Integrate mainline.

p4raw-id: //depot/perlio@8003

8 files changed, 352 insertions, 284 deletions

diff --git a/t/lib/net-hostent.t b/t/lib/net-hostent.t
index a0ec7bd970..b1c7a9db66 100644
--- a/t/lib/net-hostent.t
+++ b/t/lib/net-hostent.t
@@ -41,7 +41,7 @@ print "ok 5\n";
 # VMS returns "LOCALHOST" under tcp/ip services V4.1 ECO 2, possibly others
 # OS/390 returns localhost.YADDA.YADDA
 
-if ($^O eq 'MSWin32') {
+if ($^O eq 'MSWin32' or $^O eq 'cygwin') {
   print "ok $_ # skipped on win32\n" for (6,7);
 } else {
   print "not " unless $h->name =~ /^localhost(?:\..+)?$/i;
diff --git a/t/lib/syslfs.t b/t/lib/syslfs.t
index 39a57f36ac..cec839bc43 100644
--- a/t/lib/syslfs.t
+++ b/t/lib/syslfs.t
@@ -26,21 +26,28 @@ sub bye {
     exit(0);
 }
 
+my $explained;
+
 sub explain {
-    print <<EOM;
+    unless ($explained++) {
+	print <<EOM;
 #
-# If the lfs (large file support: large meaning larger than two gigabytes)
-# tests are skipped or fail, it may mean either that your process
-# (or process group) is not allowed to write large files (resource
-# limits) or that the file system you are running the tests on doesn't
-# let your user/group have large files (quota) or the filesystem simply
-# doesn't support large files.  You may even need to reconfigure your kernel.
-# (This is all very operating system and site-dependent.)
+# If the lfs (large file support: large meaning larger than two
+# gigabytes) tests are skipped or fail, it may mean either that your
+# process (or process group) is not allowed to write large files
+# (resource limits) or that the file system (the network filesystem?)
+# you are running the tests on doesn't let your user/group have large
+# files (quota) or the filesystem simply doesn't support large files.
+# You may even need to reconfigure your kernel.  (This is all very
+# operating system and site-dependent.)
 #
 # Perl may still be able to support large files, once you have
 # such a process, enough quota, and such a (file) system.
+# It is just that the test failed now.
 #
 EOM
+    }
+    print "1..0 # Skip: @_\n" if @_;
 }
 
 print "# checking whether we have sparse files...\n";
@@ -120,9 +127,8 @@ sysopen(BIG, "big", O_WRONLY|O_CREAT|O_TRUNC) or
 my $sysseek = sysseek(BIG, 5_000_000_000, SEEK_SET);
 unless (! $r && defined $sysseek && $sysseek == 5_000_000_000) {
     $sysseek = 'undef' unless defined $sysseek;
-    print "1..0 # Skip: seeking past 2GB failed: ",
-           $r ? 'signal '.($r & 0x7f) : "$! (sysseek returned $sysseek)", "\n";
-    explain();
+    explain("seeking past 2GB failed: ",
+	    $r ? 'signal '.($r & 0x7f) : "$! (sysseek returned $sysseek)");
     bye();
 }
 
@@ -135,11 +141,12 @@ my $close     = close BIG;
 print "# close failed: $!\n" unless $close;
 unless($syswrite && $close) {
     if ($! =~/too large/i) {
-	print "1..0 # Skip: writing past 2GB failed: process limits?\n";
+	explain("writing past 2GB failed: process limits?");
     } elsif ($! =~ /quota/i) {
-	print "1..0 # Skip: filesystem quota limits?\n";
+	explain("filesystem quota limits?");
+    } else {
+	explain("error: $!");
     }
-    explain();
     bye();
 }
 
@@ -148,8 +155,7 @@ unless($syswrite && $close) {
 print "# @s\n";
 
 unless ($s[7] == 5_000_000_003) {
-    print "1..0 # Skip: not configured to use large files?\n";
-    explain();
+    explain("kernel/fs not configured to use large files?");
     bye();
 }
 
@@ -220,7 +226,7 @@ print "ok 16\n";
 fail unless $zero eq "\0\0\0";
 print "ok 17\n";
 
-explain if $fail;
+explain() if $fail;
 
 bye(); # does the necessary cleanup
 
diff --git a/t/op/each.t b/t/op/each.t
index 879c0d0fd3..4a00a1e2c5 100755
--- a/t/op/each.t
+++ b/t/op/each.t
@@ -1,6 +1,6 @@
 #!./perl
 
-print "1..19\n";
+print "1..20\n";
 
 $h{'abc'} = 'ABC';
 $h{'def'} = 'DEF';
@@ -131,3 +131,15 @@ if ($i == 5) { print "ok 16\n" } else { print "not ok\n" }
     print "ok 19\n";
 }
 
