Commit | Line | Data |
0effba8c |
1 | BEGIN { |
2 | if (ord("A") == 193) { |
3 | print "1..0 # encoding pragma does not support EBCDIC platforms\n"; |
0f963d18 |
4 | exit(0); |
0effba8c |
5 | } |
6 | } |
7 | |
553e1bcc |
8 | print "1..29\n"; |
799ef3cb |
9 | |
0a378802 |
10 | use encoding "latin1"; # ignored (overwritten by the next line) |
f14ed3c6 |
11 | use encoding "greek"; # iso 8859-7 (no "latin" alias, surprise...) |
0a378802 |
12 | |
0a378802 |
13 | # "greek" is "ISO 8859-7", and \xDF in ISO 8859-7 is |
f14ed3c6 |
14 | # \x{3AF} in Unicode (GREEK SMALL LETTER IOTA WITH TONOS), |
0a378802 |
15 | # instead of \xDF in Unicode (LATIN SMALL LETTER SHARP S) |
16 | |
9f4817db |
17 | $a = "\xDF"; |
18 | $b = "\x{100}"; |
19 | |
20 | print "not " unless ord($a) == 0x3af; |
0a378802 |
21 | print "ok 1\n"; |
22 | |
9f4817db |
23 | print "not " unless ord($b) == 0x100; |
0a378802 |
24 | print "ok 2\n"; |
25 | |
9f4817db |
26 | my $c; |
27 | |
28 | $c = $a . $b; |
29 | |
30 | print "not " unless ord($c) == 0x3af; |
0a378802 |
31 | print "ok 3\n"; |
32 | |
9f4817db |
33 | print "not " unless length($c) == 2; |
34 | print "ok 4\n"; |
35 | |
36 | print "not " unless ord(substr($c, 1, 1)) == 0x100; |
37 | print "ok 5\n"; |
0a378802 |
38 | |
121910a4 |
39 | print "not " unless ord(chr(0xdf)) == 0x3af; # spooky |
40 | print "ok 6\n"; |
41 | |
42 | print "not " unless ord(pack("C", 0xdf)) == 0x3af; |
43 | print "ok 7\n"; |
44 | |
45 | # we didn't break pack/unpack, I hope |
46 | |
47 | print "not " unless unpack("C", pack("C", 0xdf)) == 0xdf; |
48 | print "ok 8\n"; |
49 | |
50 | # the first octet of UTF-8 encoded 0x3af |
51 | print "not " unless unpack("C", chr(0xdf)) == 0xce; |
52 | print "ok 9\n"; |
bfa383d6 |
53 | |
3de8ed06 |
54 | print "not " unless unpack("U", pack("U", 0xdf)) == 0xdf; |
55 | print "ok 10\n"; |
56 | |
57 | print "not " unless unpack("U", chr(0xdf)) == 0x3af; |
58 | print "ok 11\n"; |
59 | |
bfa383d6 |
60 | # charnames must still work |
61 | use charnames ':full'; |
62 | print "not " unless ord("\N{LATIN SMALL LETTER SHARP S}") == 0xdf; |
3de8ed06 |
63 | print "ok 12\n"; |
64 | |
65 | # combine |
66 | |
67 | $c = "\xDF\N{LATIN SMALL LETTER SHARP S}" . chr(0xdf); |
68 | |
69 | print "not " unless ord($c) == 0x3af; |
70 | print "ok 13\n"; |
71 | |
72 | print "not " unless ord(substr($c, 1, 1)) == 0xdf; |
73 | print "ok 14\n"; |
74 | |
75 | print "not " unless ord(substr($c, 2, 1)) == 0x3af; |
76 | print "ok 15\n"; |
bfa383d6 |
77 | |
a72c7584 |
78 | # regex literals |
79 | |
80 | print "not " unless "\xDF" =~ /\x{3AF}/; |
81 | print "ok 16\n"; |
82 | |
83 | print "not " unless "\x{3AF}" =~ /\xDF/; |
84 | print "ok 17\n"; |
85 | |
86 | print "not " unless "\xDF" =~ /\xDF/; |
87 | print "ok 18\n"; |
88 | |
89 | print "not " unless "\x{3AF}" =~ /\x{3AF}/; |
90 | print "ok 19\n"; |
91 | |
799ef3cb |
92 | # eq, cmp |
93 | |
553e1bcc |
94 | my ($byte,$bytes,$U,$Ub,$g1,$g2,$l) = ( |
95 | pack("C*", 0xDF ), # byte |
96 | pack("C*", 0xDF, 0x20), # ($bytes2 cmp $U) > 0 |
