1#!./perl 2 3# 4# test the bit operators '&', '|', '^', '~', '<<', and '>>' 5# 6 7use warnings; 8 9BEGIN { 10 chdir 't' if -d 't'; 11 require "./test.pl"; 12 set_up_inc('../lib'); 13 require "./charset_tools.pl"; 14 require Config; 15} 16 17# Tests don't have names yet. 18# If you find tests are failing, please try adding names to tests to track 19# down where the failure is, and supply your new names as a patch. 20# (Just-in-time test naming) 21plan tests => 502; 22 23# numerics 24ok ((0xdead & 0xbeef) == 0x9ead); 25ok ((0xdead | 0xbeef) == 0xfeef); 26ok ((0xdead ^ 0xbeef) == 0x6042); 27ok ((~0xdead & 0xbeef) == 0x2042); 28 29# shifts 30ok ((257 << 7) == 32896); 31ok ((33023 >> 7) == 257); 32 33# signed vs. unsigned 34ok ((~0 > 0 && do { use integer; ~0 } == -1)); 35 36my $bits = 0; 37for (my $i = ~0; $i; $i >>= 1) { ++$bits; } 38my $cusp = 1 << ($bits - 1); 39 40 41ok (($cusp & -1) > 0 && do { use integer; $cusp & -1 } < 0); 42ok (($cusp | 1) > 0 && do { use integer; $cusp | 1 } < 0); 43ok (($cusp ^ 1) > 0 && do { use integer; $cusp ^ 1 } < 0); 44ok ((1 << ($bits - 1)) == $cusp && 45 do { use integer; 1 << ($bits - 1) } == -$cusp); 46ok (($cusp >> 1) == ($cusp / 2) && 47 do { use integer; abs($cusp >> 1) } == ($cusp / 2)); 48 49$Aaz = chr(ord("A") & ord("z")); 50$Aoz = chr(ord("A") | ord("z")); 51$Axz = chr(ord("A") ^ ord("z")); 52 53# short strings 54is (("AAAAA" & "zzzzz"), ($Aaz x 5)); 55is (("AAAAA" | "zzzzz"), ($Aoz x 5)); 56is (("AAAAA" ^ "zzzzz"), ($Axz x 5)); 57 58# long strings 59$foo = "A" x 150; 60$bar = "z" x 75; 61$zap = "A" x 75; 62# & truncates 63is (($foo & $bar), ($Aaz x 75 )); 64# | does not truncate 65is (($foo | $bar), ($Aoz x 75 . $zap)); 66# ^ does not truncate 67is (($foo ^ $bar), ($Axz x 75 . $zap)); 68 69# string constants. These tests expect the bit patterns of these strings in 70# ASCII, so convert to that. 71sub _and($) { $_[0] & native_to_uni("+0") } 72sub _oar($) { $_[0] | native_to_uni("+0") } 73sub _xor($) { $_[0] ^ native_to_uni("+0") } 74is _and native_to_uni("waf"), native_to_uni('# '), 'str var & const str'; # [perl #20661] 75is _and native_to_uni("waf"), native_to_uni('# '), 'str var & const str again'; # [perl #20661] 76is _oar native_to_uni("yit"), native_to_uni('{yt'), 'str var | const str'; 77is _oar native_to_uni("yit"), native_to_uni('{yt'), 'str var | const str again'; 78is _xor native_to_uni("yit"), native_to_uni('RYt'), 'str var ^ const str'; 79is _xor native_to_uni("yit"), native_to_uni('RYt'), 'str var ^ const str again'; 80 81SKIP: { 82 skip "Converting a numeric doesn't work with EBCDIC unlike the above tests", 83 3 if $::IS_EBCDIC; 84 is _and 0, '0', 'num var & const str'; # [perl #20661] 85 is _oar 0, '0', 'num var | const str'; 86 is _xor 0, '0', 'num var ^ const str'; 87} 88 89# But don’t mistake a COW for a constant when assigning to it 90%h=(150=>1); 91$i=(keys %h)[0]; 92$i |= 105; 93is $i, 255, '[perl #108480] $cow |= number'; 94$i=(keys %h)[0]; 95$i &= 105; 96is $i, 0, '[perl #108480] $cow &= number'; 97$i=(keys %h)[0]; 98$i ^= 105; 99is $i, 255, '[perl #108480] $cow ^= number'; 100 101# 102is ("ok \xFF\xFF\n" & "ok 19\n", "ok 19\n"); 103is ("ok 20\n" | "ok \0\0\n", "ok 20\n"); 104is ("o\000 \0001\000" ^ "\000k\0002\000\n", "ok 21\n"); 105 106# 107is ("ok \x{FF}\x{FF}\n" & "ok 22\n", "ok 22\n"); 108is ("ok 23\n" | "ok \x{0}\x{0}\n", "ok 23\n"); 109is ("o\x{0} \x{0}4\x{0}" ^ "\x{0}k\x{0}2\x{0}\n", "ok 24\n"); 110 111# More variations on 19 and 22. 112is ("ok \xFF\x{FF}\n" & "ok 41\n", "ok 41\n"); 113is ("ok \x{FF}\xFF\n" & "ok 42\n", "ok 42\n"); 114 115# Tests to see if you really can do casts negative floats to unsigned properly 116$neg1 = -1.0; 117ok (~ $neg1 == 0); 118$neg7 = -7.0; 119ok (~ $neg7 == 6); 120 121 122# double magic tests 123 124sub TIESCALAR { bless { value => $_[1], orig => $_[1] } } 125sub STORE { $_[0]{store}++; $_[0]{value} = $_[1] } 126sub FETCH { $_[0]{fetch}++; $_[0]{value} } 127sub stores { tied($_[0])->{value} = tied($_[0])->{orig}; 128 delete(tied($_[0])->{store}) || 0 } 129sub fetches { delete(tied($_[0])->{fetch}) || 0 } 130 131# numeric double magic tests 132 133tie $x, "main", 1; 134tie $y, "main", 3; 135 136is(($x | $y), 3); 137is(fetches($x), 1); 138is(fetches($y), 1); 139is(stores($x), 0); 140is(stores($y), 0); 141 142is(($x & $y), 1); 143is(fetches($x), 1); 144is(fetches($y), 1); 145is(stores($x), 0); 146is(stores($y), 0); 147 148is(($x ^ $y), 2); 149is(fetches($x), 1); 150is(fetches($y), 1); 151is(stores($x), 0); 152is(stores($y), 0); 153 154is(($x |= $y), 3); 155is(fetches($x), 2); 156is(fetches($y), 1); 157is(stores($x), 1); 158is(stores($y), 0); 159 160is(($x &= $y), 1); 161is(fetches($x), 2); 162is(fetches($y), 1); 163is(stores($x), 1); 164is(stores($y), 0); 165 166is(($x ^= $y), 2); 167is(fetches($x), 2); 168is(fetches($y), 1); 169is(stores($x), 1); 170is(stores($y), 0); 171 172is(~~$y, 3); 173is(fetches($y), 1); 174is(stores($y), 0); 175 176{ use integer; 177 178is(($x | $y), 3); 179is(fetches($x), 1); 180is(fetches($y), 1); 181is(stores($x), 0); 182is(stores($y), 0); 183 184is(($x & $y), 1); 185is(fetches($x), 1); 186is(fetches($y), 1); 187is(stores($x), 0); 188is(stores($y), 0); 189 190is(($x ^ $y), 2); 191is(fetches($x), 1); 192is(fetches($y), 1); 193is(stores($x), 0); 194is(stores($y), 0); 195 196is(($x |= $y), 3); 197is(fetches($x), 2); 198is(fetches($y), 1); 199is(stores($x), 1); 200is(stores($y), 0); 201 202is(($x &= $y), 1); 203is(fetches($x), 2); 204is(fetches($y), 1); 205is(stores($x), 1); 206is(stores($y), 0); 207 208is(($x ^= $y), 2); 209is(fetches($x), 2); 210is(fetches($y), 1); 211is(stores($x), 1); 212is(stores($y), 0); 213 214is(~$y, -4); 215is(fetches($y), 1); 216is(stores($y), 0); 217 218} # end of use integer; 219 220# stringwise double magic tests 221 222tie $x, "main", "a"; 223tie $y, "main", "c"; 224 