# the format supported by op/regexp.t. If you want to add a test
# that does fit that format, add it to op/re_tests, not here.
-print "1..581\n";
+$| = 1;
+
+print "1..715\n";
BEGIN {
chdir 't' if -d 't';
@INC = '../lib';
}
-eval 'use Config'; # Defaults assumed if this fails
-# XXX known to leak scalars
-$ENV{PERL_DESTRUCT_LEVEL} = 0 unless $ENV{PERL_DESTRUCT_LEVEL} > 3;
+eval 'use Config'; # Defaults assumed if this fails
$x = "abc\ndef\n";
if ("ab\ncd\n" =~ /^cd/) {print "ok 24\n";} else {print "not ok 24\n";}
$* = 0;
-#$XXX{123} = 123;
-#$XXX{234} = 234;
-#$XXX{345} = 345;
-#
-#@XXX = ('ok 25','not ok 25', 'ok 26','not ok 26','not ok 27');
-#while ($_ = shift(@XXX)) {
-# ?(.*)? && (print $1,"\n");
-# /not/ && reset;
-# /not ok 26/ && reset 'X';
-#}
-#
-#while (($key,$val) = each(%XXX)) {
-# print "not ok 27\n";
-# exit;
-#}
-#
-#print "ok 27\n";
-for (25..27) { print "ok $_\n" }
+$XXX{123} = 123;
+$XXX{234} = 234;
+$XXX{345} = 345;
+
+@XXX = ('ok 25','not ok 25', 'ok 26','not ok 26','not ok 27');
+while ($_ = shift(@XXX)) {
+ ?(.*)? && (print $1,"\n");
+ /not/ && reset;
+ /not ok 26/ && reset 'X';
+}
+
+while (($key,$val) = each(%XXX)) {
+ print "not ok 27\n";
+ exit;
+}
+
+print "ok 27\n";
'cde' =~ /[^ab]*/;
'xyz' =~ //;
print "# length=$l\nnot " unless "ba$a=" =~ /a$a=/;
print "ok $test\n";
$test++;
-
+
print "not " if "b$a=" =~ /a$a=/;
print "ok $test\n";
$test++;
);
for ( keys %ans ) {
- print "# const-len `$_' not => $ans{$_}\nnot "
+ print "# const-len `$_' not => $ans{$_}\nnot "
if $ans{$_} xor /a(?=([yx]($long_constant_len)){2,4}[k-o]).*b./o;
print "ok $test\n";
$test++;
- print "# var-len `$_' not => $ans{$_}\nnot "
+ print "# var-len `$_' not => $ans{$_}\nnot "
if $ans{$_} xor /a(?=([yx]($long_var_len)){2,4}[k-o]).*b./o;
print "ok $test\n";
$test++;
$_ = " a (bla()) and x(y b((l)u((e))) and b(l(e)e)e";
$expect = "(bla()) ((l)u((e))) (l(e)e)";
-sub matchit {
+sub matchit {
m/
(
- \(
+ \(
(?{ $c = 1 }) # Initialize
(?:
(?(?{ $c == 0 }) # PREVIOUS iteration was OK, stop the loop
(?!
