3 # This file is intentionally encoded in latin-1.
8 @INC = () unless is_miniperl();
9 unshift @INC, '../lib';
10 require Config; import Config;
11 require './charset_tools.pl';
12 require './loc_tools.pl'; # Contains find_utf8_ctype_locale()
17 plan tests => 134 + 4 * 256;
19 is(lc(undef), "", "lc(undef) is ''");
20 is(lcfirst(undef), "", "lcfirst(undef) is ''");
21 is(uc(undef), "", "uc(undef) is ''");
22 is(ucfirst(undef), "", "ucfirst(undef) is ''");
26 is(CORE::fc(undef), "", "fc(undef) is ''");
27 is(CORE::fc(''), "", "fc('') is ''");
31 like($@, qr/Undefined subroutine &main::fc/, "fc() throws an exception,");
37 ok(!$@, "...but works after requesting the feature");
44 is("\Q$a\E." , "HELLO\\.\\*\\ world.", '\Q\E HELLO.* world');
45 is("\u$a" , "HELLO\.\* world", '\u');
46 is("\l$a" , "hELLO\.\* world", '\l');
47 is("\U$a" , "HELLO\.\* WORLD", '\U');
48 is("\L$a" , "hello\.\* world", '\L');
49 is("\F$a" , "hello\.\* world", '\F');
51 is(quotemeta($a) , "HELLO\\.\\*\\ world", 'quotemeta');
52 is(ucfirst($a) , "HELLO\.\* world", 'ucfirst');
53 is(lcfirst($a) , "hELLO\.\* world", 'lcfirst');
54 is(uc($a) , "HELLO\.\* WORLD", 'uc');
55 is(lc($a) , "hello\.\* world", 'lc');
56 is(fc($a) , "hello\.\* world", 'fc');
58 is("\Q$b\E." , "hello\\.\\*\\ WORLD.", '\Q\E hello.* WORLD');
59 is("\u$b" , "Hello\.\* WORLD", '\u');
60 is("\l$b" , "hello\.\* WORLD", '\l');
61 is("\U$b" , "HELLO\.\* WORLD", '\U');
62 is("\L$b" , "hello\.\* world", '\L');
63 is("\F$b" , "hello\.\* world", '\F');
65 is(quotemeta($b) , "hello\\.\\*\\ WORLD", 'quotemeta');
66 is(ucfirst($b) , "Hello\.\* WORLD", 'ucfirst');
67 is(lcfirst($b) , "hello\.\* WORLD", 'lcfirst');
68 is(uc($b) , "HELLO\.\* WORLD", 'uc');
69 is(lc($b) , "hello\.\* world", 'lc');
70 is(fc($b) , "hello\.\* world", 'fc');
72 # \x{100} is LATIN CAPITAL LETTER A WITH MACRON; its bijective lowercase is
73 # \x{101}, LATIN SMALL LETTER A WITH MACRON.
74 # Which is also its foldcase.
76 $a = "\x{100}\x{101}Aa";
77 $b = "\x{101}\x{100}aA";
79 is("\Q$a\E." , "\x{100}\x{101}Aa.", '\Q\E \x{100}\x{101}Aa');
80 is("\u$a" , "\x{100}\x{101}Aa", '\u');
81 is("\l$a" , "\x{101}\x{101}Aa", '\l');
82 is("\U$a" , "\x{100}\x{100}AA", '\U');
83 is("\L$a" , "\x{101}\x{101}aa", '\L');
84 is("\F$a" , "\x{101}\x{101}aa", '\F');
86 is(quotemeta($a) , "\x{100}\x{101}Aa", 'quotemeta');
87 is(ucfirst($a) , "\x{100}\x{101}Aa", 'ucfirst');
88 is(lcfirst($a) , "\x{101}\x{101}Aa", 'lcfirst');
89 is(uc($a) , "\x{100}\x{100}AA", 'uc');
90 is(lc($a) , "\x{101}\x{101}aa", 'lc');
91 is(fc($a) , "\x{101}\x{101}aa", 'fc');
93 is("\Q$b\E." , "\x{101}\x{100}aA.", '\Q\E \x{101}\x{100}aA');
94 is("\u$b" , "\x{100}\x{100}aA", '\u');
95 is("\l$b" , "\x{101}\x{100}aA", '\l');
96 is("\U$b" , "\x{100}\x{100}AA", '\U');
97 is("\L$b" , "\x{101}\x{101}aa", '\L');
98 is("\F$b" , "\x{101}\x{101}aa", '\F');
100 is(quotemeta($b) , "\x{101}\x{100}aA", 'quotemeta');
101 is(ucfirst($b) , "\x{100}\x{100}aA", 'ucfirst');
102 is(lcfirst($b) , "\x{101}\x{100}aA", 'lcfirst');
103 is(uc($b) , "\x{100}\x{100}AA", 'uc');
104 is(lc($b) , "\x{101}\x{101}aa", 'lc');
105 is(fc($b) , "\x{101}\x{101}aa", 'fc');
107 # \x{DF} is LATIN SMALL LETTER SHARP S, its uppercase is SS or \x{53}\x{53};
108 # \x{149} is LATIN SMALL LETTER N PRECEDED BY APOSTROPHE, its uppercase is
109 # \x{2BC}\x{E4} or MODIFIER LETTER APOSTROPHE and N.
