$| = 1;
-print "1..828\n";
+print "1..849\n";
BEGIN {
chdir 't' if -d 't';
print "# GREEK CAPITAL LETTER SIGMA vs COMBINING GREEK PERISPOMENI\n";
my $SIGMA = "\N{GREEK CAPITAL LETTER SIGMA}";
-
- my $hSIGMA = sprintf "%04x", ord $SIGMA;
-
- my $char = "\N{COMBINING GREEK PERISPOMENI}";
- my $code = sprintf "%04x", ord($char);
+ my $char = "\N{COMBINING GREEK PERISPOMENI}";
# Before #13843 this was failing by matching falsely.
print "_:$char:_" =~ m/_:$SIGMA:_/i ? "not ok 786\n" : "ok 786\n";
}
}
}
+
+{
+ print "# more SIGMAs\n";
+
+ my $SIGMA = "\x{03A3}"; # CAPITAL
+ my $Sigma = "\x{03C2}"; # SMALL FINAL
+ my $sigma = "\x{03C3}"; # SMALL
+
+ my $S3 = "$SIGMA$Sigma$sigma";
+
+ print ":$S3:" =~ /:(($SIGMA)+):/i && $1 eq $S3 && $2 eq $sigma ?
+ "ok 829\n" : "not ok 829\n";
+ print ":$S3:" =~ /:(($Sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
+ "ok 830\n" : "not ok 830\n";
+ print ":$S3:" =~ /:(($sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
+ "ok 831\n" : "not ok 831\n";
+
+ print ":$S3:" =~ /:(([$SIGMA])+):/i && $1 eq $S3 && $2 eq $sigma ?
+ "ok 832\n" : "not ok 832\n";
+ print ":$S3:" =~ /:(([$Sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
+ "ok 833\n" : "not ok 833\n";
+ print ":$S3:" =~ /:(([$sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
+ "ok 834\n" : "not ok 834\n";
+}
+
+{
+ print "# LATIN SMALL LETTER SHARP S\n";
+
+ use charnames ':full';
+
+ print "\N{LATIN SMALL LETTER SHARP S}" =~
+ /\N{LATIN SMALL LETTER SHARP S}/ ? "ok 835\n" : "not ok 835\n";
+
+ print "\N{LATIN SMALL LETTER SHARP S}" =~
+ /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 836\n" : "not ok 836\n";
+
+ print "\N{LATIN SMALL LETTER SHARP S}" =~
+ /[\N{LATIN SMALL LETTER SHARP S}]/ ? "ok 837\n" : "not ok 837\n";
+
+ print "\N{LATIN SMALL LETTER SHARP S}" =~
+ /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 838\n" : "not ok 838\n";
+
+ print "ss" =~
+ /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 839\n" : "not ok 839\n";
+
+ print "SS" =~
+ /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 840\n" : "not ok 840\n";
+
+# These are a bit tricky. Since the LATIN SMALL LETTER SHARP S is U+00DF,
+# the ANYOF reduces to a byte. The Unicodeness needs to be caught earlier.
+# print "ss" =~
+# /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 841\n" : "not ok 841\n";
+#
+# print "SS" =~
+# /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 842\n" : "not ok 842\n";
+}
+
+{
+ print "# more whitespace: U+0085, U+2028, U+2029\n";
+
+ # U+0085 needs to be forced to be Unicode, the \x{100} does that.
+ print "<\x{100}\x{0085}>" =~ /<\x{100}\s>/ ? "ok 841\n" : "not ok 841\n";
+ print "<\x{2028}>" =~ /<\s>/ ? "ok 842\n" : "not ok 842\n";
+ print "<\x{2029}>" =~ /<\s>/ ? "ok 843\n" : "not ok 843\n";
+}
+
+{
+ print "# . with /s should work on characters, as opposed to bytes\n";
+
+ my $s = "\x{e4}\x{100}";
+
+ # This is not expected to match: the point is that
+ # neither should we get "Malformed UTF-8" warnings.
+ print $s =~ /\G(.+?)\n/gcs ?
+ "not ok 844\n" : "ok 844\n";
+
+ my @c;
+
+ while ($s =~ /\G(.)/gs) {
+ push @c, $1;
+ }
+
+ print join("", @c) eq $s ? "ok 845\n" : "not ok 845\n";
+
+ my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}"; # test only chars < 256
+ my $r1 = "";
+ while ($t1 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
+ $r1 .= $1 . $2;
+ }
+
+ my $t2 = $t1 . "\x{100}"; # repeat with a larger char
+ my $r2 = "";
+ while ($t2 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
+ $r2 .= $1 . $2;
+ }
+ $r2 =~ s/\x{100}//;
+ print $r1 eq $r2 ? "ok 846\n" : "not ok 846\n";
+}
+
+{
+ print "# Unicode lookbehind\n";
+
+ print "A\x{100}B" =~ /(?<=A.)B/ ? "ok 847\n" : "not ok 847\n";
+ print "A\x{200}\x{300}B" =~ /(?<=A..)B/ ? "ok 848\n" : "not ok 848\n";
+}