225is(($x | $y), ("a" | "c")); 226is(fetches($x), 1); 227is(fetches($y), 1); 228is(stores($x), 0); 229is(stores($y), 0); 230 231is(($x & $y), ("a" & "c")); 232is(fetches($x), 1); 233is(fetches($y), 1); 234is(stores($x), 0); 235is(stores($y), 0); 236 237is(($x ^ $y), ("a" ^ "c")); 238is(fetches($x), 1); 239is(fetches($y), 1); 240is(stores($x), 0); 241is(stores($y), 0); 242 243is(($x |= $y), ("a" | "c")); 244is(fetches($x), 2); 245is(fetches($y), 1); 246is(stores($x), 1); 247is(stores($y), 0); 248 249is(($x &= $y), ("a" & "c")); 250is(fetches($x), 2); 251is(fetches($y), 1); 252is(stores($x), 1); 253is(stores($y), 0); 254 255is(($x ^= $y), ("a" ^ "c")); 256is(fetches($x), 2); 257is(fetches($y), 1); 258is(stores($x), 1); 259is(stores($y), 0); 260 261is(~~$y, "c"); 262is(fetches($y), 1); 263is(stores($y), 0); 264 265my $g; 266# Note: if the vec() reads are part of the is() calls it's treated as 267# in lvalue context, so we save it separately 268$g = vec($x, 0, 1); 269is($g, (ord("a") & 0x01), "check vec value"); 270is(fetches($x), 1, "fetches for vec read"); 271is(stores($x), 0, "stores for vec read"); 272# similarly here, and code like: 273# $g = (vec($x, 0, 1) = 0) 274# results in an extra fetch, since the inner assignment returns the LV 275vec($x, 0, 1) = 0; 276# one fetch in vec() another when the LV is assigned to 277is(fetches($x), 2, "fetches for vec write"); 278is(stores($x), 1, "stores for vec write"); 279 280{ 281 my $a = "a"; 282 utf8::upgrade($a); 283 tie $x, "main", $a; 284 $g = vec($x, 0, 1); 285 is($g, (ord("a") & 0x01), "check vec value (utf8)"); 286 is(fetches($x), 1, "fetches for vec read (utf8)"); 287 is(stores($x), 0, "stores for vec read (utf8)"); 288 vec($x, 0, 1) = 0; 289 # one fetch in vec() another when the LV is assigned to 290 is(fetches($x), 2, "fetches for vec write (utf8)"); 291 is(stores($x), 1, "stores for vec write (utf8)"); 292} 293 294$a = "\0\x{100}"; chop($a); 295ok(utf8::is_utf8($a)); # make sure UTF8 flag is still there 296$a = ~$a; 297is($a, "\xFF", "~ works with utf-8"); 298ok(! utf8::is_utf8($a), " and turns off the UTF-8 flag"); 299 300$a = "\0\x{100}"; chop($a); 301undef $b; 302$b = $a | "\xFF"; 303ok(utf8::is_utf8($b), "Verify UTF-8 | non-UTF-8 retains UTF-8 flag"); 304undef $b; 305$b = "\xFF" | $a; 306ok(utf8::is_utf8($b), "Verify non-UTF-8 | UTF-8 retains UTF-8 flag"); 307undef $b; 308$b = $a & "\xFF"; 309ok(utf8::is_utf8($b), "Verify UTF-8 & non-UTF-8 retains UTF-8 flag"); 310undef $b; 311$b = "\xFF" & $a; 312ok(utf8::is_utf8($b), "Verify non-UTF-8 & UTF-8 retains UTF-8 flag"); 313undef $b; 314$b = $a ^ "\xFF"; 315ok(utf8::is_utf8($b), "Verify UTF-8 ^ non-UTF-8 retains UTF-8 flag"); 316undef $b; 317$b = "\xFF" ^ $a; 318ok(utf8::is_utf8($b), "Verify non-UTF-8 ^ UTF-8 retains UTF-8 flag"); 319 320 321# [rt.perl.org 33003] 322# This would cause a segfault without malloc wrap 323SKIP: { 324 skip "No malloc wrap checks" unless $Config::Config{usemallocwrap}; 325 like( runperl(prog => 'eval q($#a>>=1); print 1'), qr/^1\n?