111 is(latin1_to_native("\U\x{DF}aB\x{149}cD"), latin1_to_native("SSAB\x{2BC}NCD"),
112 "multicharacter uppercase");
114 # The \x{DF} is its own lowercase, ditto for \x{149}.
115 # There are no single character -> multiple characters lowercase mappings.
117 is(latin1_to_native("\L\x{DF}aB\x{149}cD"), latin1_to_native("\x{DF}ab\x{149}cd"),
118 "multicharacter lowercase");
120 # \x{DF} is LATIN SMALL LETTER SHARP S, its foldcase is ss or \x{73}\x{73};
121 # \x{149} is LATIN SMALL LETTER N PRECEDED BY APOSTROPHE, its foldcase is
122 # \x{2BC}\x{6E} or MODIFIER LETTER APOSTROPHE and n.
123 # Note that is this further tested in t/uni/fold.t
125 is(latin1_to_native("\F\x{DF}aB\x{149}cD"), latin1_to_native("ssab\x{2BC}ncd"),
126 "multicharacter foldcase");
129 # titlecase is used for \u / ucfirst.
131 # \x{587} is ARMENIAN SMALL LIGATURE ECH YIWN and its titlecase is
132 # \x{535}\x{582} ARMENIAN CAPITAL LETTER ECH + ARMENIAN SMALL LETTER YIWN
133 # while its lowercase is
135 # and its uppercase is
136 # \x{535}\x{552} ARMENIAN CAPITAL LETTER ECH + ARMENIAN CAPITAL LETTER YIWN
137 # The foldcase is \x{565}\x{582} ARMENIAN SMALL LETTER ECH + ARMENIAN SMALL LETTER YIWN
141 is("\L\x{587}" , "\x{587}", "ligature lowercase");
142 is("\u\x{587}" , "\x{535}\x{582}", "ligature titlecase");
143 is("\U\x{587}" , "\x{535}\x{552}", "ligature uppercase");
144 is("\F\x{587}" , "\x{565}\x{582}", "ligature foldcase");
146 # mktables had problems where many-to-one case mappings didn't work right.
147 # The lib/uni/fold.t should give the fourth folding, "casefolding", a good
149 # \x{01C4} is LATIN CAPITAL LETTER DZ WITH CARON
150 # \x{01C5} is LATIN CAPITAL LETTER D WITH SMALL LETTER Z WITH CARON
151 # \x{01C6} is LATIN SMALL LETTER DZ WITH CARON
152 # \x{03A3} is GREEK CAPITAL LETTER SIGMA
153 # \x{03C2} is GREEK SMALL LETTER FINAL SIGMA
154 # \x{03C3} is GREEK SMALL LETTER SIGMA
156 is(lc("\x{1C4}") , "\x{1C6}", "U+01C4 lc is U+01C6");
157 is(lc("\x{1C5}") , "\x{1C6}", "U+01C5 lc is U+01C6, too");
159 is(ucfirst("\x{3C2}") , "\x{3A3}", "U+03C2 ucfirst is U+03A3");
160 is(ucfirst("\x{3C3}") , "\x{3A3}", "U+03C3 ucfirst is U+03A3, too");
162 is(uc("\x{1C5}") , "\x{1C4}", "U+01C5 uc is U+01C4");
163 is(uc("\x{1C6}") , "\x{1C4}", "U+01C6 uc is U+01C4, too");
165 # #18107: A host of bugs involving [ul]c{,first}. AMS 20021106
166 $a = "\x{3c3}foo.bar"; # \x{3c3} == GREEK SMALL LETTER SIGMA.
