regexp.js 27.1 KB
Newer Older
1
// Copyright 2012 the V8 project authors. All rights reserved.
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
//     * Redistributions of source code must retain the above copyright
//       notice, this list of conditions and the following disclaimer.
//     * Redistributions in binary form must reproduce the above
//       copyright notice, this list of conditions and the following
//       disclaimer in the documentation and/or other materials provided
//       with the distribution.
//     * Neither the name of Google Inc. nor the names of its
//       contributors may be used to endorse or promote products derived
//       from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

28 29
// Flags: --allow-natives-syntax

30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
function testEscape(str, regex) {
  assertEquals("foo:bar:baz", str.split(regex).join(":"));
}

testEscape("foo\nbar\nbaz", /\n/);
testEscape("foo bar baz", /\s/);
testEscape("foo\tbar\tbaz", /\s/);
testEscape("foo-bar-baz", /\u002D/);

// Test containing null char in regexp.
var s = '[' + String.fromCharCode(0) + ']';
var re = new RegExp(s);
assertEquals(s.match(re).length, 1);
assertEquals(s.match(re)[0], String.fromCharCode(0));

// Test strings containing all line separators
s = 'aA\nbB\rcC\r\ndD\u2028eE\u2029fF';
re = /^./gm; // any non-newline character at the beginning of a line
var result = s.match(re);
assertEquals(result.length, 6);
assertEquals(result[0], 'a');
assertEquals(result[1], 'b');
assertEquals(result[2], 'c');
assertEquals(result[3], 'd');
assertEquals(result[4], 'e');
assertEquals(result[5], 'f');

re = /.$/gm; // any non-newline character at the end of a line
result = s.match(re);
assertEquals(result.length, 6);
assertEquals(result[0], 'A');
assertEquals(result[1], 'B');
assertEquals(result[2], 'C');
assertEquals(result[3], 'D');
assertEquals(result[4], 'E');
assertEquals(result[5], 'F');

re = /^[^]/gm; // *any* character at the beginning of a line
result = s.match(re);
assertEquals(result.length, 7);
assertEquals(result[0], 'a');
assertEquals(result[1], 'b');
assertEquals(result[2], 'c');
assertEquals(result[3], '\n');
assertEquals(result[4], 'd');
assertEquals(result[5], 'e');
assertEquals(result[6], 'f');

re = /[^]$/gm; // *any* character at the end of a line
result = s.match(re);
assertEquals(result.length, 7);
assertEquals(result[0], 'A');
assertEquals(result[1], 'B');
assertEquals(result[2], 'C');
assertEquals(result[3], '\r');
assertEquals(result[4], 'D');
assertEquals(result[5], 'E');
assertEquals(result[6], 'F');

89
// Some tests from the Mozilla tests, where our behavior used to differ from
90 91 92 93
// SpiderMonkey.
// From ecma_3/RegExp/regress-334158.js
assertTrue(/\ca/.test( "\x01" ));
assertFalse(/\ca/.test( "\\ca" ));
94 95 96
assertFalse(/\ca/.test( "ca" ));
assertTrue(/\c[a/]/.test( "\\ca" ));
assertTrue(/\c[a/]/.test( "\\c/" ));
97

98 99 100 101 102 103 104 105 106 107
// Test \c in character class
re = /^[\cM]$/;
assertTrue(re.test("\r"));
assertFalse(re.test("M"));
assertFalse(re.test("c"));
assertFalse(re.test("\\"));
assertFalse(re.test("\x03"));  // I.e., read as \cc

re = /^[\c]]$/;
assertTrue(re.test("c]"));
108
assertTrue(re.test("\\]"));
109 110 111
assertFalse(re.test("\x1d"));  // ']' & 0x1f
assertFalse(re.test("\x03]"));  // I.e., read as \cc

112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
re = /^[\c1]$/;  // Digit control characters are masked in character classes.
assertTrue(re.test("\x11"));
assertFalse(re.test("\\"));
assertFalse(re.test("c"));
assertFalse(re.test("1"));

re = /^[\c_]$/;  // Underscore control character is masked in character classes.
assertTrue(re.test("\x1f"));
assertFalse(re.test("\\"));
assertFalse(re.test("c"));
assertFalse(re.test("_"));

re = /^[\c$]$/;  // Other characters are interpreted literally.
assertFalse(re.test("\x04"));
assertTrue(re.test("\\"));
assertTrue(re.test("c"));
assertTrue(re.test("$"));

assertTrue(/^[Z-\c-e]*$/.test("Z[\\cde"));
131

132 133 134 135 136 137 138 139 140 141 142
// Test that we handle \s and \S correctly on special Unicode characters.
re = /\s/;
assertTrue(re.test("\u2028"));
assertTrue(re.test("\u2029"));
assertTrue(re.test("\uFEFF"));

re = /\S/;
assertFalse(re.test("\u2028"));
assertFalse(re.test("\u2029"));
assertFalse(re.test("\uFEFF"));

