1#!./perl 2 3# This file is intentionally encoded in latin-1. 4 5BEGIN { 6 chdir 't'; 7 @INC = '../lib'; 8 require Config; import Config; 9 require './test.pl'; 10 require './loc_tools.pl'; # Contains find_utf8_ctype_locale() 11} 12 13use feature qw( fc ); 14 15plan tests => 134 + 4 * 256; 16 17is(lc(undef), "", "lc(undef) is ''"); 18is(lcfirst(undef), "", "lcfirst(undef) is ''"); 19is(uc(undef), "", "uc(undef) is ''"); 20is(ucfirst(undef), "", "ucfirst(undef) is ''"); 21 22{ 23 no feature 'fc'; 24 is(CORE::fc(undef), "", "fc(undef) is ''"); 25 is(CORE::fc(''), "", "fc('') is ''"); 26 27 local $@; 28 eval { fc("eeyup") }; 29 like($@, qr/Undefined subroutine &main::fc/, "fc() throws an exception,"); 30 31 { 32 use feature 'fc'; 33 local $@; 34 eval { fc("eeyup") }; 35 ok(!$@, "...but works after requesting the feature"); 36 } 37} 38 39$a = "HELLO.* world"; 40$b = "hello.* WORLD"; 41 42is("\Q$a\E." , "HELLO\\.\\*\\ world.", '\Q\E HELLO.* world'); 43is("\u$a" , "HELLO\.\* world", '\u'); 44is("\l$a" , "hELLO\.\* world", '\l'); 45is("\U$a" , "HELLO\.\* WORLD", '\U'); 46is("\L$a" , "hello\.\* world", '\L'); 47is("\F$a" , "hello\.\* world", '\F'); 48 49is(quotemeta($a) , "HELLO\\.\\*\\ world", 'quotemeta'); 50is(ucfirst($a) , "HELLO\.\* world", 'ucfirst'); 51is(lcfirst($a) , "hELLO\.\* world", 'lcfirst'); 52is(uc($a) , "HELLO\.\* WORLD", 'uc'); 53is(lc($a) , "hello\.\* world", 'lc'); 54is(fc($a) , "hello\.\* world", 'fc'); 55 56is("\Q$b\E." , "hello\\.\\*\\ WORLD.", '\Q\E hello.* WORLD'); 57is("\u$b" , "Hello\.\* WORLD", '\u'); 58is("\l$b" , "hello\.\* WORLD", '\l'); 59is("\U$b" , "HELLO\.\* WORLD", '\U'); 60is("\L$b" , "hello\.\* world", '\L'); 61is("\F$b" , "hello\.\* world", '\F'); 62 63is(quotemeta($b) , "hello\\.\\*\\ WORLD", 'quotemeta'); 64is(ucfirst($b) , "Hello\.\* WORLD", 'ucfirst'); 65is(lcfirst($b) , "hello\.\* WORLD", 'lcfirst'); 66is(uc($b) , "HELLO\.\* WORLD", 'uc'); 67is(lc($b) , "hello\.\* world", 'lc'); 68is(fc($b) , "hello\.\* world", 'fc'); 69 70# \x{100} is LATIN CAPITAL LETTER A WITH MACRON; its bijective lowercase is 71# \x{101}, LATIN SMALL LETTER A WITH MACRON. 72# Which is also its foldcase. 73 74$a = "\x{100}\x{101}Aa"; 75$b = "\x{101}\x{100}aA"; 76 77is("\Q$a\E." , "\x{100}\x{101}Aa.", '\Q\E \x{100}\x{101}Aa'); 78is("\u$a" , "\x{100}\x{101}Aa", '\u'); 79is("\l$a" , "\x{101}\x{101}Aa", '\l'); 80is("\U$a" , "\x{100}\x{100}AA", '\U'); 81is("\L$a" , "\x{101}\x{101}aa", '\L'); 82is("\F$a" , "\x{101}\x{101}aa", '\F'); 83 84is(quotemeta($a) , "\x{100}\x{101}Aa", 'quotemeta'); 85is(ucfirst($a) , "\x{100}\x{101}Aa", 'ucfirst'); 86is(lcfirst($a) , "\x{101}\x{101}Aa", 'lcfirst'); 87is(uc($a) , "\x{100}\x{100}AA", 'uc'); 88is(lc($a) , "\x{101}\x{101}aa", 'lc'); 89is(fc($a) , "\x{101}\x{101}aa", 'fc'); 90 91is("\Q$b\E." , "\x{101}\x{100}aA.", '\Q\E \x{101}\x{100}aA'); 92is("\u$b" , "\x{100}\x{100}aA", '\u'); 93is("\l$b" , "\x{101}\x{100}aA", '\l'); 94is("\U$b" , "\x{100}\x{100}AA", '\U'); 95is("\L$b" , "\x{101}\x{101}aa", '\L'); 96is("\F$b" , "\x{101}\x{101}aa", '\F'); 97 