+# Check for Unicode hash keys.
+%u = ("\x{12}", "f", "\x{123}", "fo", "\x{1234}",  "foo");
+$u{"\x{12345}"}  = "bar";
+@u{"\x{123456}"} = "zap";
+
+foreach (keys %u) {
+    unless (length() == 1) {
+	print "not ";
+	last;
+    }
+}
+print "ok 20\n";
diff --git a/t/op/length.t b/t/op/length.t
new file mode 100644
index 0000000000..ceb005ecc4
--- /dev/null
+++ b/t/op/length.t
@@ -0,0 +1,85 @@
+#!./perl
+
+BEGIN {
+    chdir 't' if -d 't';
+    @INC = '../lib';
+}
+
+print "1..13\n";
+
+print "not " unless length("")    == 0;
+print "ok 1\n";
+
+print "not " unless length("abc") == 3;
+print "ok 2\n";
+
+$_ = "foobar";
+print "not " unless length()      == 6;
+print "ok 3\n";
+
+# Okay, so that wasn't very challenging.  Let's go Unicode.
+
+{
+    my $a = "\x{41}";
+
+    print "not " unless length($a) == 1;
+    print "ok 4\n";
+    $test++;
+
+    use bytes;
+    print "not " unless $a eq "\x41" && length($a) == 1;
+    print "ok 5\n";
+    $test++;
+}
+
+{
+    my $a = "\x{80}";
+    
+    print "not " unless length($a) == 1;
+    print "ok 6\n";
+    $test++;
+    
+    use bytes;
+    print "not " unless $a eq "\xc2\x80" && length($a) == 2;
+    print "ok 7\n";
+    $test++;
+}
+
+{
+    my $a = "\x{100}";
+    
+    print "not " unless length($a) == 1;
+    print "ok 8\n";
+    $test++;
+    
+    use bytes;
+    print "not " unless $a eq "\xc4\x80" && length($a) == 2;
+    print "ok 9\n";
+    $test++;
+}
+
+{
+    my $a = "\x{100}\x{80}";
+    
+    print "not " unless length($a) == 2;
+    print "ok 10\n";
+    $test++;
+    
+    use bytes;
+    print "not " unless $a eq "\xc4\x80\xc2\x80" && length($a) == 4;
+    print "ok 11\n";
+    $test++;
+}
+
+{
+    my $a = "\x{80}\x{100}";
+    
+    print "not " unless length($a) == 2;
+    print "ok 12\n";
+    $test++;
+    
+    use bytes;
+    print "not " unless $a eq "\xc2\x80\xc4\x80" && length($a) == 4;
+    print "ok 13\n";
+    $test++;
+}
diff --git a/t/op/lfs.t b/t/op/lfs.t
index e55212fe51..e732adc798 100644
--- a/t/op/lfs.t
+++ b/t/op/lfs.t
@@ -25,21 +25,28 @@ sub bye {
     exit(0);
 }
 
+my $explained;
+
 sub explain {
-    print <<EOM;
+    unless ($explained++) {
+	print <<EOM;
 #
-# If the lfs (large file support: large meaning larger than two gigabytes)
-# tests are skipped or fail, it may mean either that your process
-# (or process group) is not allowed to write large files (resource
-# limits) or that the file system you are running the tests on doesn't
-# let your user/group have large files (quota) or the filesystem simply
-# doesn't support large files.  You may even need to reconfigure your kernel.
-# (This is all very operating system and site-dependent.)
+# If the lfs (large file support: large meaning larger than two
+# gigabytes) tests are skipped or fail, it may mean either that your
+# process (or process group) is not allowed to write large files
+# (resource limits) or that the file system (the network filesystem?)
+# you are running the tests on doesn't let your user/group have large
+# files (quota) or the filesystem simply doesn't support large files.
+# You may even need to reconfigure your kernel.  (This is all very
+# operating system and site-dependent.)
 #
 # Perl may still be able to support large files, once you have
 # such a process, enough quota, and such a (file) system.
+# It is just that the test failed now.
 #
 EOM
+    }
+    print "1..0 # Skip: @_\n" if @_;
 }
 
 print "# checking whether we have sparse files...\n";
@@ -125,8 +132,7 @@ open(BIG, ">big") or do { warn "open failed: $!\n"; bye };
 binmode BIG;
 if ($r or not seek(BIG, 5_000_000_000, $SEEK_SET)) {
     my $err = $r ? 'signal '.($r & 0x7f) : $!;
-    print "1..0 # Skip: seeking past 2GB failed: $err\n";
-    explain();
+    explain("seeking past 2GB failed: $err");
     bye();
 }
 