/ ); 326} 327 328# [perl #37616] Bug in &= (string) and/or m// 329{ 330 $a = "aa"; 331 $a &= "a"; 332 ok($a =~ /a+$/, 'ASCII "a" is NUL-terminated'); 333 334 $b = "bb\x{FF}"; 335 utf8::upgrade($b); 336 $b &= "b"; 337 ok($b =~ /b+$/, 'Unicode "b" is NUL-terminated'); 338} 339 340# New string- and number-specific bitwise ops 341{ 342 use feature "bitwise"; 343 no warnings "experimental::bitwise"; 344 is "22" & "66", 2, 'numeric & with strings'; 345 is "22" | "66", 86, 'numeric | with strings'; 346 is "22" ^ "66", 84, 'numeric ^ with strings'; 347 is ~"22" & 0xff, 233, 'numeric ~ with string'; 348 is 22 &. 66, 22, '&. with numbers'; 349 is 22 |. 66, 66, '|. with numbers'; 350 is 22 ^. 66, "\4\4", '^. with numbers'; 351 if ($::IS_EBCDIC) { 352 # ord('2') is 0xF2 on EBCDIC 353 is ~.22, "\x0d\x0d", '~. with number'; 354 } 355 else { 356 # ord('2') is 0x32 on ASCII 357 is ~.22, "\xcd\xcd", '~. with number'; 358 } 359 $_ = "22"; 360 is $_ &= "66", 2, 'numeric &= with strings'; 361 $_ = "22"; 362 is $_ |= "66", 86, 'numeric |= with strings'; 363 $_ = "22"; 364 is $_ ^= "66", 84, 'numeric ^= with strings'; 365 $_ = 22; 366 is $_ &.= 66, 22, '&.= with numbers'; 367 $_ = 22; 368 is $_ |.= 66, 66, '|.= with numbers'; 369 $_ = 22; 370 is $_ ^.= 66, "\4\4", '^.= with numbers'; 371 372 # signed vs. unsigned 373 ok ((~0 > 0 && do { use integer; ~0 } == -1)); 374 375 my $bits = 0; 376 for (my $i = ~0; $i; $i >>= 1) { ++$bits; } 377 my $cusp = 1 << ($bits - 1); 378 379 ok (($cusp & -1) > 0 && do { use integer; $cusp & -1 } < 0); 380 ok (($cusp | 1) > 0 && do { use integer; $cusp | 1 } < 0); 381 ok (($cusp ^ 1) > 0 && do { use integer; $cusp ^ 1 } < 0); 382 ok ((1 << ($bits - 1)) == $cusp && 383 do { use integer; 1 << ($bits - 1) } == -$cusp); 384 ok (($cusp >> 1) == ($cusp / 2) && 385 do { use integer; abs($cusp >> 1) } == ($cusp / 2)); 386} 387# Repeat some of those, with 'use v5.27' 388{ 389 use v5.27; 390 391 is "22" & "66", 2, 'numeric & with strings'; 392 is "22" | "66", 86, 'numeric | with strings'; 393 is "22" ^ "66", 84, 'numeric ^ with strings'; 394 is ~"22" & 0xff, 233, 'numeric ~ with string'; 395 is 22 &. 66, 22, '&. with numbers'; 396 is 22 |. 66, 66, '|. with numbers'; 397 is 22 ^. 66, "\4\4", '^. with numbers'; 398 if ($::IS_EBCDIC) { 399 # ord('2') is 0xF2 on EBCDIC 400 is ~.22, "\x0d\x0d", '~. with number'; 401 } 402 else { 403 # ord('2') is 0x32 on ASCII 404 is ~.22, "\xcd\xcd", '~. with number'; 405 } 406 $_ = "22"; 407 is $_ &= "66", 2, 'numeric &= with strings'; 408 $_ = "22"; 409 is $_ |= "66", 86, 'numeric |= with strings'; 410 $_ = "22"; 411 is $_ ^= "66", 84, 'numeric ^= with strings'; 412 $_ = 22; 413 is $_ &.= 66, 22, '&.= with numbers'; 414 $_ = 22; 415 is $_ |.= 66, 66, '|.= with numbers'; 416 $_ = 22; 417 is $_ ^.= 66, "\4\4", '^.