98is(quotemeta($b) , "\x{101}\x{100}aA", 'quotemeta'); 99is(ucfirst($b) , "\x{100}\x{100}aA", 'ucfirst'); 100is(lcfirst($b) , "\x{101}\x{100}aA", 'lcfirst'); 101is(uc($b) , "\x{100}\x{100}AA", 'uc'); 102is(lc($b) , "\x{101}\x{101}aa", 'lc'); 103is(fc($b) , "\x{101}\x{101}aa", 'fc'); 104 105# \x{DF} is LATIN SMALL LETTER SHARP S, its uppercase is SS or \x{53}\x{53}; 106# \x{149} is LATIN SMALL LETTER N PRECEDED BY APOSTROPHE, its uppercase is 107# \x{2BC}\x{E4} or MODIFIER LETTER APOSTROPHE and N. 108 109is(latin1_to_native("\U\x{DF}aB\x{149}cD"), latin1_to_native("SSAB\x{2BC}NCD"), 110 "multicharacter uppercase"); 111 112# The \x{DF} is its own lowercase, ditto for \x{149}. 113# There are no single character -> multiple characters lowercase mappings. 114 115is(latin1_to_native("\L\x{DF}aB\x{149}cD"), latin1_to_native("\x{DF}ab\x{149}cd"), 116 "multicharacter lowercase"); 117 118# \x{DF} is LATIN SMALL LETTER SHARP S, its foldcase is ss or \x{73}\x{73}; 119# \x{149} is LATIN SMALL LETTER N PRECEDED BY APOSTROPHE, its foldcase is 120# \x{2BC}\x{6E} or MODIFIER LETTER APOSTROPHE and n. 121# Note that is this further tested in t/uni/fold.t 122 123is(latin1_to_native("\F\x{DF}aB\x{149}cD"), latin1_to_native("ssab\x{2BC}ncd"), 124 "multicharacter foldcase"); 125 126 127# titlecase is used for \u / ucfirst. 128 129# \x{587} is ARMENIAN SMALL LIGATURE ECH YIWN and its titlecase is 130# \x{535}\x{582} ARMENIAN CAPITAL LETTER ECH + ARMENIAN SMALL LETTER YIWN 131# while its lowercase is 132# \x{587} itself 133# and its uppercase is 134# \x{535}\x{552} ARMENIAN CAPITAL LETTER ECH + ARMENIAN CAPITAL LETTER YIWN 135# The foldcase is \x{565}\x{582} ARMENIAN SMALL LETTER ECH + ARMENIAN SMALL LETTER YIWN 136 137$a = "\x{587}"; 138 139is("\L\x{587}" , "\x{587}", "ligature lowercase"); 140is("\u\x{587}" , "\x{535}\x{582}", "ligature titlecase"); 141is("\U\x{587}" , "\x{535}\x{552}", "ligature uppercase"); 142is("\F\x{587}" , "\x{565}\x{582}", "ligature foldcase"); 143 144# mktables had problems where many-to-one case mappings didn't work right. 145# The lib/uni/fold.t should give the fourth folding, "casefolding", a good 146# workout. 147# \x{01C4} is LATIN CAPITAL LETTER DZ WITH CARON 148# \x{01C5} is LATIN CAPITAL LETTER D WITH SMALL LETTER Z WITH CARON 149# \x{01C6} is LATIN SMALL LETTER DZ WITH CARON 150# \x{03A3} is GREEK CAPITAL LETTER SIGMA 151# \x{03C2} is GREEK SMALL LETTER FINAL SIGMA 152# \x{03C3} is GREEK SMALL LETTER SIGMA 153 154is(lc("\x{1C4}") , "\x{1C6}", "U+01C4 lc is U+01C6"); 155is(lc("\x{1C5}") , "\x{1C6}", "U+01C5 lc is U+01C6, too"); 156 157is(ucfirst("\x{3C2}") , "\x{3A3}", "U+03C2 ucfirst is U+03A3"); 158is(ucfirst("\x{3C3}") , "\x{3A3}", "U+03C3 ucfirst is U+03A3, too"); 159 160is(uc("\x{1C5}") , "\x{1C4}", "U+01C5 uc is U+01C4"); 161is(uc("\x{1C6}") , "\x{1C4}", "U+01C6 uc is U+01C4, too"); 162 163# #18107: A host of bugs involving [ul]c{,first}. AMS 20021106 164$a = "\x{3c3}foo.bar"; # \x{3c3} == GREEK SMALL LETTER SIGMA. 165$b = "\x{3a3}FOO.BAR"; # \x{3a3} == GREEK CAPITAL LETTER SIGMA. 