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180
// Test that we handle \s and \S correctly inside some bizarre
// character classes.
re = /[\s-:]/;
assertTrue(re.test('-'));
assertTrue(re.test(':'));
assertTrue(re.test(' '));
assertTrue(re.test('\t'));
assertTrue(re.test('\n'));
assertFalse(re.test('a'));
assertFalse(re.test('Z'));

re = /[\S-:]/;
assertTrue(re.test('-'));
assertTrue(re.test(':'));
assertFalse(re.test(' '));
assertFalse(re.test('\t'));
assertFalse(re.test('\n'));
assertTrue(re.test('a'));
assertTrue(re.test('Z'));

re = /[^\s-:]/;
assertFalse(re.test('-'));
assertFalse(re.test(':'));
assertFalse(re.test(' '));
assertFalse(re.test('\t'));
assertFalse(re.test('\n'));
assertTrue(re.test('a'));
assertTrue(re.test('Z'));

re = /[^\S-:]/;
assertFalse(re.test('-'));
assertFalse(re.test(':'));
assertTrue(re.test(' '));
assertTrue(re.test('\t'));
assertTrue(re.test('\n'));
assertFalse(re.test('a'));
assertFalse(re.test('Z'));

181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
re = /[\s]/;
assertFalse(re.test('-'));
assertFalse(re.test(':'));
assertTrue(re.test(' '));
assertTrue(re.test('\t'));
assertTrue(re.test('\n'));
assertFalse(re.test('a'));
assertFalse(re.test('Z'));

re = /[^\s]/;
assertTrue(re.test('-'));
assertTrue(re.test(':'));
assertFalse(re.test(' '));
assertFalse(re.test('\t'));
assertFalse(re.test('\n'));
assertTrue(re.test('a'));
assertTrue(re.test('Z'));

re = /[\S]/;
assertTrue(re.test('-'));
assertTrue(re.test(':'));
assertFalse(re.test(' '));
assertFalse(re.test('\t'));
assertFalse(re.test('\n'));
assertTrue(re.test('a'));
assertTrue(re.test('Z'));

re = /[^\S]/;
assertFalse(re.test('-'));
assertFalse(re.test(':'));
assertTrue(re.test(' '));
assertTrue(re.test('\t'));
assertTrue(re.test('\n'));
assertFalse(re.test('a'));
assertFalse(re.test('Z'));

re = /[\s\S]/;
assertTrue(re.test('-'));
assertTrue(re.test(':'));
assertTrue(re.test(' '));
assertTrue(re.test('\t'));
assertTrue(re.test('\n'));
assertTrue(re.test('a'));
assertTrue(re.test('Z'));

re = /[^\s\S]/;
assertFalse(re.test('-'));
assertFalse(re.test(':'));
assertFalse(re.test(' '));
assertFalse(re.test('\t'));
assertFalse(re.test('\n'));
assertFalse(re.test('a'));
assertFalse(re.test('Z'));

235 236 237 238 239 240 241 242 243 244 245
// First - is treated as range operator, second as literal minus.
// This follows the specification in parsing, but doesn't throw on
// the \s at the beginning of the range.
re = /[\s-0-9]/;
assertTrue(re.test(' '));
assertTrue(re.test('\xA0'));
assertTrue(re.test('-'));
assertTrue(re.test('0'));
assertTrue(re.test('9'));
assertFalse(re.test('1'));

246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
// Test beginning and end of line assertions with or without the
// multiline flag.
re = /^\d+/;
assertFalse(re.test("asdf\n123"));
re = /^\d+/m;
assertTrue(re.test("asdf\n123"));

re = /\d+$/;
assertFalse(re.test("123\nasdf"));
re = /\d+$/m;
assertTrue(re.test("123\nasdf"));

// Test that empty matches are handled correctly for multiline global
// regexps.
re = /^(.*)/mg;
assertEquals(3, "a\n\rb".match(re).length);
assertEquals("*a\n*b\r*c\n*\r*d\r*\n*e", "a\nb\rc\n\rd\r\ne".replace(re, "*$1"));

// Test that empty matches advance one character
re = new RegExp("", "g");
assertEquals("xAx", "A".replace(re, "x"));
assertEquals(3, String.fromCharCode(161).replace(re, "x").length);

// Test that we match the KJS behavior with regard to undefined constructor
// arguments:
re = new RegExp();
// KJS actually shows this as '//'.  Here we match the Firefox behavior (ie,
// giving a syntactically legal regexp literal).
assertEquals('/(?:)/', re.toString());
re = new RegExp(void 0);
assertEquals('/(?:)/', re.toString());
re.compile();
assertEquals('/(?:)/', re.toString());
re.compile(void 0);
280
assertEquals('/(?:)/', re.toString());
281 282


283 284
// Check for early syntax errors.
assertThrows("/foo(/gi");
285 286 287 288 289 290 291 292 293 294 295 296 297