) # Fail: will unwind one iteration back
- )
+ )
(?:
[^()]+ # Match a big chunk
(?=
[()]
) # Do not try to match subchunks
|
- \(
+ \(
(?{ ++$c })
|
- \)
+ \)
(?{ --$c })
)
)+ # This may not match with different subblocks
if ($code eq '=xx') {
print "#'$@','$res','$blah'\nnot " unless not $@ and $res;
} else {
- print "#'$@','$res','$blah'\nnot " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
+ print "#'$@','$res','$blah'\nnot " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
}
print "ok $test\n";
$test++;
}
sub prefixify {
- my($v,$a,$b,$res) = @_;
- $v =~ s/\Q$a\E/$b/;
- print "not " unless $res eq $v;
+ my($v,$a,$b,$res) = @_;
+ $v =~ s/\Q$a\E/$b/;
+ print "not " unless $res eq $v;
print "ok $test\n";
$test++;
}
print "ok $test\n";
$test++;
-$a=qr/(?{++$b})/;
+$a=qr/(?{++$b})/;
$b = 7;
-/$a$a/;
-print "not " unless $b eq '9';
+/$a$a/;
+print "not " unless $b eq '9';
print "ok $test\n";
$test++;
-$c="$a";
-/$a$a/;
-print "not " unless $b eq '11';
+$c="$a";
+/$a$a/;
+print "not " unless $b eq '11';
print "ok $test\n";
$test++;
{
- use re "eval";
- /$a$c$a/;
- print "not " unless $b eq '14';
+ use re "eval";
+ /$a$c$a/;
+ print "not " unless $b eq '14';
print "ok $test\n";
$test++;
$test++;
- no re "eval";
+ no re "eval";
$match = eval { /$a$c$a/ };
- print "not "
+ print "not "
unless $b eq '14' and $@ =~ /Eval-group not allowed/ and not $match;
print "ok $test\n";
$test++;
$test++;
print "not " unless $c == 3;
print "ok $test\n";
-$test++;
-
+$test++;
+
sub must_warn_pat {
my $warn_pat = shift;
return sub { print "not " unless $_[0] =~ /$warn_pat/ }
print "ok $test\n";
$test++;
-print "not "
+print "not "
if defined $+[1] or defined $-[1] or defined $+[2] or defined $-[2];
print "ok $test\n";
$test++;
print "ok $test\n";
$test++;
-print "not "
+print "not "
if defined $+[3] or defined $-[3] or defined $+[4] or defined $-[4];
print "ok $test\n";
$test++;
print "ok $test\n";
$test++;
-print "not "
+print "not "
if defined $+[2] or defined $-[2] or defined $+[4] or defined $-[4];
print "ok $test\n";
$test++;
print "ok $test\n";
$test++;
-print "not "
+print "not "
if defined $+[2] or defined $-[2] or defined $+[3] or defined $-[3];
print "ok $test\n";
$test++;
eval { $+[0] = 13; };
-print "not "
+print "not "
if $@ !~ /^Modification of a read-only value attempted/;
print "ok $test\n";
$test++;
eval { $-[0] = 13; };
-print "not "
+print "not "
if $@ !~ /^Modification of a read-only value attempted/;
print "ok $test\n";
$test++;
eval { @+ = (7, 6, 5); };
-print "not "
+print "not "
if $@ !~ /^Modification of a read-only value attempted/;
print "ok $test\n";
$test++;
eval { @- = qw(foo bar); };
-print "not "
+print "not "
if $@ !~ /^Modification of a read-only value attempted/;
print "ok $test\n";
$test++;
undef $foo; undef $bar;
print "#'$str','$foo','$bar'\nnot "
- unless $str =~ /b(?{$foo = $_; $bar = pos})c/
+ unless $str =~ /b(?{$foo = $_; $bar = pos})c/
and $foo eq 'abcde' and $bar eq 2;
print "ok $test\n";
$test++;
undef $foo; undef $bar;
pos $str = undef;
print "#'$str','$foo','$bar'\nnot "
- unless $str =~ /b(?{$foo = $_; $bar = pos})c/g
+ unless $str =~ /b(?{$foo = $_; $bar = pos})c/g
