11 # Think carefully before adding tests here. In general this should be
12 # used only for about three categories of tests:
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
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)
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
38 { use bytes; $a = "\xc3\xa4" }
39 { use utf8; $b = "\xe4" }
45 { use utf8; ok($a ne $b) }
52 my $smiley = "\x{263a}";
54 for my $s ("\x{263a}",
63 my $length_chars = length($s);
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 $split_chars = @split_chars;
70 ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq
74 for my $s ("\x{263a}" . "\x{263a}",
83 my $length_chars = length($s);
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 $split_chars = @split_chars;
90 ok("$length_chars/$regex_chars/$split_chars/$length_bytes" eq
98 local $SIG{__WARN__} = sub { print "#($_[0])\n"; $w++ };
99 my $x = eval q/"\\/ . "\x{100}" . q/"/;;
101 ok($w == 0 && $x eq "\x{100}");
106 my $progfile = 'utf' . $$;
107 END {unlink_all $progfile}
109 # If I'm right 60 is '>' in ASCII, ' ' in EBCDIC
110 # 173 is not punctuation in either ASCII or EBCDIC
112 foreach (60, 173, 257, 65532) {
115 # I don't want to use map {ord} and I've no need to hardcode the UTF
117 my $charsubst = $char;
118 $charsubst =~ s/(.)/ord ($1) . ','/ge;
120 push @char, [$_, $char, $charsubst];
123 ['check our detection program works',
124 '@a = ("'.chr(60).'\x2A", ""); display @a', qr/^>60,42<><$/],
125 ['check literal 8 bit input',
126 '$a = "' . chr (173) . '"; display $a', qr/^>173<$/],
127 ['check no utf8; makes no change',
128 'no utf8; $a = "' . chr (173) . '"; display $a', qr/^>173<$/],
129 # Now we do the real byte sequences that are valid UTF8
131 ["the utf8 sequence for chr $_->[0]",
132 qq(\$a = "$_->[1]"; display \$a), qr/^>$_->[2]<$/],
133 ["no utf8; for the utf8 sequence for chr $_->[0]",
134 qq(no utf8; \$a = "$_->[1]"; display \$a), qr/^>$_->[2]<$/],
135 ["use utf8; for the utf8 sequence for chr $_->[0]",
136 qq(use utf8; \$a = "$_->[1]"; display \$a), qr/^>$_->[0]<$/],
138 # Interpolation of hex characters needs to take place now, as we're
139 # testing feeding malformed utf8 into perl. Bug now fixed was an
140 # "out of memory" error. We really need the "" [rather than qq()
141 # or q()] to get the best explosion.
142 ["!Feed malformed utf8 into perl.", <<"BANG",
143 use utf8; %a = ("\xE1\xA0"=>"sterling");
144 print 'start'; printf '%x,', ord \$_ foreach keys %a; print "end\n";
146 qr/^Malformed UTF-8 character \(2 bytes, need 3\).*start\d+,end$/s
149 my ($why, $prog, $expect) = @$_;
150 open P, ">$progfile" or die "Can't open '$progfile': $!";
153 print '>' . join (',', map {ord} split //, $_) . '<'
158 close P or die "Can't close '$progfile': $!";
159 if ($why =~ s/^!//) {
160 print "# Possible delay...\n";
164 my $result = runperl ( stderr => 1, progfile => $progfile );
165 like ($result, $expect, $why);