
use POSIX;
use Socket;
use Getopt::Std;

my $HOME = $ENV{HOME};
my $quit = 0;
my $debug;
my $hosts_file = "$HOME/bin/whois.db";
my $hosts_log_file = "$HOME/bin/whois.log";
my $read_log_mode = 0;
my $wait_sec = 10;
my $did_whois = 0; # whois ¼Â¹Ô¤·¤¿


sub nz($) {
	return defined($_[0]) ? $_[0] : "";
}

sub p(@) {
	my @v = @_;
	my $i;
	for ($i = 0; $i <= $#v; $i++) {	
		printf("%d %s\n", $i, $v[$i]);
	}
}

sub tobin($$) {
	my($x, $col) = @_;
	my($s, $i);
	$s = "";
	for ($i = 0; $i < $col; $i++) {
		$s = (($x & 1) ? "1" : "0") . $s;
		$x >>= 1;
	}
	return $s;
}

sub tobin8($) {
	my($x) = @_;
	my $i;
	my $s = "";
	for ($i = 0; $i < 4; $i++) {
		$s = ":" . tobin($x & 0xff, 8) . $s;
		$x >>= 8;
	}
	$s =~ s/.//;
	return $s;
}

sub inet_addr($) {
	my($ipstr) = @_;
	my(@v, $i, $s_addr);
	@v = split(/\./, $ipstr . ".0.0.0.0");
	$s_addr = 0;
	for ($i = 0; $i < 4; $i++) {
		$s_addr = ($s_addr * 256) + $v[$i];
	}
	return $s_addr;
}

sub s_addr_to_v($) {
	my($s_addr) = @_;
	my(@v);
	$v[0] = ($s_addr >> 24) & 0xff;
	$v[1] = ($s_addr >> 16) & 0xff;
	$v[2] = ($s_addr >> 8) & 0xff;
	$v[3] = $s_addr & 0xff;
	return @v;
}

sub s_addr_to_dec($) {
	my($s_addr) = @_;
	my(@v) = s_addr_to_v($s_addr);
	return sprintf("%d.%d.%d.%d", $v[0], $v[1], $v[2], $v[3]);
}

my @net_lo;
my @net_hi;
my @net_co;
my $net_used = 0;
my $tab_changed = 0;

sub net_add($$$) {
	my($blo, $bhi, $co) = @_;
	$net_lo[$net_used] = $blo;
	$net_hi[$net_used] = $bhi;
	$net_co[$net_used] = $co;
	$net_used++;
	return $net_used;
}

sub net_lookup($) {
	my($ip) = @_;
	my($i) = 0;
	my($found) = -1;
	while ($found == -1 && $i < $net_used) {
		if ($ip >= $net_lo[$i] && $ip <= $net_hi[$i]) {
			$found = $i;
		}
		$i++;
	}
	return $found;
}

sub read_hosts($) {
	my($file, $line, @v) = @_;
	if (open(F, $file)) {
		while ($line = <F>) {
			if ($line =~ /^$|^#/) {
				# ignore
			}
			else {
				chomp($line);
				@v = split(/[ \t]+/, $line);
				if (nz($v[2]) ne "") {
					net_add(inet_addr($v[0]), inet_addr($v[1]), $v[2]);
				}
			}
		}
		close(F);
	}
	else {
		printf(STDERR "%s: $!\n", $file);
	}
}

sub write_hosts($) {
	my($file, $i, $ip) = @_;
	if (open(F, ">$file")) {
		my(%ptr_tabx);
		for ($i = 0; $i < $net_used; $i++) {
			$ptr_tabx{$net_lo[$i]} = $i;
		}
		foreach $ip (sort {$a <=> $b} @net_lo) {
			my($lo, $hi, $co);
			$i = $ptr_tabx{$ip};
			$lo = s_addr_to_dec($net_lo[$i]);
			$hi = s_addr_to_dec($net_hi[$i]);
			$co = $net_co[$i];
			printf(F "%s %s %s\n", $lo, $hi, $co);
		}
		close(F);
		$tab_changed = 0;
	}
	else {
		printf(STDERR "%s: $!\n", $file);
	}
}

