Commit | Line | Data |
---|---|---|
19692e8d NC |
1 | #!./perl -w |
2 | ||
cb0a5b5c NIS |
3 | BEGIN { |
4 | chdir 't' if -d 't'; | |
5 | @INC = '../lib'; | |
6 | require './test.pl'; | |
7 | ||
eaf7a4d2 | 8 | plan(tests => 99); |
cb0a5b5c NIS |
9 | } |
10 | ||
19692e8d NC |
11 | use strict; |
12 | ||
cb0a5b5c NIS |
13 | # Two hashes one will all keys 8-bit possible (initially), other |
14 | # with a utf8 requiring key from the outset. | |
15 | ||
16 | my %hash8 = ( "\xff" => 0xff, | |
17 | "\x7f" => 0x7f, | |
18 | ); | |
19 | my %hashu = ( "\xff" => 0xff, | |
20 | "\x7f" => 0x7f, | |
21 | "\x{1ff}" => 0x1ff, | |
22 | ); | |
23 | ||
93f09d7b | 24 | # Check that we can find the 8-bit things by various literals |
cb0a5b5c NIS |
25 | is($hash8{"\x{00ff}"},0xFF); |
26 | is($hash8{"\x{007f}"},0x7F); | |
27 | is($hash8{"\xff"},0xFF); | |
28 | is($hash8{"\x7f"},0x7F); | |
29 | is($hashu{"\x{00ff}"},0xFF); | |
30 | is($hashu{"\x{007f}"},0x7F); | |
31 | is($hashu{"\xff"},0xFF); | |
32 | is($hashu{"\x7f"},0x7F); | |
33 | ||
34 | # Now try same thing with variables forced into various forms. | |
5fec3b1d | 35 | foreach ("\x7f","\xff") |
cb0a5b5c | 36 | { |
5fec3b1d | 37 | my $a = $_; # Force a copy |
cb0a5b5c NIS |
38 | utf8::upgrade($a); |
39 | is($hash8{$a},ord($a)); | |
40 | is($hashu{$a},ord($a)); | |
41 | utf8::downgrade($a); | |
42 | is($hash8{$a},ord($a)); | |
43 | is($hashu{$a},ord($a)); | |
44 | my $b = $a.chr(100); | |
45 | chop($b); | |
46 | is($hash8{$b},ord($b)); | |
47 | is($hashu{$b},ord($b)); | |
48 | } | |
49 | ||
50 | # Check we have not got an spurious extra keys | |
20b5b8d0 JH |
51 | is(join('',sort { ord $a <=> ord $b } keys %hash8),"\x7f\xff"); |
52 | is(join('',sort { ord $a <=> ord $b } keys %hashu),"\x7f\xff\x{1ff}"); | |
cb0a5b5c NIS |
53 | |
54 | # Now add a utf8 key to the 8-bit hash | |
55 | $hash8{chr(0x1ff)} = 0x1ff; | |
56 | ||
57 | # Check we have not got an spurious extra keys | |
20b5b8d0 | 58 | is(join('',sort { ord $a <=> ord $b } keys %hash8),"\x7f\xff\x{1ff}"); |
cb0a5b5c | 59 | |
5fec3b1d | 60 | foreach ("\x7f","\xff","\x{1ff}") |
cb0a5b5c | 61 | { |
5fec3b1d | 62 | my $a = $_; |
cb0a5b5c NIS |
63 | utf8::upgrade($a); |
64 | is($hash8{$a},ord($a)); | |
65 | my $b = $a.chr(100); | |
66 | chop($b); | |
67 | is($hash8{$b},ord($b)); | |
68 | } | |
69 | ||
70 | # and remove utf8 from the other hash | |
71 | is(delete $hashu{chr(0x1ff)},0x1ff); | |
72 | is(join('',sort keys %hashu),"\x7f\xff"); | |
73 | ||
5fec3b1d | 74 | foreach ("\x7f","\xff") |
cb0a5b5c | 75 | { |
5fec3b1d | 76 | my $a = $_; |
cb0a5b5c NIS |
77 | utf8::upgrade($a); |
78 | is($hashu{$a},ord($a)); | |
79 | utf8::downgrade($a); | |
80 | is($hashu{$a},ord($a)); | |
81 | my $b = $a.chr(100); | |
82 | chop($b); | |
83 | is($hashu{$b},ord($b)); | |
84 | } | |
85 | ||
86 | ||
4c79aee6 NC |
87 | |
88 | { | |
19692e8d NC |
89 | print "# Unicode hash keys and \\w\n"; |
90 | # This is not really a regex test but regexes bring | |
91 | # out the issue nicely. | |
92 | use strict; | |
93 | my $u3 = "f\x{df}\x{100}"; | |
94 | my $u2 = substr($u3,0,2); | |
95 | my $u1 = substr($u2,0,1); | |
96 | my $u0 = chr (0xdf)x4; # Make this 4 chars so that all lengths are distinct. | |
97 | ||
98 | my @u = ($u0, $u1, $u2, $u3); | |
99 | ||
100 | while (@u) { | |
101 | my %u = (map {( $_, $_)} @u); | |
102 | my $keys = scalar @u; | |
103 | $keys .= ($keys == 1) ? " key" : " keys"; | |
4c79aee6 NC |
104 | |
105 | for (keys %u) { | |
19692e8d NC |
106 | my $l = 0 + /^\w+$/; |
107 | my $r = 0 + $u{$_} =~ /^\w+$/; | |
108 | is ($l, $r, "\\w on keys with $keys, key of length " . length $_); | |
109 | } | |
110 | ||
111 | my $more; | |
112 | do { | |
113 | $more = 0; | |