and $foo eq 'abcde' and $bar eq 2 and pos $str eq 3;
print "ok $test\n";
$test++;
undef $foo; undef $bar;
print "#'$str','$foo','$bar'\nnot "
- unless /b(?{$foo = $_; $bar = pos})c/
+ unless /b(?{$foo = $_; $bar = pos})c/
and $foo eq 'abcde' and $bar eq 2;
print "ok $test\n";
$test++;
undef $foo; undef $bar;
print "#'$str','$foo','$bar'\nnot "
- unless /b(?{$foo = $_; $bar = pos})c/g
+ unless /b(?{$foo = $_; $bar = pos})c/g
and $foo eq 'abcde' and $bar eq 2 and pos eq 3;
print "ok $test\n";
$test++;
undef $foo; undef $bar;
$_ = 'abcde|abcde';
print "#'$str','$foo','$bar','$_'\nnot "
- unless s/b(?{$foo = $_; $bar = pos})c/x/g and $foo eq 'abcde|abcde'
+ unless s/b(?{$foo = $_; $bar = pos})c/x/g and $foo eq 'abcde|abcde'
and $bar eq 8 and $_ eq 'axde|axde';
print "ok $test\n";
$test++;
print "ok $test\n";
$test++;
-$_='123x123';
+$_='123x123';
@res = /(\d*|x)/g;
print "not " unless('123||x|123|' eq join '|', @res);
print "ok $test\n";
print "not " unless "@space2" eq "spc tab";
print "ok $test # @space2\n";
$test++;
-
+
# bugid 20001021.005 - this caused a SEGV
print "not " unless undef =~ /^([^\/]*)(.*)$/;
print "ok $test\n";
print "ok $test\n";
$test++;
+my $ordA = ord('A');
+
$_ = "a\x{100}b";
if (/(.)(\C)(\C)(.)/) {
print "ok 232\n";
} else {
print "not ok 233\n";
}
- if ($2 eq "\xC4") {
- print "ok 234\n";
- } else {
- print "not ok 234\n";
- }
- if ($3 eq "\x80") {
- print "ok 235\n";
+ if ($ordA == 65) { # ASCII (or equivalent), should be UTF-8
+ if ($2 eq "\xC4") {
+ print "ok 234\n";
+ } else {
+ print "not ok 234\n";
+ }
+ if ($3 eq "\x80") {
+ print "ok 235\n";
+ } else {
+ print "not ok 235\n";
+ }
+ } elsif ($ordA == 193) { # EBCDIC (or equivalent), should be UTF-EBCDIC
+ if ($2 eq "\x8C") {
+ print "ok 234\n";
+ } else {
+ print "not ok 234\n";
+ }
+ if ($3 eq "\x41") {
+ print "ok 235\n";
+ } else {
+ print "not ok 235\n";
+ }
} else {
- print "not ok 235\n";
+ for (234..235) {
+ print "not ok $_ # ord('A') == $ordA\n";
+ }
}
if ($4 eq "b") {
print "ok 236\n";
$_ = "\x{100}";
if (/(\C)/g) {
print "ok 237\n";
- if ($1 eq "\xC4") {
- print "ok 238\n";
+ # currently \C are still tagged as UTF-8
+ if ($ordA == 65) {
+ if ($1 eq "\xC4") {
+ print "ok 238\n";
+ } else {
+ print "not ok 238\n";
+ }
+ } elsif ($ordA == 193) {
+ if ($1 eq "\x8C") {
+ print "ok 238\n";
+ } else {
+ print "not ok 238\n";
+ }
} else {
- print "not ok 238\n";
+ print "not ok 238 # ord('A') == $ordA\n";
}
} else {
for (237..238) {
}
if (/(\C)/g) {
print "ok 239\n";
- if ($1 eq "\x80") {
- print "ok 240\n";
+ # currently \C are still tagged as UTF-8
+ if ($ordA == 65) {
+ if ($1 eq "\x80") {
+ print "ok 240\n";
+ } else {
+ print "not ok 240\n";
+ }
+ } elsif ($ordA == 193) {
+ if ($1 eq "\x41") {
+ print "ok 240\n";
+ } else {
+ print "not ok 240\n";
+ }
} else {
- print "not ok 240\n";
+ print "not ok 240 # ord('A') == $ordA\n";
}
} else {
for (239..240) {
}
} else {
for (244..245) {
- print "ok $_ # Skip: not EBCDIC\n";
+ print "ok $_ # Skip: only in EBCDIC\n";
}
}
"#latin[$latin]\nnot ok $test\n";
$test++;
$latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a
- use utf8;
+ use utf8; # needed for the raw UTF-8
$latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a
}
}
{
# the second half of 20001028.003
+ my $X = '';
$X =~ s/^/chr(1488)/e;
print "not " unless length $X == 1 && ord($X) == 1488;
print "ok 260\n";
" " => 'Zs',
"\0" => 'Cc',
);
-
- for my $char (keys %s) {
+
+ for my $char (map { s/^\S+ //; $_ }
+ sort map { sprintf("%06x", ord($_))." $_" } keys %s) {
my $class = $s{$char};
- my $code = sprintf("%04x", ord($char));
- printf "# 0x$code\n";
+ my $code = sprintf("%06x", ord($char));
+ printf "#\n# 0x$code\n#\n";
print "# IsAlpha\n";
if ($class =~ /^[LM]/) {
print "not " unless $char =~ /\p{IsAlpha}/;
print "ok $test\n"; $test++;
}
print "# IsASCII\n";
- if ($code <= 127) {
+ if ($code le '00007f') {
print "not " unless $char =~ /\p{IsASCII}/;
print "ok $test\n"; $test++;
print "not " if $char =~ /\P{IsASCII}/;
print "ok 576\n";
print "not " unless $` eq "abc\x{100}" && length($`) == 4;
- print "ok 577\n";
+ print "ok 577\n";
print "not " unless $& eq "\x{200}\x{300}\x{380}" && length($&) == 3;
- print "ok 578\n";
+ print "ok 578\n";
print "not " unless $' eq "\x{400}defg" && length($') == 5;
- print "ok 579\n";
+ print "ok 579\n";
print "not " unless $1 eq "\x{200}\x{300}" && length($1) == 2;
- print "ok 580\n";
+ print "ok 580\n";
} else {
for (576..580) { print "not ok $_\n" }
}
print "ok 581\n";
}
+
+{
+ $test = 582;
+
+ # bugid 20010410.006
+ for my $rx (
+ '/(.*?)\{(.*?)\}/csg',
+ '/(.*?)\{(.*?)\}/cg',
+ '/(.*?)\{(.*?)\}/sg',
+ '/(.*?)\{(.*?)\}/g',
+ '/(.+?)\{(.+?)\}/csg',
+ )
+ {
+ my($input, $i);
+
+ $i = 0;
+ $input = "a{b}c{d}";
+ eval <<EOT;
+ while (eval \$input =~ $rx) {
+ print "# \\\$1 = '\$1' \\\$2 = '\$2'\n";
+ ++\$i;
+ }
+EOT
+ print "not " unless $i == 2;
+ print "ok " . $test++ . "\n";
+ }
+}
+
+{
+ # from Robin Houston
+
+ my $x = "\x{12345678}";
+ $x =~ s/(.)/$1/g;
+ print "not " unless ord($x) == 0x12345678 && length($x) == 1;
+ print "ok 587\n";
+}
+
+{
+ my $x = "\x7f";
+
+ print "not " if $x =~ /[\x80-\xff]/;
+ print "ok 588\n";
+
+ print "not " if $x =~ /[\x80-\x{100}]/;
+ print "ok 589\n";
+
+ print "not " if $x =~ /[\x{100}]/;
+ print "ok 590\n";
+
+ print "not " if $x =~ /\p{InLatin1Supplement}/;
+ print "ok 591\n";
+
+ print "not " unless $x =~ /\P{InLatin1Supplement}/;
+ print "ok 592\n";
+
+ print "not " if $x =~ /\p{InLatinExtendedA}/;
+ print "ok 593\n";
+
+ print "not " unless $x =~ /\P{InLatinExtendedA}/;
+ print "ok 594\n";
+}
+
+{
+ my $x = "\x80";
+
+ print "not " unless $x =~ /[\x80-\xff]/;
+ print "ok 595\n";
+
+ print "not " unless $x =~ /[\x80-\x{100}]/;
+ print "ok 596\n";
+
+ print "not " if $x =~ /[\x{100}]/;
+ print "ok 597\n";
+
+ print "not " unless $x =~ /\p{InLatin1Supplement}/;
+ print "ok 598\n";
+
+ print "not " if $x =~ /\P{InLatin1Supplement}/;
+ print "ok 599\n";
+
+ print "not " if $x =~ /\p{InLatinExtendedA}/;
+ print "ok 600\n";
+
+ print "not " unless $x =~ /\P{InLatinExtendedA}/;
+ print "ok 601\n";
+}
+
+{
+ my $x = "\xff";
+
+ print "not " unless $x =~ /[\x80-\xff]/;
+ print "ok 602\n";
+
+ print "not " unless $x =~ /[\x80-\x{100}]/;
+ print "ok 603\n";
+
+ print "not " if $x =~ /[\x{100}]/;
+ print "ok 604\n";
+
+ print "not " unless $x =~ /\p{InLatin1Supplement}/;