sub cmd($) {
	my($cmd) = @_;
	my(@v);
	if (open(F, "$cmd")) {
		my ($i);
		@v = <F>;
		close(F);
		for ($i = 0; $i <= $#v; $i++) {	
			chomp($v[$i]);
		}
	}
	else {
		logger("$cmd: $!\n");
	}
	return @v;
}

sub getval($@) {
	my($k, @v) = @_;
	my $ret;
	my @ni = grep(/$k/, @v);
	if ($#ni >= 0) {
		my $x = $ni[0];
		$x =~ s/[ \t]+(.*)//;
		$ret = $1;
	}
	return $ret;
}

sub whois($) {
	my($ip) = @_;
	my(@v) = cmd("whois $ip |");
	$did_whois = 1;
	if ($#v == -1) {
		$quit = 1;
	}
	else {
		if (open(LOG, ">>$hosts_log_file")) {
			my($i);
			printf(LOG ">>> %s\n", $ip);
			printf(LOG "### %s\n", strftime("%Y/%m/%d %H:%M:%S", localtime()));
			for ($i = 0; $i <= $#v; $i++) {
				printf(LOG "%s\n", $v[$i]);
			}
			close(LOG);
		}
		else {
			printf(STDERR "%s: $!\n", $hosts_log_file);
		}
	}
	return @v;
}

sub whois_from_log($) {
	my($ip) = @_;
	my(@v);
	if (open(LOG, $hosts_log_file)) {
		my($lbuf, $found, $head);
		$found = -1;
		$head = ">>> $ip\n";
		# search header
		while (defined($lbuf = <LOG>) && $lbuf ne $head) {
		}
		if (defined($lbuf)) {
			while (defined($lbuf = <LOG>) && $lbuf !~ /^>>> /) {
				chomp($lbuf);
				push(@v, $lbuf);
			}
		}
		close(LOG);
	}
	else {
		printf(STDERR "%s: $!\n", $hosts_log_file);
	}
	return @v;
}

sub do_whois($) {
	my($ip) = @_;
	my(@v);
	if ($read_log_mode) {
		@v = whois_from_log($ip);
	}
	else {
		@v = whois($ip);
	}
	return @v;
}


sub parse_net($) {
	my($range) = @_;
	my($blo, $bhi, $lo, $hi);
	($lo = $range) =~ s/[ \t]*-.*//;
	($hi = $range) =~ s/.*-[ \t]*//;
	$blo = inet_addr($lo);
	$bhi = inet_addr($hi);
	return ($blo, $bhi);
}

sub lookup_inetnum(@) {
	my(@v) = @_;
	my($net, $co);
	my @x = grep(/^inetnum:/, @v);
	if ($#x >= 0) {
		my $i = 0;
		my $found1 = -1;
		my $found2 = -1;
		for ($i = 0; $i <= $#v; $i++) {
			if ($v[$i] =~ /^inetnum:/) {
				$found1 = $i;
			}
			if ($found1 >= 0 && $found2 == -1 && $v[$i] =~ /^country:/) {
				$found2 = $i;
			}
		}
		if ($found1 >= 0 && $found2 >= 0) {
			($net = $v[$found1]) =~ s/^inetnum:[ \t]*//;
			($co = $v[$found2]) =~ s/^country:[ \t]*//;
		}
	}
	return ($net, $co);
}

sub lookup_ARIN(@) {
	my(@v) = @_;
	my($net, $co);
	my @x = grep(/^# ARIN Whois/, @v);
	if ($#x >= 0) {
		my $i = 0;
		my $found = -1;
		for ($i = 0; $i <= $#v; $i++) {
			if ($v[$i] =~ /\d+\.\d+\.\d+\.\d+[ \t]*-[ \t]*\d+\.\d+\.\d+\.\d+/) {
				$found = $i;
			}
		}
		if ($found >= 0) {
			($net = $v[$found]) =~ s/^[ \t]+//;
			$co = nz($v[$found - 1]);
		}
	}
	return ($net, $co);
}