@@ -138,11 +144,12 @@ my $close = close BIG;
 print "# close failed: $!\n" unless $close;
 unless ($print && $close) {
     if ($! =~/too large/i) {
-	print "1..0 # Skip: writing past 2GB failed: process limits?\n";
+	explain("writing past 2GB failed: process limits?");
     } elsif ($! =~ /quota/i) {
-	print "1..0 # Skip: filesystem quota limits?\n";
+	explain("filesystem quota limits?");
+    } else {
+	explain("error: $!");
     }
-    explain();
     bye();
 }
 
@@ -151,8 +158,7 @@ unless ($print && $close) {
 print "# @s\n";
 
 unless ($s[7] == 5_000_000_003) {
-    print "1..0 # Skip: not configured to use large files?\n";
-    explain();
+    explain("kernel/fs not configured to use large files?");
     bye();
 }
 
@@ -224,7 +230,7 @@ print "ok 16\n";
 fail unless $zero eq "\0\0\0";
 print "ok 17\n";
 
-explain if $fail;
+explain() if $fail;
 
 bye(); # does the necessary cleanup
 
diff --git a/t/op/ref.t b/t/op/ref.t
index a2baab8e3b..8ae90424eb 100755
--- a/t/op/ref.t
+++ b/t/op/ref.t
@@ -1,6 +1,6 @@
 #!./perl
 
-print "1..56\n";
+print "1..61\n";
 
 # Test glob operations.
 
@@ -279,14 +279,34 @@ print $$_,"\n";
     print ${\$_} for @a;
 }
 
+# This test is the reason for postponed destruction in sv_unref
+$a = [1,2,3];
+$a = $a->[1];
+print "not " unless $a == 2;
+print "ok 54\n";
+
+sub x::DESTROY {print "ok ", 54 + shift->[0], "\n"}
+{ my $a1 = bless [4],"x";
+  my $a2 = bless [3],"x";
+  { my $a3 = bless [2],"x";
+    my $a4 = bless [1],"x";
+    567;
+  }
+}
+
+
 # test global destruction
 
+my $test = 59;
+my $test1 = $test + 1;
+my $test2 = $test + 2;
+
 package FINALE;
 
 {
-    $ref3 = bless ["ok 56\n"];		# package destruction
-    my $ref2 = bless ["ok 55\n"];	# lexical destruction
-    local $ref1 = bless ["ok 54\n"];	# dynamic destruction
+    $ref3 = bless ["ok $test2\n"];	# package destruction
+    my $ref2 = bless ["ok $test1\n"];	# lexical destruction
+    local $ref1 = bless ["ok $test\n"];	# dynamic destruction
     1;					# flush any temp values on stack
 }
 