= with numbers'; 418} 419 420# ref tests 421 422my %res; 423 424for my $str ("x", "\x{B6}") { 425 utf8::upgrade($str) if $str !~ /x/; 426 for my $chr (qw/S A H G X ( * F/) { 427 for my $op (qw/| & ^/) { 428 my $co = ord $chr; 429 my $so = ord $str; 430 $res{"$chr$op$str"} = eval qq/chr($co $op $so)/; 431 } 432 } 433 $res{"undef|$str"} = $str; 434 $res{"undef&$str"} = ""; 435 $res{"undef^$str"} = $str; 436} 437 438sub PVBM () { "X" } 4391 if index "foo", PVBM; 440 441my $warn = 0; 442local $^W = 1; 443local $SIG{__WARN__} = sub { $warn++ }; 444 445sub is_first { 446 my ($got, $orig, $op, $str, $name) = @_; 447 is(substr($got, 0, 1), $res{"$orig$op$str"}, $name); 448} 449 450for ( 451 # [object to test, first char of stringification, name] 452 [undef, "undef", "undef" ], 453 [\1, "S", "scalar ref" ], 454 [[], "A", "array ref" ], 455 [{}, "H", "hash ref" ], 456 [qr/x/, "(", "qr//" ], 457 [*foo, "*", "glob" ], 458 [\*foo, "G", "glob ref" ], 459 [PVBM, "X", "PVBM" ], 460 [\PVBM, "S", "PVBM ref" ], 461 [bless([], "Foo"), "F", "object" ], 462) { 463 my ($val, $orig, $type) = @$_; 464 465 for (["x", "string"], ["\x{B6}", "utf8"]) { 466 my ($str, $desc) = @$_; 467 utf8::upgrade($str) if $desc =~ /utf8/; 468 469 $warn = 0; 470 471 is_first($val | $str, $orig, "|", $str, "$type | $desc"); 472 is_first($val & $str, $orig, "&", $str, "$type & $desc"); 473 is_first($val ^ $str, $orig, "^", $str, "$type ^ $desc"); 474 475 is_first($str | $val, $orig, "|", $str, "$desc | $type"); 476 is_first($str & $val, $orig, "&", $str, "$desc & $type"); 477 is_first($str ^ $val, $orig, "^", $str, "$desc ^ $type"); 478 479 my $new; 480 ($new = $val) |= $str; 481 is_first($new, $orig, "|", $str, "$type |= $desc"); 482 ($new = $val) &= $str; 483 is_first($new, $orig, "&", $str, "$type &= $desc"); 484 ($new = $val) ^= $str; 485 is_first($new, $orig, "^", $str, "$type ^= $desc"); 486 487 ($new = $str) |= $val; 488 is_first($new, $orig, "|", $str, "$desc |= $type"); 489 ($new = $str) &= $val; 490 is_first($new, $orig, "&", $str, "$desc &= $type"); 491 ($new = $str) ^= $val; 492 is_first($new, $orig, "^", $str, "$desc ^= $type"); 493 494 if ($orig eq "undef") { 495 # undef |= and undef ^= don't warn 496 is($warn, 10, "no duplicate warnings"); 497 } 498 else { 499 is($warn, 0, "no warnings"); 500 } 501 } 502} 503 504delete $SIG{__WARN__}; 505 506my $strval; 507 508{ 509 package Bar; 510 use overload q/""/ => sub { $strval }; 511 512 package Baz; 513 use overload q/|/ => sub { "y" }; 514} 515 516ok(!eval { 1 if bless([], "Bar") | "x"; 1 },"string overload can't use |"); 517like($@, qr/no method found/, "correct error"); 518is(eval { bless([], "Baz") | "x" }, "y", "| overload works"); 519 520my $obj = bless [], "Bar"; 521$strval = "x"; 522eval { $obj |= "Q" }; 523$strval = "z"; 524is("$obj", "z", "|= doesn't break string overload"); 525 526# [perl #29070] 527$^A .