Commit | Line | Data |
f96ec2a2 |
1 | #!./perl |
2 | |
3 | BEGIN { |
4 | chdir 't' if -d 't'; |
20822f61 |
5 | @INC = '../lib'; |
f96ec2a2 |
6 | $ENV{PERL5LIB} = '../lib'; |
f70c35af |
7 | if ( ord("\t") != 9 ) { # skip on ebcdic platforms |
8 | print "1..0 # Skip utf8 tests on ebcdic platform.\n"; |
9 | exit; |
10 | } |
f96ec2a2 |
11 | } |
12 | |
4765795a |
13 | # NOTE! |
14 | # |
15 | # Think carefully before adding tests here. In general this should be |
16 | # used only for about three categories of tests: |
17 | # |
18 | # (1) tests that absolutely require 'use utf8', and since that in general |
19 | # shouldn't be needed as the utf8 is being obsoleted, this should |
20 | # have rather few tests. If you want to test Unicode and regexes, |
21 | # you probably want to go to op/regexp or op/pat; if you want to test |
22 | # split, go to op/split; pack, op/pack; appending or joining, |
23 | # op/append or op/join, and so forth |
24 | # |
25 | # (2) tests that have to do with Unicode tokenizing (though it's likely |
26 | # that all the other Unicode tests sprinkled around the t/**/*.t are |
27 | # going to catch that) |
28 | # |
29 | # (3) complicated tests that simultaneously stress so many Unicode features |
30 | # that deciding into which other test script the tests should go to |
31 | # is hard -- maybe consider breaking up the complicated test |
32 | # |
33 | # |
34 | |
35 | use Test; |
36 | plan tests => 15; |
31067593 |
37 | |
7bbb0251 |
38 | { |
da450f52 |
39 | # bug id 20001009.001 |
40 | |
89491803 |
41 | my ($a, $b); |
42 | |
43 | { use bytes; $a = "\xc3\xa4" } |
4765795a |
44 | { use utf8; $b = "\xe4" } |
89491803 |
45 | |
4765795a |
46 | my $test = 68; |
31067593 |
47 | |
4765795a |
48 | ok($a ne $b); |
da450f52 |
49 | |
4765795a |
50 | { use utf8; ok($a ne $b) } |
da450f52 |
51 | } |
52 | |
60ff4832 |
53 | |
54 | { |
55 | # bug id 20000730.004 |
56 | |
60ff4832 |
57 | my $smiley = "\x{263a}"; |
58 | |
4765795a |
59 | for my $s ("\x{263a}", |
60 | $smiley, |
60ff4832 |
61 | |
4765795a |
62 | "" . $smiley, |
63 | "" . "\x{263a}", |
60ff4832 |
64 | |
4765795a |
65 | $smiley . "", |
66 | "\x{263a}" . "", |
60ff4832 |
67 | ) { |
68 | my $length_chars = length($s); |
69 | my $length_bytes; |
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 $split_chars = @split_chars; |
4765795a |
75 | ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq |
76 | "1/1/1/3"); |
60ff4832 |
77 | } |
78 | |
4765795a |
79 | for my $s ("\x{263a}" . "\x{263a}", |
80 | $smiley . $smiley, |
60ff4832 |
81 | |
4765795a |
82 | "\x{263a}\x{263a}", |
83 | "$smiley$smiley", |
60ff4832 |
84 | |
4765795a |
85 | "\x{263a}" x 2, |
86 | $smiley x 2, |
60ff4832 |
87 | ) { |
88 | my $length_chars = length($s); |
89 | my $length_bytes; |
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 $split_chars = @split_chars; |
4765795a |
95 | ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq |
96 | "2/2/2/6"); |
60ff4832 |
97 | } |
98 | } |
ffc61ed2 |
99 | |
ffc61ed2 |
100 | |
101 | { |
f9a63242 |
102 | my $w = 0; |
103 | local $SIG{__WARN__} = sub { print "#($_[0])\n"; $w++ }; |
104 | my $x = eval q/"\\/ . "\x{100}" . q/"/;; |
105 | |
4765795a |
106 | ok($w == 0 && $x eq "\x{100}"); |
f9a63242 |
107 | } |
108 | |