summaryrefslogtreecommitdiff
path: root/t/op/pat.t
diff options
context:
space:
mode:
Diffstat (limited to 't/op/pat.t')
-rwxr-xr-xt/op/pat.t43
1 files changed, 28 insertions, 15 deletions
diff --git a/t/op/pat.t b/t/op/pat.t
index 467e0a29d0..d9e8c3d43d 100755
--- a/t/op/pat.t
+++ b/t/op/pat.t
@@ -6,7 +6,7 @@
$| = 1;
-print "1..846\n";
+print "1..854\n";
BEGIN {
chdir 't' if -d 't';
@@ -2602,33 +2602,37 @@ print "# some Unicode properties\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 "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 "\N{LATIN SMALL LETTER SHARP S}" =~ /ss/i ?
+ "ok 843\n" : "not ok 843\n";
+
+ print "\N{LATIN SMALL LETTER SHARP S}" =~ /SS/i ?
+ "ok 844\n" : "not ok 844\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 "<\x{100}\x{0085}>" =~ /<\x{100}\s>/ ? "ok 845\n" : "not ok 845\n";
+ print "<\x{2028}>" =~ /<\s>/ ? "ok 846\n" : "not ok 846\n";
+ print "<\x{2029}>" =~ /<\s>/ ? "ok 847\n" : "not ok 847\n";
}
{
- print "# . with /s should work on characters, not bytes\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";
+ "not ok 848\n" : "ok 848\n";
my @c;
@@ -2636,7 +2640,7 @@ print "# some Unicode properties\n";
push @c, $1;
}
- print join("", @c) eq $s ? "ok 845\n" : "not ok 845\n";
+ print join("", @c) eq $s ? "ok 849\n" : "not ok 849\n";
my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}"; # test only chars < 256
my $r1 = "";
@@ -2650,5 +2654,14 @@ print "# some Unicode properties\n";
$r2 .= $1 . $2;
}
$r2 =~ s/\x{100}//;
- print $r1 eq $r2 ? "ok 846\n" : "not ok 846\n";
+ print $r1 eq $r2 ? "ok 850\n" : "not ok 850\n";
+}
+
+{
+ print "# Unicode lookbehind\n";
+
+ print "A\x{100}B" =~ /(?<=A.)B/ ? "ok 851\n" : "not ok 851\n";
+ print "A\x{200}\x{300}B" =~ /(?<=A..)B/ ? "ok 852\n" : "not ok 852\n";
+ print "\x{400}AB" =~ /(?<=\x{400}.)B/ ? "ok 853\n" : "not ok 853\n";
+ print "\x{500\x{600}}B" =~ /(?<=\x{500}.)B/ ? "ok 854\n" : "not ok 854\n";
}