114 | # Want to do this direct, rather than copying to a temporary variable | |
115 | # The first time each will return key and value at the start of the hash. | |
116 | # each will return () after we've done the last pair. $more won't get | |
117 | # set then, and the do will exit. | |
118 | for (each %u) { | |
119 | $more = 1; | |
120 | my $l = 0 + /^\w+$/; | |
121 | my $r = 0 + $u{$_} =~ /^\w+$/; | |
122 | is ($l, $r, "\\w on each, with $keys, key of length " . length $_); | |
123 | } | |
124 | } while ($more); | |
4c79aee6 NC |
125 | |
126 | for (%u) { | |
19692e8d NC |
127 | my $l = 0 + /^\w+$/; |
128 | my $r = 0 + $u{$_} =~ /^\w+$/; | |
129 | is ($l, $r, "\\w on hash with $keys, key of length " . length $_); | |
130 | } | |
131 | pop @u; | |
132 | undef %u; | |
133 | } | |
134 | } | |
135 | ||
136 | { | |
137 | my $utf8_sz = my $bytes_sz = "\x{df}"; | |
138 | $utf8_sz .= chr 256; | |
139 | chop ($utf8_sz); | |
140 | ||
141 | my (%bytes_first, %utf8_first); | |
142 | ||
143 | $bytes_first{$bytes_sz} = $bytes_sz; | |
144 | ||
145 | for (keys %bytes_first) { | |
146 | my $l = 0 + /^\w+$/; | |
147 | my $r = 0 + $bytes_first{$_} =~ /^\w+$/; | |
148 | is ($l, $r, "\\w on each, bytes"); | |
149 | } | |
150 | ||
151 | $bytes_first{$utf8_sz} = $utf8_sz; | |
152 | ||
153 | for (keys %bytes_first) { | |
154 | my $l = 0 + /^\w+$/; | |
155 | my $r = 0 + $bytes_first{$_} =~ /^\w+$/; | |
156 | is ($l, $r, "\\w on each, bytes now utf8"); | |
157 | } | |
158 | ||
159 | $utf8_first{$utf8_sz} = $utf8_sz; | |
160 | ||
161 | for (keys %utf8_first) { | |
162 | my $l = 0 + /^\w+$/; | |
163 | my $r = 0 + $utf8_first{$_} =~ /^\w+$/; | |
164 | is ($l, $r, "\\w on each, utf8"); | |
165 | } | |
166 | ||
167 | $utf8_first{$bytes_sz} = $bytes_sz; | |
168 | ||
169 | for (keys %utf8_first) { | |
170 | my $l = 0 + /^\w+$/; | |
171 | my $r = 0 + $utf8_first{$_} =~ /^\w+$/; | |
172 | is ($l, $r, "\\w on each, utf8 now bytes"); | |
173 | } | |
174 | ||
4c79aee6 | 175 | } |
5464dbd2 RGS |
176 | |
177 | { | |
250d67eb JH |
178 | local $/; # Slurp. |
179 | my $utf8 = <DATA>; | |
180 | my $utfebcdic = <DATA>; | |
181 | if (ord('A') == 65) { | |
182 | eval $utf8; | |
183 | } elsif (ord('A') == 193) { | |
184 | eval $utfebcdic; | |
185 | } | |
186 | } | |
187 | __END__ | |
188 | { | |
5464dbd2 RGS |
189 | # See if utf8 barewords work [perl #22969] |
190 | use utf8; | |
191 | my %hash = (тест => 123); | |
192 | is($hash{тест}, $hash{'тест'}); | |
193 | is($hash{тест}, 123); | |
194 | is($hash{'тест'}, 123); | |
195 | %hash = (тест => 123); | |
196 | is($hash{тест}, $hash{'тест'}); | |
197 | is($hash{тест}, 123); | |
198 | is($hash{'тест'}, 123); | |
eaf7a4d2 CS |
199 | |
200 | # See if plain ASCII strings quoted with '=>' erroneously get utf8 flag [perl #68812] | |
201 | my %foo = (a => 'b', 'c' => 'd'); | |
202 | for my $key (keys %foo) { | |
203 | ok !utf8::is_utf8($key), "'$key' shouldn't have utf8 flag"; | |
204 | } | |
5464dbd2 | 205 | } |
250d67eb JH |
206 | __END__ |
207 | { | |
208 | # See if utf8 barewords work [perl #22969] | |
209 | use utf8; # UTF-EBCDIC, really. | |
210 | my %hash = (½ää½âÀ½äâ½ää => 123); | |
211 | is($hash{½ää½âÀ½äâ½ää}, $hash{'½ää½âÀ½äâ½ää'}); | |
212 | is($hash{½ää½âÀ½äâ½ää}, 123); | |
213 | is($hash{'½ää½âÀ½äâ½ää'}, 123); | |
214 | %hash = (½ää½âÀ½äâ½ää => 123); | |
215 | is($hash{½ää½âÀ½äâ½ää}, $hash{'½ää½âÀ½äâ½ää'}); | |
216 | is($hash{½ää½âÀ½äâ½ää}, 123); | |
217 | is($hash{'½ää½âÀ½äâ½ää'}, 123); | |
eaf7a4d2 CS |
218 | |
219 | # See if plain ASCII strings quoted with '=>' erroneously get utf8 flag [perl #68812] | |
220 | my %foo = (a => 'b', 'c' => 'd'); | |
221 | for my $key (keys %foo) { | |
222 | ok !utf8::is_utf8($key), "'$key' shouldn't have utf8 flag"; | |
223 | } | |
250d67eb | 224 | } |