diff --git a/t/op/utf8decode.t b/t/op/utf8decode.t
new file mode 100644
index 0000000000..c631c0a7a9
--- /dev/null
+++ b/t/op/utf8decode.t
@@ -0,0 +1,181 @@
+#!./perl
+
+BEGIN {
+    chdir 't' if -d 't';
+    @INC = '../lib';
+}
+
+print "1..78\n";
+
+my $test = 1;
+
+# This table is based on Markus Kuhn's UTF-8 Decode Stress Tester,
+# http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt,
+# version dated 2000-09-02. 
+
+# Note the \0 instead of a raw zero byte in 2.1.1: for example
+# GNU patch v2.1 has "issues" with raw zero bytes.
+
+my @MK = split(/\n/, <<__EOMK__);
+1	Correct UTF-8
+1.1.1 y "κόσμε"	-		11	ce:ba:e1:bd:b9:cf:83:ce:bc:ce:b5	5
+2	Boundary conditions 
+2.1	First possible sequence of certain length
+2.1.1 y "\0"			0		1	00	1
+2.1.2 y "€"			80		2	c2:80	1
+2.1.3 y "ࠀ"		800		3	e0:a0:80	1
+2.1.4 y "𐀀"		10000		4	f0:90:80:80	1
+2.1.5 y "�����"	200000		5	f8:88:80:80:80	1
+2.1.6 y "������"	4000000		6	fc:84:80:80:80:80	1
+2.2	Last possible sequence of certain length
+2.2.1 y ""			7f		1	7f	1
+2.2.2 y "߿"			7ff		2	df:bf	1
+# The ffff is illegal unless UTF8_ALLOW_FFFF
+2.2.3 n "￿"			ffff		3	ef:bf:bf	1	character 0xffff
+2.2.4 y "����"			1fffff		4	f7:bf:bf:bf	1
+2.2.5 y "�����"			3ffffff		5	fb:bf:bf:bf:bf	1
+2.2.6 y "������"		7fffffff	6	fd:bf:bf:bf:bf:bf	1
+2.3	Other boundary conditions
+2.3.1 y "퟿"		d7ff		3	ed:9f:bf	1
+2.3.2 y ""		e000		3	ee:80:80	1
+2.3.3 y "�"			fffd		3	ef:bf:bd	1
+2.3.4 y "􏿿"		10ffff		4	f4:8f:bf:bf	1
+2.3.5 y "����"		110000		4	f4:90:80:80	1
+3	Malformed sequences
+3.1	Unexpected continuation bytes
+3.1.1 n "�"			-		1	80	-	unexpected continuation byte 0x80
+3.1.2 n "�"			-		1	bf	-	unexpected continuation byte 0xbf
+3.1.3 n "��"			-		2	80:bf	-	unexpected continuation byte 0x80
+3.1.4 n "���"		-		3	80:bf:80	-	unexpected continuation byte 0x80
+3.1.5 n "����"		-		4	80:bf:80:bf	-	unexpected continuation byte 0x80
+3.1.6 n "�����"	-		5	80:bf:80:bf:80	-	unexpected continuation byte 0x80
+3.1.7 n "������"	-		6	80:bf:80:bf:80:bf	-	unexpected continuation byte 0x80
+3.1.8 n "�������"	-		7	80:bf:80:bf:80:bf:80	-	unexpected continuation byte 0x80
+3.1.9 n "����������������������������������������������������������������"				-	64	80:81:82:83:84:85:86:87:88:89:8a:8b:8c:8d:8e:8f:90:91:92:93:94:95:96:97:98:99:9a:9b:9c:9d:9e:9f:a0:a1:a2:a3:a4:a5:a6:a7:a8:a9:aa:ab:ac:ad:ae:af:b0:b1:b2:b3:b4:b5:b6:b7:b8:b9:ba:bb:bc:bd:be:bf	-	unexpected continuation byte 0x80
+3.2	Lonely start characters
+3.2.1 n "� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � "	-	64 	c0:20:c1:20:c2:20:c3:20:c4:20:c5:20:c6:20:c7:20:c8:20:c9:20:ca:20:cb:20:cc:20:cd:20:ce:20:cf:20:d0:20:d1:20:d2:20:d3:20:d4:20:d5:20:d6:20:d7:20:d8:20:d9:20:da:20:db:20:dc:20:dd:20:de:20:df:20	-	unexpected non-continuation byte 0x20 after byte 0xc0
