6 $ENV{PERL5LIB} = '../lib';
7 if ( ord("\t") != 9 ) { # skip on ebcdic platforms
8 print "1..0 # Skip utf8 tests on ebcdic platform.\n";
15 # Think carefully before adding tests here. In general this should be
16 # used only for about three categories of tests:
18 # (1) tests that absolutely require 'use utf8', and since that in general
19 # shouldn't be needed as the utf8 is being obsoleted, this should
20 # have rather few tests. If you want to test Unicode and regexes,
21 # you probably want to go to op/regexp or op/pat; if you want to test
22 # split, go to op/split; pack, op/pack; appending or joining,
23 # op/append or op/join, and so forth
25 # (2) tests that have to do with Unicode tokenizing (though it's likely
26 # that all the other Unicode tests sprinkled around the t/**/*.t are
27 # going to catch that)
29 # (3) complicated tests that simultaneously stress so many Unicode features
30 # that deciding into which other test script the tests should go to
31 # is hard -- maybe consider breaking up the complicated test
43 { use bytes; $a = "\xc3\xa4" }
44 { use utf8; $b = "\xe4" }
50 { use utf8; ok($a ne $b) }
57 my $smiley = "\x{263a}";
59 for my $s ("\x{263a}",
68 my $length_chars = length($s);
70 { use bytes; $length_bytes = length($s) }
71 my @regex_chars = $s =~ m/(.)/g;
72 my $regex_chars = @regex_chars;
73 my @split_chars = split //, $s;
74 my $split_chars = @split_chars;
75 ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq
79 for my $s ("\x{263a}" . "\x{263a}",
88 my $length_chars = length($s);
90 { use bytes; $length_bytes = length($s) }
91 my @regex_chars = $s =~ m/(.)/g;
92 my $regex_chars = @regex_chars;
93 my @split_chars = split //, $s;
94 my $split_chars = @split_chars;
95 ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq
103 local $SIG{__WARN__} = sub { print "#($_[0])\n"; $w++ };
104 my $x = eval q/"\\/ . "\x{100}" . q/"/;;
106 ok($w == 0 && $x eq "\x{100}");