sub count_mask($$) {
	my($bip, $x) = @_;
	my($i) = 0;
	my($bw) = 32;
	while ($i < $bw && ($bip & 1) == $x) {
		$bip = $bip >> 1;
		$i++;
	}
	return $bw - $i;
}

my @bitmask;

sub bitmask_setup() {
	my $i;
	my $bm = 0;
	for ($i = 0; $i < 32; $i++) {
		$bitmask[$i] = $bm;
		$bm = ($bm >> 1) | 0x80000000;
	}
	$bitmask[32] = 0xffffffff;
}

sub netaddr($$) {
	my($bip, $mlen) = @_;
	return $bip & $bitmask[$mlen];
}

sub broaddr($$) {
	my($bip, $mlen) = @_;
	return ($bip & $bitmask[$mlen]) | ~$bitmask[$mlen];
}

sub find_split_point($$$) {
	my($blo, $bhi, $l0) = @_;
	while (broaddr($blo, $l0) > $bhi) {
		$l0++;
	}
	return $l0;
}

sub range_to_cidr($$$$);

sub range_to_cidr($$$$) {
	my($blo, $bhi, $lvl, $co) = @_;
	my($l0, $l1, $mlen);
	my($bnet, $bbro);
	#printf("%d>>%s-%s\n", $lvl, s_addr_to_dec($blo), s_addr_to_dec($bhi));
	$l0 = count_mask($blo, 0);
	$l1 = count_mask($bhi, 1);
	#printf("%s/%d\n", tobin8($blo), $l0);
	#printf("%s/%d\n", tobin8($bhi), $l1);
	$mlen = ($l0 > $l1 ? $l0 : $l1);
	$bnet = netaddr($blo, $mlen);
	$bbro = broaddr($blo, $mlen);
	#printf("net:%s/%d\n", s_addr_to_dec($bnet), $mlen);
	#printf("bro:%s/%d\n", s_addr_to_dec($bbro), $mlen);
	if ($blo != $bnet || $bhi != $bbro) {
		my $split_mlen = find_split_point($blo, $bhi, $l0);
		my $split_bbro = broaddr($blo, $split_mlen);
		#printf("split1:%s/%d\n", s_addr_to_dec($split_bbro), $split_mlen);
		#printf("split2:%s-%s\n", s_addr_to_dec($split_bbro+1), s_addr_to_dec($bhi));
		range_to_cidr($blo, $split_bbro, $lvl + 1, $co);
		range_to_cidr($split_bbro + 1, $bhi, $lvl + 1, $co);
	}
	else {
		my $anet = s_addr_to_dec($bnet);
		my $abro = s_addr_to_dec($bbro);
		printf("%s/%d - %s %s\n", $anet, $mlen, $abro, $co);
	}
}
sub net_to_cidr($$) {
	my($blo, $bhi) = @_;
	my($i) = 0;
	my($bm) = 0x80000000;
	while ($bm != 0 && ($blo & $bm) == ($bhi & $bm)) {
		$bm = 0x80000000 | ($bm >>= 1);
		$i++;
	}
	$bm = ($bm << 1) & 0xffffffff;
	my $net = $blo & $bm;
	my $bro = ($blo & $bm) | ~$bm;
	if ($net != $blo || $bro != $bhi) {
		# not CIDR
		$i = -$i;
	}
	return $i;
}

sub idx_to_cidr($) {
	my($i) = @_;
	my($blo, $bhi, $co, $lo, $hi, $mask);
	$blo = $net_lo[$i];
	$bhi = $net_hi[$i];
	$co = $net_co[$i];
	$lo = s_addr_to_dec($blo);
	$hi = s_addr_to_dec($bhi);
	$mask = net_to_cidr($blo, $bhi);
	return sprintf("%s/%d %s %s", $lo, $mask, $hi, $co);
}