+3.2.2 n "� � � � � � � � � � � � � � � � "	-	32	e0:20:e1:20:e2:20:e3:20:e4:20:e5:20:e6:20:e7:20:e8:20:e9:20:ea:20:eb:20:ec:20:ed:20:ee:20:ef:20	-	unexpected non-continuation byte 0x20 after byte 0xe0
+3.2.3 n "� � � � � � � � "	-	16	f0:20:f1:20:f2:20:f3:20:f4:20:f5:20:f6:20:f7:20	-	unexpected non-continuation byte 0x20 after byte 0xf0
+3.2.4 n "� � � � "		-	8	f8:20:f9:20:fa:20:fb:20	-	unexpected non-continuation byte 0x20 after byte 0xf8
+3.2.5 n "� � "			-	4	fc:20:fd:20	-	unexpected non-continuation byte 0x20 after byte 0xfc
+3.3	Sequences with last continuation byte missing
+3.3.1 n "�"			-	1	c0	-	1 byte, need 2
+3.3.2 n "��"			-	2	e0:80	-	2 bytes, need 3
+3.3.3 n "���"		-	3	f0:80:80	-	3 bytes, need 4
+3.3.4 n "����"		-	4	f8:80:80:80	-	4 bytes, need 5
+3.3.5 n "�����"	-	5	fc:80:80:80:80	-	5 bytes, need 6
+3.3.6 n "�"			-	1	df	-	1 byte, need 2
+3.3.7 n "�"			-	2	ef:bf	-	2 bytes, need 3
+3.3.8 n "���"			-	3	f7:bf:bf	-	3 bytes, need 4
+3.3.9 n "����"			-	4	fb:bf:bf:bf	-	4 bytes, need 5
+3.3.10 n "�����"		-	5	fd:bf:bf:bf:bf	-	5 bytes, need 6
+3.4	Concatenation of incomplete sequences
+3.4.1 n "�����������������������������"	-	30	c0:e0:80:f0:80:80:f8:80:80:80:fc:80:80:80:80:df:ef:bf:f7:bf:bf:fb:bf:bf:bf:fd:bf:bf:bf:bf	-	unexpected continuation byte 0xe0
+3.5	Impossible bytes
+3.5.1 n "�"			-	1	fe	-	byte 0xfe
+3.5.2 n "�"			-	1	ff	-	byte 0xff
+3.5.3 n "����"			-	4	fe:fe:ff:ff	-	byte 0xfe
+4	Overlong sequences
+4.1	Examples of an overlong ASCII character
+4.1.1 n "��"			-	2	c0:af	-	2 bytes, need 1
+4.1.2 n "���"		-	3	e0:80:af	-	3 bytes, need 1
+4.1.3 n "����"		-	4	f0:80:80:af	-	4 bytes, need 1
+4.1.4 n "�����"	-	5	f8:80:80:80:af	-	5 bytes, need 1
+4.1.5 n "������"	-	6	fc:80:80:80:80:af	-	6 bytes, need 1
+4.2	Maximum overlong sequences
+4.2.1 n "��"			-	2	c1:bf	-	2 bytes, need 1
+4.2.2 n "���"		-	3	e0:9f:bf	-	3 bytes, need 2
+4.2.3 n "����"		-	4	f0:8f:bf:bf	-	4 bytes, need 3
+4.2.4 n "�����"		-	5	f8:87:bf:bf:bf	-	5 bytes, need 4
+4.2.5 n "������"		-	6	fc:83:bf:bf:bf:bf	-	6 bytes, need 5
+4.3	Overlong representation of the NUL character
+4.3.1 n "��"			-	2	c0:80	-	2 bytes, need 1
+4.3.2 n "���"		-	3	e0:80:80	-	3 bytes, need 1
+4.3.3 n "����"		-	4	f0:80:80:80	-	4 bytes, need 1
+4.3.4 n "�����"	-	5	f8:80:80:80:80	-	5 bytes, need 1
+4.3.5 n "������"	-	6	fc:80:80:80:80:80	-	6 bytes, need 1
+5	Illegal code positions
+5.1	Single UTF-16 surrogates
+5.1.1 n "���"		-	3	ed:a0:80	-	UTF-16 surrogate 0xd800
+5.1.2 n "���"			-	3	ed:ad:bf	-	UTF-16 surrogate 0xdb7f