= new version ~$_ for eval sprintf('"\\x%02x"', 0xff - ord("1")), 528 $::IS_EBCDIC ? v13 : v205, # 255 - ord('2') 529 eval sprintf('"\\x%02x"', 0xff - ord("3")); 530is $^A, "123", '~v0 clears vstring magic on retval'; 531 532{ 533 my $w = $Config::Config{ivsize} * 8; 534 535 fail("unexpected w $w") unless $w == 32 || $w == 64; 536 537 is(1 << 1, 2, "UV 1 left shift 1"); 538 is(1 >> 1, 0, "UV 1 right shift 1"); 539 540 is(0x7b << -4, 0x007, "UV left negative shift == right shift"); 541 is(0x7b >> -4, 0x7b0, "UV right negative shift == left shift"); 542 543 is(0x7b << 0, 0x07b, "UV left zero shift == identity"); 544 is(0x7b >> 0, 0x07b, "UV right zero shift == identity"); 545 546 is(0x0 << -1, 0x0, "zero left negative shift == zero"); 547 is(0x0 >> -1, 0x0, "zero right negative shift == zero"); 548 549 cmp_ok(1 << $w - 1, '==', 2 ** ($w - 1), # not is() because NV stringify. 550 "UV left $w - 1 shift == 2 ** ($w - 1)"); 551 is(1 << $w, 0, "UV left shift $w == zero"); 552 is(1 << $w + 1, 0, "UV left shift $w + 1 == zero"); 553 554 is(1 >> $w - 1, 0, "UV right shift $w - 1 == zero"); 555 is(1 >> $w, 0, "UV right shift $w == zero"); 556 is(1 >> $w + 1, 0, "UV right shift $w + 1 == zero"); 557 558 # Negative shiftees get promoted to UVs before shifting. This is 559 # not necessarily the ideal behavior, but that is what is happening. 560 if ($w == 64) { 561 no warnings "portable"; 562 no warnings "overflow"; # prevent compile-time warning for ivsize=4 563 is(-1 << 1, 0xFFFF_FFFF_FFFF_FFFE, 564 "neg UV (sic) left shift = 0xFF..E"); 565 is(-1 >> 1, 0x7FFF_FFFF_FFFF_FFFF, 566 "neg UV (sic) right shift = 0x7F..F"); 567 } elsif ($w == 32) { 568 no warnings "portable"; 569 is(-1 << 1, 0xFFFF_FFFE, "neg left shift == 0xFF..E"); 570 is(-1 >> 1, 0x7FFF_FFFF, "neg right shift == 0x7F..F"); 571 } 572 573 { 574 # 'use integer' means use IVs instead of UVs. 575 use integer; 576 577 # No surprises here. 578 is(1 << 1, 2, "IV 1 left shift 1 == 2"); 579 is(1 >> 1, 0, "IV 1 right shift 1 == 0"); 580 581 # The left overshift should behave like without 'use integer', 582 # that is, return zero. 583 is(1 << $w, 0, "IV 1 left shift $w == 0"); 584 is(1 << $w + 1, 0, "IV 1 left shift $w + 1 == 0"); 585 is(-1 << $w, 0, "IV -1 left shift $w == 0"); 586 is(-1 << $w + 1, 0, "IV -1 left shift $w + 1 == 0"); 587 588 # Even for negative IVs, left shift is multiplication. 589 # But right shift should display the stuckiness to -1. 590 is(-1 << 1, -2, "IV -1 left shift 1 == -2"); 591 is(-1 >> 1, -1, "IV -1 right shift 1 == -1"); 592 593 # As for UVs, negative shifting means the reverse shift. 594 is(-1 << -1, -1, "IV -1 left shift -1 == -1"); 595 is(-1 >> -1, -2, "IV -1 right shift -1 == -2"); 596 597 # Test also at and around wordsize, expect stuckiness to -1. 