+ print "ok 605\n";
+
+ print "not " if $x =~ /\P{InLatin1Supplement}/;
+ print "ok 606\n";
+
+ print "not " if $x =~ /\p{InLatinExtendedA}/;
+ print "ok 607\n";
+
+ print "not " unless $x =~ /\P{InLatinExtendedA}/;
+ print "ok 608\n";
+}
+
+{
+ my $x = "\x{100}";
+
+ print "not " if $x =~ /[\x80-\xff]/;
+ print "ok 609\n";
+
+ print "not " unless $x =~ /[\x80-\x{100}]/;
+ print "ok 610\n";
+
+ print "not " unless $x =~ /[\x{100}]/;
+ print "ok 611\n";
+
+ print "not " if $x =~ /\p{InLatin1Supplement}/;
+ print "ok 612\n";
+
+ print "not " unless $x =~ /\P{InLatin1Supplement}/;
+ print "ok 613\n";
+
+ print "not " unless $x =~ /\p{InLatinExtendedA}/;
+ print "ok 614\n";
+
+ print "not " if $x =~ /\P{InLatinExtendedA}/;
+ print "ok 615\n";
+}
+
+{
+ # from japhy
+ my $w;
+ use warnings;
+ local $SIG{__WARN__} = sub { $w .= shift };
+
+ $w = "";
+ eval 'qr/(?c)/';
+ print "not " if $w !~ /^Useless \(\?c\)/;
+ print "ok 616\n";
+
+ $w = "";
+ eval 'qr/(?-c)/';
+ print "not " if $w !~ /^Useless \(\?-c\)/;
+ print "ok 617\n";
+
+ $w = "";
+ eval 'qr/(?g)/';
+ print "not " if $w !~ /^Useless \(\?g\)/;
+ print "ok 618\n";
+
+ $w = "";
+ eval 'qr/(?-g)/';
+ print "not " if $w !~ /^Useless \(\?-g\)/;
+ print "ok 619\n";
+
+ $w = "";
+ eval 'qr/(?o)/';
+ print "not " if $w !~ /^Useless \(\?o\)/;
+ print "ok 620\n";
+
+ $w = "";
+ eval 'qr/(?-o)/';
+ print "not " if $w !~ /^Useless \(\?-o\)/;
+ print "ok 621\n";
+
+ # now test multi-error regexes
+
+ $w = "";
+ eval 'qr/(?g-o)/';
+ print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-o\)/;
+ print "ok 622\n";
+
+ $w = "";
+ eval 'qr/(?g-c)/';
+ print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-c\)/;
+ print "ok 623\n";
+
+ $w = "";
+ eval 'qr/(?o-cg)/'; # (?c) means (?g) error won't be thrown
+ print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?-c\)/;
+ print "ok 624\n";
+
+ $w = "";
+ eval 'qr/(?ogc)/';
+ print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?g\).*\nUseless \(\?c\)/;
+ print "ok 625\n";
+}
+
+# More Unicode "class" tests
+
+{
+ use charnames ':full';
+
+ print "not " unless "\N{LATIN CAPITAL LETTER A}" =~ /\p{InBasicLatin}/;
+ print "ok 626\n";
+
+ print "not " unless "\N{LATIN CAPITAL LETTER A WITH GRAVE}" =~ /\p{InLatin1Supplement}/;
+ print "ok 627\n";
+
+ print "not " unless "\N{LATIN CAPITAL LETTER A WITH MACRON}" =~ /\p{InLatinExtendedA}/;
+ print "ok 628\n";
+
+ print "not " unless "\N{LATIN SMALL LETTER B WITH STROKE}" =~ /\p{InLatinExtendedB}/;
+ print "ok 629\n";
+
+ print "not " unless "\N{KATAKANA LETTER SMALL A}" =~ /\p{InKatakana}/;
+ print "ok 630\n";
+}
+
+$_ = "foo";
+
+eval <<"EOT"; die if $@;
+ /f
+ o\r
+ o
+ \$
+ /x && print "ok 631\n";
+EOT
+
+eval <<"EOT"; die if $@;
+ /f
+ o
+ o
+ \$\r
+ /x && print "ok 632\n";
+EOT
+
+#test /o feature
+sub test_o { $_[0] =~/$_[1]/o; return $1}
+if(test_o('abc','(.)..') eq 'a') {
+ print "ok 633\n";
+} else {
+ print "not ok 633\n";
+}
+if(test_o('abc','..(.)') eq 'a') {
+ print "ok 634\n";
+} else {
+ print "not ok 634\n";
+}
+
+# 635..639: ID 20010619.003 (only the space character is
+# supposed to be [:print:], not the whole isprint()).