167 $b = "\x{3a3}FOO.BAR"; # \x{3a3} == GREEK CAPITAL LETTER SIGMA.
169 ($c = $b) =~ s/(\w+)/lc($1)/ge;
170 is($c , $a, "Using s///e to change case.");
172 ($c = $a) =~ s/(\p{IsWord}+)/uc($1)/ge;
173 is($c , $b, "Using s///e to change case.");
175 ($c = $a) =~ s/(\p{IsWord}+)/fc($1)/ge;
176 is($c , $a, "Using s///e to foldcase.");
178 ($c = $b) =~ s/(\p{IsWord}+)/lcfirst($1)/ge;
179 is($c , "\x{3c3}FOO.bAR", "Using s///e to change case.");
181 ($c = $a) =~ s/(\p{IsWord}+)/ucfirst($1)/ge;
182 is($c , "\x{3a3}foo.Bar", "Using s///e to change case.");
184 # #18931: perl5.8.0 bug in \U..\E processing
185 # Test case from Nicholas Clark.
191 is(uc($1), "ABCDEFGH", "[perl #18931]");
198 $a =~ s/^(\s*)(\w*)/$1\u$2/;
199 is($a, v10, "[perl #18857]");
204 # [perl #38619] Bug in lc and uc (interaction between UTF-8, substr, and lc/uc)
206 for ("a\x{100}", "xyz\x{100}") {
207 is(substr(uc($_), 0), uc($_), "[perl #38619] uc");
209 for ("A\x{100}", "XYZ\x{100}") {
210 is(substr(lc($_), 0), lc($_), "[perl #38619] lc");
212 for ("a\x{100}", "ßyz\x{100}") { # ß to Ss (different length)
213 is(substr(ucfirst($_), 0), ucfirst($_), "[perl #38619] ucfirst");
216 #fc() didn't exist back then, but coverage is coverage.
217 for ("a\x{100}", "ßyz\x{100}", "xyz\x{100}", "XYZ\x{100}") { # ß to Ss (different length)
218 is(substr(fc($_), 0), fc($_), "[perl #38619] fc");
221 # Related to [perl #38619]
222 # the original report concerns PERL_MAGIC_utf8.
223 # these cases concern PERL_MAGIC_regex_global.
225 for (map { $_ } "a\x{100}", "abc\x{100}", "\x{100}") {
226 chop; # get ("a", "abc", "") in utf8
227 my $return = uc($_) =~ /\G(.?)/g;
228 my $result = $return ? $1 : "not";
229 my $expect = (uc($_) =~ /(.?)/g)[0];
230 is($return, 1, "[perl #38619]");
231 is($result, $expect, "[perl #38619]");
234 for (map { $_ } "A\x{100}", "ABC\x{100}", "\x{100}") {
235 chop; # get ("A", "ABC", "") in utf8
236 my $return = lc($_) =~ /\G(.?)/g;
237 my $result = $return ? $1 : "not";
238 my $expect = (lc($_) =~ /(.?)/g)[0];
239 is($return, 1, "[perl #38619]");
240 is($result, $expect, "[perl #38619]");
243 for (map { $_ } "A\x{100}", "ABC\x{100}", "\x{100}") {
244 chop; # get ("A", "ABC", "") in utf8
245 my $return = fc($_) =~ /\G(.?)/g;
246 my $result = $return ? $1 : "not";
247 my $expect = (fc($_) =~ /(.?)/g)[0];
248 is($return, 1, "[perl #38619]");
249 is($result, $expect, "[perl #38619]");
252 for (1, 4, 9, 16, 25) {
253 is(uc "\x{03B0}" x $_, "\x{3a5}\x{308}\x{301}" x $_,
254 'uc U+03B0 grows threefold');
256 is(lc "\x{0130}" x $_, "i\x{307}" x $_, 'lc U+0130 grows');
258 is(fc "\x{03B0}" x $_, "\x{3C5}\x{308}\x{301}" x $_,
259 'fc U+03B0 grows threefold');
266 is($_, "HellO", '[perl #43207] lc($_) modifying $_');