97 | pack("U*", 0x3AF), # $U eq $byte |
98 | pack("U*", 0xDF ), # $Ub would eq $bytev w/o use encoding |
99 | pack("U*", 0x3B1), # ($g1 cmp $byte) > 0; === chr(0xe1) |
100 | pack("U*", 0x3AF, 0x20), # ($g2 cmp $byte) > 0; |
101 | pack("U*", 0x3AB), # ($l cmp $byte) < 0; === chr(0xdb) |
102 | ); |
103 | |
104 | # all the tests in this section that compare a byte encoded string |
105 | # ato UTF-8 encoded are run in all possible vairants |
106 | # all of the eq, ne, cmp operations tested, |
107 | # $v z $u tested as well as $u z $v |
108 | |
109 | sub alleq($$){ |
110 | my ($a,$b) = (shift, shift); |
111 | $a eq $b && $b eq $a && |
112 | !( $a ne $b ) && !( $b ne $a ) && |
113 | ( $a cmp $b ) == 0 && ( $b cmp $a ) == 0; |
114 | } |
115 | |
116 | sub anyeq($$){ |
117 | my ($a,$b) = (shift, shift); |
118 | $a eq $b || $b eq $a || |
119 | !( $a ne $b ) || !( $b ne $a ) || |
120 | ( $a cmp $b ) == 0 || ( $b cmp $a ) == 0; |
121 | } |
122 | |
123 | sub allgt($$){ |
124 | my ($a,$b) = (shift, shift); |
125 | ( $a cmp $b ) == 1 && ( $b cmp $a ) == -1; |
126 | } |
127 | #match the correct UTF-8 string |
128 | print "not " unless alleq($byte, $U); |
129 | print "ok 20\n"; |
130 | |
131 | #do not match a wrong UTF-8 string |
132 | print "not " if anyeq($byte, $Ub); |
133 | print "ok 21\n"; |
134 | |
135 | #string ordering |
136 | print "not " unless allgt ( $g1, $byte ) && |
137 | allgt ( $g2, $byte ) && |
138 | allgt ( $byte, $l ) && |
139 | allgt ( $bytes, $U ); |
140 | print "ok 22\n"; |
141 | |
142 | # upgrade, downgrade |
143 | |
144 | my ($u,$v,$v2); |
145 | $u = $v = $v2 = pack("C*", 0xDF); |
146 | utf8::upgrade($v); #explicit upgrade |
147 | $v2 = substr( $v2."\x{410}", 0, -1); #implicit upgrade |
148 | |
149 | # implicit upgrade === explicit upgrade |
150 | print "not " if do{{use bytes; $v ne $v2}} || $v ne $v2; |
151 | print "ok 23\n"; |
152 | |
153 | # utf8::upgrade is transparent and does not break equality |
154 | print "not " unless alleq( $u, $v ); |
155 | print "ok 24\n"; |
156 | |
157 | $u = $v = pack("C*", 0xDF); |
158 | utf8::upgrade($v); |
159 | #test for a roundtrip, we should get back from where we left |
160 | eval {utf8::downgrade( $v )}; |
161 | print "not " if $@ !~ /^Wide / || do{{use bytes; $u eq $v}} || $u ne $v; |
162 | print "ok 25\n"; |
163 | |
164 | # some more eq, cmp |
165 | |
799ef3cb |
166 | my $byte=pack("C*", 0xDF); |
167 | |
168 | print "not " unless pack("U*", 0x3AF) eq $byte; |
553e1bcc |
169 | print "ok 26\n"; |
799ef3cb |
170 | |
171 | print "not " if chr(0xDF) cmp $byte; |
553e1bcc |
172 | print "ok 27\n"; |
799ef3cb |
173 | |
174 | print "not " unless ((pack("U*", 0x3B0) cmp $byte) == 1) && |
175 | ((pack("U*", 0x3AE) cmp $byte) == -1) && |
176 | ((pack("U*", 0x3AF, 0x20) cmp $byte) == 1) && |
177 | ((pack("U*", 0x3AF) cmp pack("C*",0xDF,0x20))==-1); |
553e1bcc |
178 | print "ok 28\n"; |
799ef3cb |
179 | |
180 | # Used to core dump in 5.7.3 |
553e1bcc |
181 | print ord undef == 0 ? "ok 29\n" : "not ok 29\n"; |