+
+print "not " if "\n" =~ /[[:print:]]/;
+print "ok 635\n";
+
+print "not " if "\t" =~ /[[:print:]]/;
+print "ok 636\n";
+
+# Amazingly vertical tabulator is the same in ASCII and EBCDIC.
+print "not " if "\014" =~ /[[:print:]]/;
+print "ok 637\n";
+
+print "not " if "\r" =~ /[[:print:]]/;
+print "ok 638\n";
+
+print "not " unless " " =~ /[[:print:]]/;
+print "ok 639\n";
+
+##
+## Test basic $^N usage outside of a regex
+##
+$x = "abcdef";
+$T="ok 640\n";if ($x =~ /cde/ and not defined $^N) {print $T} else {print "not $T"};
+$T="ok 641\n";if ($x =~ /(cde)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 642\n";if ($x =~ /(c)(d)(e)/ and $^N eq "e") {print $T} else {print "not $T"};
+$T="ok 643\n";if ($x =~ /(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 644\n";if ($x =~ /(foo)|(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 645\n";if ($x =~ /(c(d)e)|(foo)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 646\n";if ($x =~ /(c(d)e)|(abc)/ and $^N eq "abc") {print $T} else {print "not $T"};
+$T="ok 647\n";if ($x =~ /(c(d)e)|(abc)x/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 648\n";if ($x =~ /(c(d)e)(abc)?/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 649\n";if ($x =~ /(?:c(d)e)/ and $^N eq "d" ) {print $T} else {print "not $T"};
+$T="ok 650\n";if ($x =~ /(?:c(d)e)(?:f)/ and $^N eq "d" ) {print $T} else {print "not $T"};
+$T="ok 651\n";if ($x =~ /(?:([abc])|([def]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
+$T="ok 652\n";if ($x =~ /(?:([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
+$T="ok 653\n";if ($x =~ /(([ace])|([bd]))*/ and $^N eq "e" ){print $T} else {print "not $T"};
+{
+ $T="ok 654\n";if($x =~ /(([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
+}
+## test to see if $^N is automatically localized -- it should now
+## have the value set in test 653
+$T="ok 655\n";if ($^N eq "e" ){print $T} else {print "not $T"};
+
+##
+## Now test inside (?{...})
+##
+$T="ok 656\n";if ($x =~ /a([abc])(?{$y=$^N})c/ and $y eq "b" ){print $T} else {print "not $T"};
+$T="ok 657\n";if ($x =~ /a([abc]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
+$T="ok 658\n";if ($x =~ /a([abcdefg]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
+$T="ok 659\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq "bc" and $z eq "abcd")
+ {print $T} else {print "not $T"};
+$T="ok 660\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq "bc" and $z eq "abcde")
+ {print $T} else {print "not $T"};
+
+# Test the Unicode script classes
+
+print "not " unless chr(0x100) =~ /\p{InLatin}/; # outside Latin-1
+print "ok 661\n";
+
+print "not " unless chr(0x212b) =~ /\p{InLatin}/; # Angstrom sign, very outside
+print "ok 662\n";
+
+print "not " unless chr(0x5d0) =~ /\p{InHebrew}/; # inside HebrewBlock
+print "ok 663\n";
+
+print "not " unless chr(0xfb4f) =~ /\p{InHebrew}/; # outside HebrewBlock
+print "ok 664\n";
+
+print "not " unless chr(0xb5) =~ /\p{InGreek}/; # singleton (not in a range)
+print "ok 665\n";
+
+print "not " unless chr(0x37a) =~ /\p{InGreek}/; # singleton
+print "ok 666\n";
+
+print "not " unless chr(0x386) =~ /\p{InGreek}/; # singleton
+print "ok 667\n";
+
+print "not " unless chr(0x387) =~ /\P{InGreek}/; # not there
+print "ok 668\n";
+
+print "not " unless chr(0x388) =~ /\p{InGreek}/; # range
+print "ok 669\n";
+
+print "not " unless chr(0x38a) =~ /\p{InGreek}/; # range
+print "ok 670\n";
+
+print "not " unless chr(0x38b) =~ /\P{InGreek}/; # not there
+print "ok 671\n";
+
+print "not " unless chr(0x38c) =~ /\p{InGreek}/; # singleton
+print "ok 672\n";
+
+##
+## Test [:cntrl:]...
