This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Update Changes.
[perl5.git] / lib / utf8.t
CommitLineData
f96ec2a2
GS
1#!./perl
2
3BEGIN {
4 chdir 't' if -d 't';
20822f61 5 @INC = '../lib';
8ae6c9f9 6 require './test.pl';
f96ec2a2
GS
7}
8
4765795a
JH
9# NOTE!
10#
11# Think carefully before adding tests here. In general this should be
12# used only for about three categories of tests:
13#
14# (1) tests that absolutely require 'use utf8', and since that in general
15# shouldn't be needed as the utf8 is being obsoleted, this should
16# have rather few tests. If you want to test Unicode and regexes,
17# you probably want to go to op/regexp or op/pat; if you want to test
18# split, go to op/split; pack, op/pack; appending or joining,
19# op/append or op/join, and so forth
20#
21# (2) tests that have to do with Unicode tokenizing (though it's likely
22# that all the other Unicode tests sprinkled around the t/**/*.t are
23# going to catch that)
24#
25# (3) complicated tests that simultaneously stress so many Unicode features
26# that deciding into which other test script the tests should go to
27# is hard -- maybe consider breaking up the complicated test
28#
29#
30
8ae6c9f9 31plan tests => 16;
31067593 32
7bbb0251 33{
da450f52
JH
34 # bug id 20001009.001
35
89491803
SC
36 my ($a, $b);
37
38 { use bytes; $a = "\xc3\xa4" }
4765795a 39 { use utf8; $b = "\xe4" }
89491803 40
4765795a 41 my $test = 68;
31067593 42
4765795a 43 ok($a ne $b);
da450f52 44
4765795a 45 { use utf8; ok($a ne $b) }
da450f52
JH
46}
47
60ff4832
JH
48
49{
50 # bug id 20000730.004
51
60ff4832
JH
52 my $smiley = "\x{263a}";
53
4765795a
JH
54 for my $s ("\x{263a}",
55 $smiley,
60ff4832 56
4765795a
JH
57 "" . $smiley,
58 "" . "\x{263a}",
60ff4832 59
4765795a
JH
60 $smiley . "",
61 "\x{263a}" . "",
60ff4832
JH
62 ) {
63 my $length_chars = length($s);
64 my $length_bytes;
65 { use bytes; $length_bytes = length($s) }
66 my @regex_chars = $s =~ m/(.)/g;
67 my $regex_chars = @regex_chars;
68 my @split_chars = split //, $s;
69 my $split_chars = @split_chars;
4765795a
JH
70 ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq
71 "1/1/1/3");
60ff4832
JH
72 }
73
4765795a
JH
74 for my $s ("\x{263a}" . "\x{263a}",
75 $smiley . $smiley,
60ff4832 76
4765795a
JH
77 "\x{263a}\x{263a}",
78 "$smiley$smiley",
60ff4832 79
4765795a
JH
80 "\x{263a}" x 2,
81 $smiley x 2,
60ff4832
JH
82 ) {
83 my $length_chars = length($s);
84 my $length_bytes;
85 { use bytes; $length_bytes = length($s) }
86 my @regex_chars = $s =~ m/(.)/g;
87 my $regex_chars = @regex_chars;
88 my @split_chars = split //, $s;
89 my $split_chars = @split_chars;
4765795a
JH
90 ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq
91 "2/2/2/6");
60ff4832
JH
92 }
93}
ffc61ed2 94
ffc61ed2
JH
95
96{
f9a63242
JH
97 my $w = 0;
98 local $SIG{__WARN__} = sub { print "#($_[0])\n"; $w++ };
99 my $x = eval q/"\\/ . "\x{100}" . q/"/;;
100
4765795a 101 ok($w == 0 && $x eq "\x{100}");
f9a63242
JH
102}
103
8ae6c9f9
NC
104{
105 my $progfile = 'utf' . $$;
106 END {unlink $progfile}
107 open P, ">$progfile" or die "Can't open '$progfile': $!";
108 # Interpolation of hex characters needs to take place now, as we're
109 # testing feeding malformed utf8 into perl. Bug now fixed was an
110 # "out of memory" error. We really need the "" [rather than qq() or q()]
111 # to get the best explosion.
112 print P <<"BANG";
113 use utf8; %a = ("\xE1\xA0"=>"sterling");
114 print 'start'; printf '%x,', ord $_ foreach keys %a; print "end\n";
115BANG
116 print "# Possible delay...\n";
117 my $result = runperl ( verbose => 1, stderr => 1, progfile => $progfile );
118 like ($result,
119 qr/^Malformed UTF-8 character \(2 bytes, need 3\).*start\d+,end$/s);
120}