270 is($_, "HellO", '[perl #43207] fc($_) modifying $_');
274 is($_, "HellO", '[perl #43207] uc($_) modifying $_');
278 is($_, "HellO", '[perl #43207] ucfirst($_) modifying $_');
282 is($_, "HellO", '[perl #43207] lcfirst($_) modifying $_');
285 # new in Unicode 5.1.0
286 is(lc("\x{1E9E}"), "\x{df}", "lc(LATIN CAPITAL LETTER SHARP S)");
289 use feature 'unicode_strings';
291 is(lc("\xc0"), "\xc0", "lc of above-ASCII Latin1 is itself under use bytes");
292 is(lcfirst("\xc0"), "\xc0", "lcfirst of above-ASCII Latin1 is itself under use bytes");
293 is(uc("\xe0"), "\xe0", "uc of above-ASCII Latin1 is itself under use bytes");
294 is(ucfirst("\xe0"), "\xe0", "ucfirst of above-ASCII Latin1 is itself under use bytes");
297 # Brought up in ticket #117855: Constant folding applied to uc() should use
298 # the right set of hints.
299 fresh_perl_like(<<'constantfolding', qr/^(\d+),\1\z/, {},
303 use feature 'unicode_strings';
304 print ord uc($char), ",",
305 ord eval "$function('$char')", "\n";
308 'folded uc() in string eval uses the right hints');
310 # In-place lc/uc should not corrupt string buffers when given a non-utf8-
311 # flagged thingy that stringifies to utf8
312 $h{k} = bless[], "\x{3b0}\x{3b0}\x{3b0}bcde"; # U+03B0 grows with uc()
313 # using delete marks it as TEMP, so uc-in-place is permitted
314 like uc delete $h{k}, qr "^(?:\x{3a5}\x{308}\x{301}){3}BCDE=ARRAY\(.*\)",
315 'uc(TEMP ref) does not produce a corrupt string';
316 $h{k} = bless[], "\x{130}bcde"; # U+0130 grows with lc()
317 # using delete marks it as TEMP, so uc-in-place is permitted
318 like lc delete $h{k}, qr "^i\x{307}bcde=array\(.*\)",
319 'lc(TEMP ref) does not produce a corrupt string';
322 my $utf8_locale = find_utf8_ctype_locale();
325 skip 'Can\'t find a UTF-8 locale', 4*256 unless defined $utf8_locale;
327 eval { require POSIX; import POSIX 'locale_h'; };
328 unless (defined &POSIX::LC_CTYPE) {
329 skip "no POSIX (maybe no Fcntl, or no dynamic loading)", 4*256;
332 use feature qw( unicode_strings );
341 # Get all the values outside of 'locale'
342 for my $i (0 .. 255) {
343 push @unicode_lc, lc(chr $i);
344 push @unicode_uc, uc(chr $i);
345 push @unicode_lcfirst, lcfirst(chr $i);
346 push @unicode_ucfirst, ucfirst(chr $i);
350 setlocale(LC_CTYPE, $utf8_locale);
352 for my $i (0 .. 255) {
353 is(lc(chr $i), $unicode_lc[$i], "In a UTF-8 locale, lc(chr $i) is the same as official Unicode");
354 is(uc(chr $i), $unicode_uc[$i], "In a UTF-8 locale, uc(chr $i) is the same as official Unicode");
355 is(lcfirst(chr $i), $unicode_lcfirst[$i], "In a UTF-8 locale, lcfirst(chr $i) is the same as official Unicode");
356 is(ucfirst(chr $i), $unicode_ucfirst[$i], "In a UTF-8 locale, ucfirst(chr $i) is the same as official Unicode");