+##
+## Should probably put in tests for all the POSIX stuff, but not sure how to
+## guarantee a specific locale......
+##
+$AllBytes = join('', map { chr($_) } 0..255);
+($x = $AllBytes) =~ s/[[:cntrl:]]//g;
+if ($x ne join('', map { chr($_) } 0x20..0x7E, 0x80..0xFF)) { print "not " };
+print "ok 673\n";
+
+($x = $AllBytes) =~ s/[^[:cntrl:]]//g;
+if ($x ne join('', map { chr($_) } 0..0x1F, 0x7F)) { print "not " };
+print "ok 674\n";
+
+# With /s modifier UTF8 chars were interpreted as bytes
+{
+ my $a = "Hello \x{263A} World";
+
+ my @a = ($a =~ /./gs);
+
+ print "not " unless $#a == 12;
+ print "ok 675\n";
+}
+
+@a = ("foo\nbar" =~ /./g);
+print "ok 676\n" if @a == 6 && "@a" eq "f o o b a r";
+
+@a = ("foo\nbar" =~ /./gs);
+print "ok 677\n" if @a == 7 && "@a" eq "f o o \n b a r";
+
+@a = ("foo\nbar" =~ /\C/g);
+print "ok 678\n" if @a == 7 && "@a" eq "f o o \n b a r";
+
+@a = ("foo\nbar" =~ /\C/gs);
+print "ok 679\n" if @a == 7 && "@a" eq "f o o \n b a r";
+
+@a = ("foo\n\x{100}bar" =~ /./g);
+print "ok 680\n" if @a == 7 && "@a" eq "f o o \x{100} b a r";
+
+@a = ("foo\n\x{100}bar" =~ /./gs);
+print "ok 681\n" if @a == 8 && "@a" eq "f o o \n \x{100} b a r";
+
+($a, $b) = map { chr } ord('A') == 65 ? (0xc4, 0x80) : (0x8c, 0x41);
+
+@a = ("foo\n\x{100}bar" =~ /\C/g);
+print "ok 682\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
+
+@a = ("foo\n\x{100}bar" =~ /\C/gs);
+print "ok 683\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
+
+{
+ # [ID 20010814.004] pos() doesn't work when using =~m// in list context
+ $_ = "ababacadaea";
+ $a = join ":", /b./gc;
+ $b = join ":", /a./gc;
+ $c = pos;
+ print "$a $b $c" eq 'ba:ba ad:ae 10' ? "ok 684\n" : "not ok 684\t# $a $b $c\n";
+}
+
+{
+ # [ID 20010407.006] matching utf8 return values from functions does not work
+
+ package ID_20010407_006;
+
+ sub x {
+ "a\x{1234}";
+ }
+
+ my $x = x;
+ my $y;
+
+ $x =~ /(..)/; $y = $1;
+ print "not " unless length($y) == 2 && $y eq $x;
+ print "ok 685\n";
+
+ x =~ /(..)/; $y = $1;
+ print "not " unless length($y) == 2 && $y eq $x;
+ print "ok 686\n";
+}
+
+
+my $test = 687;
+
+# Force scalar context on the patern match
+sub ok ($$) {
+ my($ok, $name) = @_;
+
+ printf "%sok %d - %s\n", ($ok ? "" : "not "), $test, $name;
+
+ printf "# Failed test at line %d\n", (caller)[2] unless $ok;
+
+ $test++;
+ return $ok;
+}
+
+{
+ # Check that \x## works. 5.6.1 and 5.005_03 fail some of these.
