$| = 1;
-print "1..903\n";
+print "1..941\n";
BEGIN {
chdir 't' if -d 't';
}
}
-
-{
- print "# Unicode hash keys and \\w\n";
- # This is not really a regex test but regexes bring
- # out the issue nicely.
- use strict;
- my $u3 = "f\x{df}\x{100}";
- my $u2 = substr($u3,0,2);
- my $u1 = substr($u2,0,1);
- my %u = ( $u1 => $u1, $u2 => $u2, $u3 => $u3 );
-
- for (keys %u) {
- print /^\w+$/ && $u{$_} =~ /^\w+$/ ?
- "ok $test\n" : "not ok $test\n";
- $test++;
- }
-
- for (each %u) {
- print /^\w+$/ && $u{$_} =~ /^\w+$/ ?
- "ok $test\n" : "not ok $test\n";
- $test++;
- }
-
- for (%u) {
- print /^\w+$/ && $u{$_} =~ /^\w+$/ ?
- "ok $test\n" : "not ok $test\n";
- $test++;
- }
-}
-
{
print "# qr/.../x\n";
+ my $test = 893;
my $R = qr/ A B C # D E/x;
print eval {"ABCDE" =~ m/($R)/} ? "ok $test\n" : "not ok $test\n";
$test++;
}
+
+{
+ print "# illegal Unicode properties\n";
+ my $test = 896;
+
+ print eval { "a" =~ /\pq / } ? "not ok $test\n" : "ok $test\n";
+ $test++;
+
+ print eval { "a" =~ /\p{qrst} / } ? "not ok $test\n" : "ok $test\n";
+ $test++;
+}
+
+{
+ print "# [ID 20020412.005] wrong pmop flags checked when empty pattern\n";
+ # requires reuse of last successful pattern
+ my $test = 898;
+ $test =~ /\d/;
+ for (0 .. 1) {
+ my $match = ?? + 0;
+ if ($match != $_) {
+ print "ok $test\n";
+ } else {
+ printf "not ok %s\t# 'match once' %s on %s iteration\n", $test,
+ $match ? 'succeeded' : 'failed', $_ ? 'second' : 'first';
+ }
+ ++$test;
+ }
+ $test =~ /(\d)/;
+ my $result = join '', $test =~ //g;
+ if ($result eq $test) {
+ print "ok $test\n";
+ } else {
+ printf "not ok %s\t# expected '%s', got '%s'\n", $test, $test, $result;
+ }
+ ++$test;
+}
+
+print "# user-defined character properties\n";
+
+sub InKana1 {
+ return <<'END';
+3040 309F
+30A0 30FF
+END
+}
+
+sub InKana2 {
+ return <<'END';
++utf8::InHiragana
++utf8::InKatakana
+END
+}
+
+sub InKana3 {
+ return <<'END';
++utf8::InHiragana
++utf8::InKatakana
+-utf8::IsCn
+END
+}
+
+sub InNotKana {
+ return <<'END';
+!utf8::InHiragana
+-utf8::InKatakana
++utf8::IsCn
+END
+}
+
+$test = 901;
+
+print "\x{3040}" =~ /\p{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "\x{303F}" =~ /\P{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+print "\x{3040}" =~ /\p{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "\x{303F}" =~ /\P{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+print "\x{3041}" =~ /\p{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "\x{3040}" =~ /\P{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+print "\x{3040}" =~ /\p{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "\x{3041}" =~ /\P{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+sub InConsonant { # Not EBCDIC-aware.