166 167($c = $b) =~ s/(\w+)/lc($1)/ge; 168is($c , $a, "Using s///e to change case."); 169 170($c = $a) =~ s/(\p{IsWord}+)/uc($1)/ge; 171is($c , $b, "Using s///e to change case."); 172 173($c = $a) =~ s/(\p{IsWord}+)/fc($1)/ge; 174is($c , $a, "Using s///e to foldcase."); 175 176($c = $b) =~ s/(\p{IsWord}+)/lcfirst($1)/ge; 177is($c , "\x{3c3}FOO.bAR", "Using s///e to change case."); 178 179($c = $a) =~ s/(\p{IsWord}+)/ucfirst($1)/ge; 180is($c , "\x{3a3}foo.Bar", "Using s///e to change case."); 181 182# #18931: perl5.8.0 bug in \U..\E processing 183# Test case from Nicholas Clark. 184for my $a (0,1) { 185 $_ = 'abcdefgh'; 186 $_ .= chr 256; 187 chop; 188 /(.*)/; 189 is(uc($1), "ABCDEFGH", "[perl #18931]"); 190} 191 192{ 193 foreach (0, 1) { 194 $a = v10.v257; 195 chop $a; 196 $a =~ s/^(\s*)(\w*)/$1\u$2/; 197 is($a, v10, "[perl #18857]"); 198 } 199} 200 201 202# [perl #38619] Bug in lc and uc (interaction between UTF-8, substr, and lc/uc) 203 204for ("a\x{100}", "xyz\x{100}") { 205 is(substr(uc($_), 0), uc($_), "[perl #38619] uc"); 206} 207for ("A\x{100}", "XYZ\x{100}") { 208 is(substr(lc($_), 0), lc($_), "[perl #38619] lc"); 209} 210for ("a\x{100}", "�yz\x{100}") { # � to Ss (different length) 211 is(substr(ucfirst($_), 0), ucfirst($_), "[perl #38619] ucfirst"); 212} 213 214#fc() didn't exist back then, but coverage is coverage. 215for ("a\x{100}", "�yz\x{100}", "xyz\x{100}", "XYZ\x{100}") { # � to Ss (different length) 216 is(substr(fc($_), 0), fc($_), "[perl #38619] fc"); 217} 218 219# Related to [perl #38619] 220# the original report concerns PERL_MAGIC_utf8. 221# these cases concern PERL_MAGIC_regex_global. 222 223for (map { $_ } "a\x{100}", "abc\x{100}", "\x{100}") { 224 chop; # get ("a", "abc", "") in utf8 225 my $return = uc($_) =~ /\G(.?)/g; 226 my $result = $return ? $1 : "not"; 227 my $expect = (uc($_) =~ /(.?)/g)[0]; 228 is($return, 1, "[perl #38619]"); 229 is($result, $expect, "[perl #38619]"); 230} 231 232for (map { $_ } "A\x{100}", "ABC\x{100}", "\x{100}") { 233 chop; # get ("A", "ABC", "") in utf8 234 my $return = lc($_) =~ /\G(.?)/g; 235 my $result = $return ? $1 : "not"; 236 my $expect = (lc($_) =~ /(.?)/g)[0]; 237 is($return, 1, "[perl #38619]"); 238 is($result, $expect, "[perl #38619]"); 239} 240 241for (map { $_ } "A\x{100}", "ABC\x{100}", "\x{100}") { 242 chop; # get ("A", "ABC", "") in utf8 243 my $return = fc($_) =~ /\G(.?)/g; 244 my $result = $return ? $1 : "not"; 245 my $expect = (fc($_) =~ /(.?)/g)[0]; 246 is($return, 1, "[perl #38619]"); 247 is($result, $expect, "[perl #38619]"); 248} 249 250for (1, 4, 9, 16, 25) { 251 is(uc "\x{03B0}" x $_, "\x{3a5}\x{308}\x{301}" x $_, 252 'uc U+03B0 grows threefold'); 253 254 is(lc "\x{0130}" x $_, "i\x{307}" x $_, 'lc U+0130 grows'); 255 256 is(fc "\x{03B0}" x $_, "\x{3C5}\x{308}\x{301}" x $_, 257 'fc U+03B0 grows threefold'); 258} 259 260# bug #43207 261my $temp = "HellO"; 262for ("$temp") { 263 lc $_; 264 is($_, "HellO", '[perl #43207] lc($_) modifying $_'); 265} 266for ("$temp") { 267 fc $_; 268 