+5.1.3 n "���"		-	3	ed:ae:80	-	UTF-16 surrogate 0xdb80
+5.1.4 n "���"			-	3	ed:af:bf	-	UTF-16 surrogate 0xdbff
+5.1.5 n "���"		-	3	ed:b0:80	-	UTF-16 surrogate 0xdc00
+5.1.6 n "���"		-	3	ed:be:80	-	UTF-16 surrogate 0xdf80
+5.1.7 n "���"			-	3	ed:bf:bf	-	UTF-16 surrogate 0xdfff
+5.2	Paired UTF-16 surrogates
+5.2.1 n "������"		-	6	ed:a0:80:ed:b0:80	-	UTF-16 surrogate 0xd800
+5.2.2 n "������"		-	6	ed:a0:80:ed:bf:bf	-	UTF-16 surrogate 0xd800
+5.2.3 n "������"		-	6	ed:ad:bf:ed:b0:80	-	UTF-16 surrogate 0xdb7f
+5.2.4 n "������"		-	6	ed:ad:bf:ed:bf:bf	-	UTF-16 surrogate 0xdb7f
+5.2.5 n "������"		-	6	ed:ae:80:ed:b0:80	-	UTF-16 surrogate 0xdb80
+5.2.6 n "������"		-	6	ed:ae:80:ed:bf:bf	-	UTF-16 surrogate 0xdb80
+5.2.7 n "������"		-	6	ed:af:bf:ed:b0:80	-	UTF-16 surrogate 0xdbff
+5.2.8 n "������"		-	6	ed:af:bf:ed:bf:bf	-	UTF-16 surrogate 0xdbff
+5.3	Other illegal code positions
+5.3.1 n "￾"			-	3	ef:bf:be	-	byte order mark 0xfffe
+# The ffff is illegal unless UTF8_ALLOW_FFFF
+5.3.2 n "￿"			-	3	ef:bf:bf	-	character 0xffff
+__EOMK__
+
+# 104..181
+{
+    my $WARNCNT;
+    my $id;
+
+    local $SIG{__WARN__} =
+	sub {
+	    # print "# $id: @_";
+	    $WARNCNT++;
+	    $WARNMSG = "@_";
+	};
+
+    sub moan {
+	print "$id: @_";
+    }
+    
+    sub test_unpack_U {
+	$WARNCNT = 0;
+	$WARNMSG = "";
+	unpack('U*', $_[0]);
+    }
+
+    for (@MK) {
+	if (/^(?:\d+(?:\.\d+)?)\s/ || /^#/) {
+	    # print "# $_\n";
+	} elsif (/^(\d+\.\d+\.\d+[bu]?)\s+([yn])\s+"(.+)"\s+([0-9a-f]{1,8}|-)\s+(\d+)\s+([0-9a-f]{2}(?::[0-9a-f]{2})*)(?:\s+((?:\d+|-)(?:\s+(.+))?))?$/) {
+	    $id = $1;
+	    my ($okay, $bytes, $Unicode, $byteslen, $hex, $charslen, $error) =
+		($2, $3, $4, $5, $6, $7, $8);
+	    my @hex = split(/:/, $hex);
+	    unless (@hex == $byteslen) {
+		my $nhex = @hex;
+		moan "amount of hex ($nhex) not equal to byteslen ($byteslen)\n";
+	    }
+	    {
+		use bytes;
+		my $bytesbyteslen = length($bytes);
+		unless ($bytesbyteslen == $byteslen) {
+		    moan "bytes length() ($bytesbyteslen) not equal to $byteslen\n";
+		}
+	    }
+	    if ($okay eq 'y') {
+		test_unpack_U($bytes);
+		if ($WARNCNT) {
+		    moan "unpack('U*') false negative\n";
+		    print "not ";
+		}
+	    } elsif ($okay eq 'n') {
+		test_unpack_U($bytes);
+		if ($WARNCNT == 0 || ($error ne '' && $WARNMSG !~ /$error/)) {
+		    moan "unpack('U*') false positive\n";
+		    print "not ";
+		}
+	    }
+	    print "ok $test\n";
+	    $test++;
+ 	} else {
+	    moan "unknown format\n";
+	}
+    }
+}
diff --git a/t/pragma/utf8.t b/t/pragma/utf8.t
index 60cbd8c97a..8efd571901 100755
--- a/t/pragma/utf8.t
+++ b/t/pragma/utf8.t
@@ -10,7 +10,7 @@ BEGIN {
     }
 }
 