+ $x = "\x4e" . "E";
+ ok ($x =~ /^\x4EE$/, "Check only 2 bytes of hex are matched.");
+
+ $x = "\x4e" . "i";
+ ok ($x =~ /^\x4Ei$/, "Check that invalid hex digit stops it (2)");
+
+ $x = "\x4" . "j";
+ ok ($x =~ /^\x4j$/, "Check that invalid hex digit stops it (1)");
+
+ $x = "\x0" . "k";
+ ok ($x =~ /^\xk$/, "Check that invalid hex digit stops it (0)");
+
+ $x = "\x0" . "x";
+ ok ($x =~ /^\xx$/, "\\xx isn't to be treated as \\0");
+
+ $x = "\x0" . "xa";
+ ok ($x =~ /^\xxa$/, "\\xxa isn't to be treated as \\xa");
+
+ $x = "\x9" . "_b";
+ ok ($x =~ /^\x9_b$/, "\\x9_b isn't to be treated as \\x9b");
+
+ print "# and now again in [] ranges\n";
+
+ $x = "\x4e" . "E";
+ ok ($x =~ /^[\x4EE]{2}$/, "Check only 2 bytes of hex are matched.");
+
+ $x = "\x4e" . "i";
+ ok ($x =~ /^[\x4Ei]{2}$/, "Check that invalid hex digit stops it (2)");
+
+ $x = "\x4" . "j";
+ ok ($x =~ /^[\x4j]{2}$/, "Check that invalid hex digit stops it (1)");
+
+ $x = "\x0" . "k";
+ ok ($x =~ /^[\xk]{2}$/, "Check that invalid hex digit stops it (0)");
+
+ $x = "\x0" . "x";
+ ok ($x =~ /^[\xx]{2}$/, "\\xx isn't to be treated as \\0");
+
+ $x = "\x0" . "xa";
+ ok ($x =~ /^[\xxa]{3}$/, "\\xxa isn't to be treated as \\xa");
+
+ $x = "\x9" . "_b";
+ ok ($x =~ /^[\x9_b]{3}$/, "\\x9_b isn't to be treated as \\x9b");
+
+}
+
+{
+ # Check that \x{##} works. 5.6.1 fails quite a few of these.
+
+ $x = "\x9b";
+ ok ($x =~ /^\x{9_b}$/, "\\x{9_b} is to be treated as \\x9b");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^\x{9_b}y$/, "\\x{9_b} is to be treated as \\x9b (again)");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^\x{9b_}y$/, "\\x{9b_} is to be treated as \\x9b");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^\x{9_bq}y$/, "\\x{9_bc} is to be treated as \\x9b");
+
+ $x = "\x0" . "y";
+ ok ($x =~ /^\x{x9b}y$/, "\\x{x9b} is to be treated as \\x0");
+
+ $x = "\x0" . "y";
+ ok ($x =~ /^\x{0x9b}y$/, "\\x{0x9b} is to be treated as \\x0");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^\x{09b}y$/, "\\x{09b} is to be treated as \\x9b");
+
+ print "# and now again in [] ranges\n";
+
+ $x = "\x9b";
+ ok ($x =~ /^[\x{9_b}]$/, "\\x{9_b} is to be treated as \\x9b");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^[\x{9_b}y]{2}$/, "\\x{9_b} is to be treated as \\x9b (again)");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^[\x{9b_}y]{2}$/, "\\x{9b_} is to be treated as \\x9b");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^[\x{9_bq}y]{2}$/, "\\x{9_bc} is to be treated as \\x9b");
+
+ $x = "\x0" . "y";
+ ok ($x =~ /^[\x{x9b}y]{2}$/, "\\x{x9b} is to be treated as \\x0");
+
+ $x = "\x0" . "y";
+ ok ($x =~ /^[\x{0x9b}y]{2}$/, "\\x{0x9b} is to be treated as \\x0");
+
+ $x = "\x9b" . "y";
+ ok ($x =~ /^[\x{09b}y]{2}$/, "\\x{09b} is to be treated as \\x9b");
+}
+
+{
+ # high bit bug -- japhy
+ my $x = "ab\200d";
+ $x =~ /.*?\200/ or print "not ";
+ print "ok 715\n";
+}
+