+ return <<EOF;
+0061 007f
+-0061
+-0065
+-0069
+-006f
+-0075
+EOF
+}
+
+print "d" =~ /\p{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "e" =~ /\P{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+{
+ print "# [ID 20020630.002] utf8 regex only matches 32k\n";
+ $test = 911;
+ for ([ 'byte', "\x{ff}" ], [ 'utf8', "\x{1ff}" ]) {
+ my($type, $char) = @$_;
+ for my $len (32000, 32768, 33000) {
+ my $s = $char . "f" x $len;
+ my $r = $s =~ /$char([f]*)/gc;
+ print $r ? "ok $test\n" : "not ok $test\t# <$type x $len> fail\n";
+ ++$test;
+ print +(!$r or pos($s) == $len + 1) ? "ok $test\n"
+ : "not ok $test\t# <$type x $len> pos @{[ pos($s) ]}\n";
+ ++$test;
+ }
+ }
+}
+
+$test = 923;
+
+$a = bless qr/foo/, 'Foo';
+print(('goodfood' =~ $a ? '' : 'not '),
+ "ok $test\t# reblessed qr// matches\n");
+++$test;
+
+print(($a eq '(?-xism:foo)' ? '' : 'not '),
+ "ok $test\t# reblessed qr// stringizes\n");
+++$test;
+
+$x = "\x{3fe}";
+$z=$y = "\317\276"; # $y is byte representation of $x
+
+$a = qr/$x/;
+print(($x =~ $a ? '' : 'not '), "ok $test - utf8 interpolation in qr//\n");
+++$test;
+
+print(("a$a" =~ $x ? '' : 'not '),
+ "ok $test - stringifed qr// preserves utf8\n");
+++$test;
+
+print(("a$x" =~ /^a$a\z/ ? '' : 'not '),
+ "ok $test - interpolated qr// preserves utf8\n");
+++$test;
+
+print(("a$x" =~ /^a(??{$a})\z/ ? '' : 'not '),
+ "ok $test - postponed interpolation of qr// preserves utf8\n");
+++$test;
+
+print((length(qr/##/x) == 12 ? '' : 'not '),
+ "ok $test - ## in qr// doesn't corrupt memory [perl #17776]\n");
+++$test;
+
+{ use re 'eval';
+
+print(("$x$x" =~ /^$x(??{$x})\z/ ? '' : 'not '),
+ "ok $test - postponed utf8 string in utf8 re matches utf8\n");
+++$test;
+
+print(("$y$x" =~ /^$y(??{$x})\z/ ? '' : 'not '),
+ "ok $test - postponed utf8 string in non-utf8 re matches utf8\n");
+++$test;
+
+print(("$y$x" !~ /^$y(??{$y})\z/ ? '' : 'not '),
+ "ok $test - postponed non-utf8 string in non-utf8 re doesn't match utf8\n");
+++$test;
+
+print(("$x$x" !~ /^$x(??{$y})\z/ ? '' : 'not '),
+ "ok $test - postponed non-utf8 string in utf8 re doesn't match utf8\n");
+++$test;
+
+print(("$y$y" =~ /^$y(??{$y})\z/ ? '' : 'not '),
+ "ok $test - postponed non-utf8 string in non-utf8 re matches non-utf8\n");
+++$test;
+
+print(("$x$y" =~ /^$x(??{$y})\z/ ? '' : 'not '),
+ "ok $test - postponed non-utf8 string in utf8 re matches non-utf8\n");
+++$test;
+$y = $z; # reset $y after upgrade
+
+print(("$x$y" !~ /^$x(??{$x})\z/ ? '' : 'not '),
+ "ok $test - postponed utf8 string in utf8 re doesn't match non-utf8\n");
+++$test;
+$y = $z; # reset $y after upgrade
+
+print(("$y$y" !~ /^$y(??{$x})\z/ ? '' : 'not '),
+ "ok $test - postponed utf8 string in non-utf8 re doesn't match non-utf8\n");
+++$test;
+
+} # no re 'eval'
+
+print "# more user-defined character properties\n";
+
+sub IsSyriac1 {
+ return <<'END';
+0712 072C
+0730 074A
+END
+}
+
+print "\x{0712}" =~ /\p{IsSyriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "\x{072F}" =~ /\P{IsSyriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+sub Syriac1 {
+ return <<'END';
+0712 072C
+0730 074A
+END
+}
+
+print "\x{0712}" =~ /\p{Syriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "\x{072F}" =~ /\P{Syriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+
+# last test 940