sub whereis($) {
	my($ip) = @_;
	my($bip) = inet_addr($ip);
	my ($idx) = net_lookup($bip);
	my $co;
	if ($idx == -1) {
		my @v = do_whois($ip);
		my $net;
		($net, $co) = lookup_inetnum(@v);
		if (!defined($net) || !defined($co)) {
			$net = getval("^NetRange:", @v);
			$co = getval("^NetName:", @v);
		}
		if (!defined($net) || !defined($co)) {
			($net, $co) = lookup_ARIN(@v);
		}
		if (defined($net) && defined($co)) {
			my($blo, $bhi) = parse_net($net);
			$idx = net_add($blo, $bhi, $co);
			$tab_changed++;
		}
	}
	else {
		$co = $net_co[$idx];
	}
	if ($::opt_v) {
		if ($idx >= 0) {
			$co = $ip . " " . idx_to_cidr($idx);
		}
	}
	return $co;
}

sub nain($$) {
	my($type, $idx) = @_;
	my($line);
	read_hosts($hosts_file);
	while ($quit == 0 && defined($line = <>)) {
		my($i, @v, $host, $fqdn);
		$did_whois = 0;
		chomp($line);
		@v = split(/[ \t]+/, $line);
		$host = nz($v[$idx]);
		$fqdn = whereis($host);
		if (defined($fqdn)) {
			$v[$idx] .= " ". $fqdn;
		}
		$i = join(" ", @v);
		printf("%s\n", $i);
		if ($tab_changed) {
			write_hosts($hosts_file);
		}
		if ($wait_sec && $quit == 0 && $did_whois) {
			sleep($wait_sec);
		}
	}
	return 0;
}

sub whereis_to_cidr() {
	read_hosts($hosts_file);
	my($i, $blo, $bhi, $co, $lo, $hi, $mask);
	for ($i = 0; $i < $net_used; $i++) {
		my($blo, $bhi, $co, $lo, $hi, $mask);
		$blo = $net_lo[$i];
		$bhi = $net_hi[$i];
		$co = $net_co[$i];
		$lo = s_addr_to_dec($blo);
		if ($::opt_v) {
			printf(">%s-%s %s\n", s_addr_to_dec($blo), s_addr_to_dec($bhi), $co);
			printf("%s\n", idx_to_cidr($i));
		}
		range_to_cidr($blo, $bhi, 0, $co);
	}
	return 0;
}

my $usage =
	"Version 0.8087\n"
	."usage: %s [-pL][-f whois.db][-l whois.log][-w wait] [file]\n"
	."\t-p\twhois.db to CIDR\n"
	."\t-L\tRead log instead of fork(2)ing whois(1)\n"
	."\t-f whois.db\tSpecify db\n"
	."\t-l whois.log\tSpecify log\n"
	."\t-w n\tWait n sec\n"
	;

sub main() {
	my($ex,  $type, $idx);
	$ex = 1;
	getopts('f:l:Lpw:vV');
	nz($::opt_f);
	nz($::opt_l);
	nz($::opt_L);
	nz($::opt_p);
	nz($::opt_v);
	nz($::opt_V);
	nz($::opt_w);
	if (nz($::opt_V)) {
		printf(STDERR $usage, $0);
	}
	else {
		if (nz($::opt_f) ne "") {
			$hosts_file = $::opt_f;
		}
		if (nz($::opt_l) ne "") {
			$hosts_log_file = $::opt_l;
		}
		if (nz($::opt_L)) {
			$read_log_mode = 1;
		}
		if (nz($::opt_w)) {
			$wait_sec = $::opt_w;
		}
		$type = 0; # apache
		$idx = 0;
		bitmask_setup();
		if (nz($::opt_p)) {
			$ex = whereis_to_cidr();
		}
		else {
			$ex = nain($type, $idx) < 0;
		}
	}
	return $ex;
}

exit(main());