is($_, "HellO", '[perl #43207] fc($_) modifying $_'); 269} 270for ("$temp") { 271 uc $_; 272 is($_, "HellO", '[perl #43207] uc($_) modifying $_'); 273} 274for ("$temp") { 275 ucfirst $_; 276 is($_, "HellO", '[perl #43207] ucfirst($_) modifying $_'); 277} 278for ("$temp") { 279 lcfirst $_; 280 is($_, "HellO", '[perl #43207] lcfirst($_) modifying $_'); 281} 282 283# new in Unicode 5.1.0 284is(lc("\x{1E9E}"), "\x{df}", "lc(LATIN CAPITAL LETTER SHARP S)"); 285 286{ 287 use feature 'unicode_strings'; 288 use bytes; 289 is(lc("\xc0"), "\xc0", "lc of above-ASCII Latin1 is itself under use bytes"); 290 is(lcfirst("\xc0"), "\xc0", "lcfirst of above-ASCII Latin1 is itself under use bytes"); 291 is(uc("\xe0"), "\xe0", "uc of above-ASCII Latin1 is itself under use bytes"); 292 is(ucfirst("\xe0"), "\xe0", "ucfirst of above-ASCII Latin1 is itself under use bytes"); 293} 294 295# Brought up in ticket #117855: Constant folding applied to uc() should use 296# the right set of hints. 297fresh_perl_like(<<'constantfolding', qr/^(\d+),\1\z/, {}, 298 my $function = "uc"; 299 my $char = "\xff"; 300 { 301 use feature 'unicode_strings'; 302 print ord uc($char), ",", 303 ord eval "$function('$char')", "\n"; 304 } 305constantfolding 306 'folded uc() in string eval uses the right hints'); 307 308# In-place lc/uc should not corrupt string buffers when given a non-utf8- 309# flagged thingy that stringifies to utf8 310$h{k} = bless[], "\x{3b0}\x{3b0}\x{3b0}bcde"; # U+03B0 grows with uc() 311 # using delete marks it as TEMP, so uc-in-place is permitted 312like uc delete $h{k}, qr "^(?:\x{3a5}\x{308}\x{301}){3}BCDE=ARRAY\(.*\)", 313 'uc(TEMP ref) does not produce a corrupt string'; 314$h{k} = bless[], "\x{130}bcde"; # U+0130 grows with lc() 315 # using delete marks it as TEMP, so uc-in-place is permitted 316like lc delete $h{k}, qr "^i\x{307}bcde=array\(.*\)", 317 'lc(TEMP ref) does not produce a corrupt string'; 318 319 320my $utf8_locale = find_utf8_ctype_locale(); 321 322SKIP: { 323 skip 'Can\'t find a UTF-8 locale', 4*256 unless defined $utf8_locale; 324 325 use feature qw( unicode_strings ); 326 327 no locale; 328 329 my @unicode_lc; 330 my @unicode_uc; 331 my @unicode_lcfirst; 332 my @unicode_ucfirst; 333 334 # Get all the values outside of 'locale' 335 for my $i (0 .. 255) { 336 push @unicode_lc, lc(chr $i); 337 push @unicode_uc, uc(chr $i); 338 push @unicode_lcfirst, lcfirst(chr $i); 339 push @unicode_ucfirst, ucfirst(chr $i); 340 } 341 342 use if $Config{d_setlocale}, qw(POSIX locale_h); 343 use locale; 344 setlocale(LC_CTYPE, $utf8_locale); 345 346 for my $i (0 .. 255) { 347 is(lc(chr $i), $unicode_lc[$i], "In a UTF-8 locale, lc(chr $i) is the same as official Unicode"); 348 is(uc(chr $i), $unicode_uc[$i], "In a UTF-8 locale, uc(chr $i) is the same as official Unicode"); 349 is(lcfirst(chr $i), $unicode_lcfirst[$i], "In a UTF-8 locale, lcfirst(chr $i) is the same as official Unicode"); 350 is(ucfirst(chr $i), $unicode_ucfirst[$i], "In a UTF-8 locale, ucfirst(chr $i) is the same as official Unicode"); 351 } 352} 353