-print "1..191\n";
+print "1..103\n";
 
 my $test = 1;
 
@@ -104,6 +104,7 @@ sub nok_bytes {
     ok $1, '123alpha';
     $test++;				# 12
 }
+
 {
     use utf8;
 
@@ -204,10 +205,8 @@ sub nok_bytes {
 
 	ok $1, pack("C*", 0342);
 	$test++;				# 40
-
     }
 
-
     {
 	no utf8;
 	$_="\342\230\272>\342\230\272\342\230\272";
@@ -262,6 +261,7 @@ sub nok_bytes {
 	ok $tmp, pack("C*", 0342, 0230, 0272);
 	$test++;				# 54
     }
+
     {
 	use bytes;
 	no utf8;
@@ -295,7 +295,6 @@ sub nok_bytes {
 
 	ok $1, pack("C*", 0342);
 	$test++;				# 64
-
     }
 
     ok "\x{ab}" =~ /^\x{ab}$/, 1;
@@ -389,8 +388,6 @@ sub nok_bytes {
 {
     # bug id 20000323.056
 
-    use utf8;
-
     print "not " unless "\x{41}" eq +v65;
     print "ok $test\n";
     $test++;
@@ -564,242 +561,3 @@ sub nok_bytes {
     print "ok $test\n";
     $test++;
 }
-
-# This table is based on Markus Kuhn's UTF-8 Decode Stress Tester,
-# http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt,
-# version dated 2000-09-02. 
-
-# Note the \0 instead of a raw zero byte in 2.1.1: for example
-# GNU patch v2.1 has "issues" with raw zero bytes.
-
-my @MK = split(/\n/, <<__EOMK__);
-1	Correct UTF-8
-1.1.1 y "κόσμε"	-		11	ce:ba:e1:bd:b9:cf:83:ce:bc:ce:b5	5
-2	Boundary conditions 
-2.1	First possible sequence of certain length
-2.1.1 y "\0"			0		1	00	1
-2.1.2 y "€"			80		2	c2:80	1
-2.1.3 y "ࠀ"		800		3	e0:a0:80	1
-2.1.4 y "𐀀"		10000		4	f0:90:80:80	1
-2.1.5 y "�����"	200000		5	f8:88:80:80:80	1
-2.1.6 y "������"	4000000		6	fc:84:80:80:80:80	1
-2.2	Last possible sequence of certain length
-2.2.1 y ""			7f		1	7f	1
-2.2.2 y "߿"			7ff		2	df:bf	1
-# The ffff is illegal unless UTF8_ALLOW_FFFF
-2.2.3 n "￿"			ffff		3	ef:bf:bf	1
-2.2.4 y "����"			1fffff		4	f7:bf:bf:bf	1
-2.2.5 y "�����"			3ffffff		5	fb:bf:bf:bf:bf	1
-2.2.6 y "������"		7fffffff	6	fd:bf:bf:bf:bf:bf	1
-2.3	Other boundary conditions
-2.3.1 y "퟿"		d7ff		3	ed:9f:bf	1
-2.3.2 y ""		e000		3	ee:80:80	1
-2.3.3 y "�"			fffd		3	ef:bf:bd	1
-2.3.4 y "􏿿"		10ffff		4	f4:8f:bf:bf	1
-2.3.5 y "����"		110000		4	f4:90:80:80	1
-3	Malformed sequences
-3.1	Unexpected continuation bytes
-3.1.1 n "�"			-		1	80
-3.1.2 n "�"			-		1	bf
-3.1.3 n "��"			-		2	80:bf
-3.1.4 n "���"		-		3	80:bf:80
-3.1.5 n "����"		-		4	80:bf:80:bf
-3.1.6 n "�����"	-		5	80:bf:80:bf:80
-3.1.7 n "������"	-		6	80:bf:80:bf:80:bf
-3.1.8 n "�������"	-		7	80:bf:80:bf:80:bf:80
-3.1.9 n "����������������������������������������������������������������"				-	64	80:81:82:83:84:85:86:87:88:89:8a:8b:8c:8d:8e:8f:90:91:92:93:94:95:96:97:98:99:9a:9b:9c:9d:9e:9f:a0:a1:a2:a3:a4:a5:a6:a7:a8:a9:aa:ab:ac:ad:ae:af:b0:b1:b2:b3:b4:b5:b6:b7:b8:b9:ba:bb:bc:bd:be:bf
-3.2	Lonely start characters
-3.2.1 n "� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � "	-	64 	c0:20:c1:20:c2:20:c3:20:c4:20:c5:20:c6:20:c7:20:c8:20:c9:20:ca:20:cb:20:cc:20:cd:20:ce:20:cf:20:d0:20:d1:20:d2:20:d3:20:d4:20:d5:20:d6:20:d7:20:d8:20:d9:20:da:20:db:20:dc:20:dd:20:de:20:df:20
-3.2.2 n "� � � � � � � � � � � � � � � � "	-	32	e0:20:e1:20:e2:20:e3:20:e4:20:e5:20:e6:20:e7:20:e8:20:e9:20:ea:20:eb:20:ec:20:ed:20:ee:20:ef:20
-3.2.3 n "� � � � � � � � "	-	16	f0:20:f1:20:f2:20:f3:20:f4:20:f5:20:f6:20:f7:20
-3.2.4 n "� � � � "		-	8	f8:20:f9:20:fa:20:fb:20
-3.2.5 n "� � "			-	4	fc:20:fd:20
-3.3	Sequences with last continuation byte missing
-3.3.1 n "�"			-	1	c0
-3.3.2 n "��"			-	2	e0:80
-3.3.3 n "���"		-	3	f0:80:80
-3.3.4 n "����"		-	4	f8:80:80:80
-3.3.5 n "�����"	-	5	fc:80:80:80:80
-3.3.6 n "�"			-	1	df
-3.3.7 n "�"			-	2	ef:bf
-3.3.8 n "���"			-	3	f7:bf:bf
-3.3.9 n "����"			-	4	fb:bf:bf:bf
-3.3.10 n "�����"		-	5	fd:bf:bf:bf:bf
-3.4	Concatenation of incomplete sequences
-3.4.1 n "�����������������������������"	-	30	c0:e0:80:f0:80:80:f8:80:80:80:fc:80:80:80:80:df:ef:bf:f7:bf:bf:fb:bf:bf:bf:fd:bf:bf:bf:bf
-3.5	Impossible bytes
-3.5.1 n "�"			-	1	fe
-3.5.2 n "�"			-	1	ff
-3.5.3 n "����"			-	4	fe:fe:ff:ff