598 is(-1 >> $w - 1, -1, "IV -1 right shift $w - 1 == -1"); 599 is(-1 >> $w, -1, "IV -1 right shift $w == -1"); 600 is(-1 >> $w + 1, -1, "IV -1 right shift $w + 1 == -1"); 601 } 602} 603 604# [perl #129287] UTF8 & was not providing a trailing null byte. 605# This test is a bit convoluted, as we want to make sure that the string 606# allocated for &’s target contains memory initialised to something other 607# than a null byte. Uninitialised memory does not make for a reliable 608# test. So we do &. on a longer non-utf8 string first. 609for (["aaa","aaa"],[substr ("a\x{100}",0,1), "a"]) { 610 use feature "bitwise"; 611 no warnings "experimental::bitwise", "pack"; 612 $byte = substr unpack("P2", pack "P", $$_[0] &. $$_[1]), -1; 613} 614is $byte, "\0", "utf8 &. appends null byte"; 615 616# only visible under sanitize 617fresh_perl_is('$x = "UUUUUUUV"; $y = "xxxxxxx"; $x |= $y; print $x', 618 ( $::IS_EBCDIC) ? 'XXXXXXXV' : '}}}}}}}V', 619 {}, "[perl #129995] access to freed memory"); 620 621 622# 623# Using code points above 0xFF is fatal 624# 625foreach my $op_info ([and => "&"], [or => "|"], [xor => "^"]) { 626 my ($op_name, $op) = @$op_info; 627 local $@; 628 eval '$_ = "\xFF" ' . $op . ' "\x{100}";'; 629 like $@, qr /^Use of strings with code points over 0xFF as arguments (?# 630 )to bitwise $op_name \Q($op)\E operator is not allowed/, 631 "Use of code points above 0xFF as arguments to bitwise " . 632 "$op_name ($op) is not allowed"; 633} 634 635{ 636 local $@; 637 eval '$_ = ~ "\x{100}";'; 638 like $@, qr /^Use of strings with code points over 0xFF as arguments (?# 639 )to 1's complement \(~\) operator is not allowed/, 640 "Use of code points above 0xFF as argument to 1's complement " . 641 "(~) is not allowed"; 642} 643 644{ 645 # RT 134140 fatalizations 646 my %op_pairs = ( 647 and => { low => 'and', high => '&', regex => qr/&/ }, 648 or => { low => 'or', high => '|', regex => qr/\|/ }, 649 xor => { low => 'xor', high => '^', regex => qr/\^/ }, 650 ); 651 my @combos = ( 652 { string => '"abc" & "abc\x{100}"', op_pair => $op_pairs{and} }, 653 { string => '"abc" | "abc\x{100}"', op_pair => $op_pairs{or} }, 654 { string => '"abc" ^ "abc\x{100}"', op_pair => $op_pairs{xor} }, 655 { string => '"abc\x{100}" & "abc"', op_pair => $op_pairs{and} }, 656 { string => '"abc\x{100}" | "abc"', op_pair => $op_pairs{or} }, 657 { string => '"abc\x{100}" ^ "abc"', op_pair => $op_pairs{xor} }, 658 659 ); 660 661 # Use of strings with code points over 0xFF as arguments to %s operator is not allowed 662 for my $h (@combos) { 663 my $s1 = "Use of strings with code points over 0xFF as arguments to bitwise"; 664 my $s2 = "operator is not allowed"; 665 my $expected = qr/$s1 $h->{op_pair}->{low} \($h->{op_pair}->{regex}\) $s2/; 666 my $description = "$s1 $h->{op_pair}->{low} ($h->{op_pair}->{high}) operator is not allowed"; 667 local $@; 668 eval $h->{string}; 669 like $@, $expected, $description; 670 } 671} 672 673{ 674 # perl #17844 - only visible with valgrind/ASAN 675 fresh_perl_is(<<'EOS', 676formline X000n^\\0,\\0^\\0for\0,0..10 677EOS 678 '', 679 {}, "[perl #17844] access beyond end of block"); 680} 681