-4	Overlong sequences
-4.1	Examples of an overlong ASCII character
-4.1.1 n "��"			-	2	c0:af
-4.1.2 n "���"		-	3	e0:80:af
-4.1.3 n "����"		-	4	f0:80:80:af
-4.1.4 n "�����"	-	5	f8:80:80:80:af
-4.1.5 n "������"	-	6	fc:80:80:80:80:af
-4.2	Maximum overlong sequences
-4.2.1 n "��"			-	2	c1:bf
-4.2.2 n "���"		-	3	e0:9f:bf
-4.2.3 n "����"		-	4	f0:8f:bf:bf
-4.2.4 n "�����"		-	5	f8:87:bf:bf:bf
-4.2.5 n "������"		-	6	fc:83:bf:bf:bf:bf
-4.3	Overlong representation of the NUL character
-4.3.1 n "��"			-	2	c0:80
-4.3.2 n "���"		-	3	e0:80:80
-4.3.3 n "����"		-	4	f0:80:80:80
-4.3.4 n "�����"	-	5	f8:80:80:80:80
-4.3.5 n "������"	-	6	fc:80:80:80:80:80
-5	Illegal code positions
-5.1	Single UTF-16 surrogates
-5.1.1 n "���"		-	3	ed:a0:80
-5.1.2 n "���"			-	3	ed:ad:bf
-5.1.3 n "���"		-	3	ed:ae:80
-5.1.4 n "���"			-	3	ed:af:bf
-5.1.5 n "���"		-	3	ed:b0:80
-5.1.6 n "���"		-	3	ed:be:80
-5.1.7 n "���"			-	3	ed:bf:bf
-5.2	Paired UTF-16 surrogates
-5.2.1 n "������"		-	6	ed:a0:80:ed:b0:80
-5.2.2 n "������"		-	6	ed:a0:80:ed:bf:bf
-5.2.3 n "������"		-	6	ed:ad:bf:ed:b0:80
-5.2.4 n "������"		-	6	ed:ad:bf:ed:bf:bf
-5.2.5 n "������"		-	6	ed:ae:80:ed:b0:80
-5.2.6 n "������"		-	6	ed:ae:80:ed:bf:bf
-5.2.7 n "������"		-	6	ed:af:bf:ed:b0:80
-5.2.8 n "������"		-	6	ed:af:bf:ed:bf:bf
-5.3	Other illegal code positions
-5.3.1 n "￾"			-	3	ef:bf:be
-# The ffff is illegal unless UTF8_ALLOW_FFFF
-5.3.2 n "￿"			-	3	ef:bf:bf
-__EOMK__
-
-# 104..181
-{
-    my $WARN;
-    my $id;
-
-    local $SIG{__WARN__} =
-	sub {
-	    # print "# $id: @_";
-	    $WARN++;
-	};
-
-    sub moan {
-	print "$id: @_";
-    }
-    
-    sub test_unpack_U {
-	$WARN = 0;
-	unpack('U*', $_[0]);
-    }
-
-    for (@MK) {
-	if (/^(?:\d+(?:\.\d+)?)\s/ || /^#/) {
-	    # print "# $_\n";
-	} elsif (/^(\d+\.\d+\.\d+[bu]?)\s+([yn])\s+"(.+)"\s+([0-9a-f]{1,8}|-)\s+(\d+)\s+([0-9a-f]{2}(?::[0-9a-f]{2})*)(?:\s+(\d+))?$/) {
-	    $id = $1;
-	    my ($okay, $bytes, $Unicode, $byteslen, $hex, $charslen) =
-		($2, $3, $4, $5, $6, $7);
-	    my @hex = split(/:/, $hex);
-	    unless (@hex == $byteslen) {
-		my $nhex = @hex;
-		moan "amount of hex ($nhex) not equal to byteslen ($byteslen)\n";
-	    }
-	    {
-		use bytes;
-		my $bytesbyteslen = length($bytes);
-		unless ($bytesbyteslen == $byteslen) {
-		    moan "bytes length() ($bytesbyteslen) not equal to $byteslen\n";
-		}
-	    }
-	    if ($okay eq 'y') {
-		test_unpack_U($bytes);
-		unless ($WARN == 0) {
-		    moan "unpack('U*') false negative\n";
-		    print "not ";
-		}
-	    } elsif ($okay eq 'n') {
-		test_unpack_U($bytes);
-		unless ($WARN) {
-		    moan "unpack('U*') false positive\n";
-		    print "not ";
-		}
-	    }
-	    print "ok $test\n";
-	    $test++;
- 	} else {
-	    moan "unknown format\n";
-	}
-    }
-}
-
-{
-    # tests 182..191
-
-    {
-	my $a = "\x{41}";
-
-	print "not " unless length($a) == 1;
-	print "ok $test\n";
-	$test++;
-
-	use bytes;
-	print "not " unless $a eq "\x41" && length($a) == 1;
-	print "ok $test\n";
-	$test++;
-    }
-
-    {
-	my $a = "\x{80}";
-
-	print "not " unless length($a) == 1;
-	print "ok $test\n";
-	$test++;
-
-	use bytes;
-	print "not " unless $a eq "\xc2\x80" && length($a) == 2;
-	print "ok $test\n";
-	$test++;
-    }
-
-    {
-	my $a = "\x{100}";
-
-	print "not " unless length($a) == 1;
-	print "ok $test\n";
-	$test++;
-
-	use bytes;
-	print "not " unless $a eq "\xc4\x80" && length($a) == 2;
-	print "ok $test\n";
-	$test++;
-    }
-
-    {
-	my $a = "\x{100}\x{80}";
-
-	print "not " unless length($a) == 2;
-	print "ok $test\n";
-	$test++;
-
-	use bytes;
-	print "not " unless $a eq "\xc4\x80\xc2\x80" && length($a) == 4;
-	print "ok $test\n";
-	$test++;
-    }
-
-    {
-	my $a = "\x{80}\x{100}";
-
-	print "not " unless length($a) == 2;
-	print "ok $test\n";
-	$test++;
-
-	use bytes;
-	print "not " unless $a eq "\xc2\x80\xc4\x80" && length($a) == 4;
-	print "ok $test\n";
-	$test++;
-    }
-}
-