// Check $01 and $10
re = new RegExp("(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)");
assertEquals("t", "123456789t".replace(re, "$10"), "$10");
assertEquals("15", "123456789t".replace(re, "$15"), "$10");
assertEquals("1", "123456789t".replace(re, "$01"), "$01");
assertEquals("$001", "123456789t".replace(re, "$001"), "$001");
re = new RegExp("foo(.)");
assertEquals("bar$0", "foox".replace(re, "bar$0"), "$0");
assertEquals("bar$00", "foox".replace(re, "bar$00"), "$00");
assertEquals("bar$000", "foox".replace(re, "bar$000"), "$000");
assertEquals("barx", "foox".replace(re, "bar$01"), "$01 2");
assertEquals("barx5", "foox".replace(re, "bar$15"), "$15");
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316

assertFalse(/()foo$\1/.test("football"), "football1");
assertFalse(/foo$(?=ball)/.test("football"), "football2");
assertFalse(/foo$(?!bar)/.test("football"), "football3");
assertTrue(/()foo$\1/.test("foo"), "football4");
assertTrue(/foo$(?=(ball)?)/.test("foo"), "football5");
assertTrue(/()foo$(?!bar)/.test("foo"), "football6");
assertFalse(/(x?)foo$\1/.test("football"), "football7");
assertFalse(/foo$(?=ball)/.test("football"), "football8");
assertFalse(/foo$(?!bar)/.test("football"), "football9");
assertTrue(/(x?)foo$\1/.test("foo"), "football10");
assertTrue(/foo$(?=(ball)?)/.test("foo"), "football11");
assertTrue(/foo$(?!bar)/.test("foo"), "football12");

// Check that the back reference has two successors.  See
// BackReferenceNode::PropagateForward.
assertFalse(/f(o)\b\1/.test('foo'));
assertTrue(/f(o)\B\1/.test('foo'));

317 318
// Back-reference, ignore case:
// ASCII
319 320 321 322 323
assertEquals("xaAx,a", String(/x(a)\1x/i.exec("xaAx")), "backref-ASCII");
assertFalse(/x(...)\1/i.test("xaaaaa"), "backref-ASCII-short");
assertTrue(/x((?:))\1\1x/i.test("xx"), "backref-ASCII-empty");
assertTrue(/x(?:...|(...))\1x/i.test("xabcx"), "backref-ASCII-uncaptured");
assertTrue(/x(?:...|(...))\1x/i.test("xabcABCx"), "backref-ASCII-backtrack");
324 325
assertEquals("xaBcAbCABCx,aBc",
             String(/x(...)\1\1x/i.exec("xaBcAbCABCx")),
326
             "backref-ASCII-twice");
327 328

for (var i = 0; i < 128; i++) {
329
  var testName = "backref-ASCII-char-" + i + "," + (i^0x20);
330 331 332 333 334 335 336 337 338
  var test = /^(.)\1$/i.test(String.fromCharCode(i, i ^ 0x20))
  var c = String.fromCharCode(i);
  if (('A' <= c && c <= 'Z') || ('a' <= c && c <= 'z')) {
    assertTrue(test, testName);
  } else {
    assertFalse(test, testName);
  }
}

339
assertFalse(/f(o)$\1/.test('foo'), "backref detects at_end");
erik.corry@gmail.com's avatar
erik.corry@gmail.com committed
340

341
// Check decimal escapes doesn't overflow.
342
// (Note: \214 is interpreted as octal).
343 344 345
assertArrayEquals(["\x8c7483648"],
                  /\2147483648/.exec("\x8c7483648"),
                  "Overflow decimal escape");
346

347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379

// Check numbers in quantifiers doesn't overflow and doesn't throw on
// too large numbers.
assertFalse(/a{111111111111111111111111111111111111111111111}/.test('b'),
            "overlarge1");
assertFalse(/a{999999999999999999999999999999999999999999999}/.test('b'),
            "overlarge2");
assertFalse(/a{1,111111111111111111111111111111111111111111111}/.test('b'),
            "overlarge3");
assertFalse(/a{1,999999999999999999999999999999999999999999999}/.test('b'),
            "overlarge4");
assertFalse(/a{2147483648}/.test('b'),
            "overlarge5");
assertFalse(/a{21474836471}/.test('b'),
            "overlarge6");
assertFalse(/a{1,2147483648}/.test('b'),
            "overlarge7");
assertFalse(/a{1,21474836471}/.test('b'),
            "overlarge8");
assertFalse(/a{2147483648,2147483648}/.test('b'),
            "overlarge9");
assertFalse(/a{21474836471,21474836471}/.test('b'),
            "overlarge10");
assertFalse(/a{2147483647}/.test('b'),
            "overlarge11");
assertFalse(/a{1,2147483647}/.test('b'),
            "overlarge12");
assertTrue(/a{1,2147483647}/.test('a'),
            "overlarge13");
assertFalse(/a{2147483647,2147483647}/.test('a'),
            "overlarge14");


erik.corry@gmail.com's avatar
erik.corry@gmail.com committed
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398
// Check that we don't read past the end of the string.
assertFalse(/f/.test('b'));
assertFalse(/[abc]f/.test('x'));
assertFalse(/[abc]f/.test('xa'));
assertFalse(/[abc]</.test('x'));
assertFalse(/[abc]</.test('xa'));
assertFalse(/f/i.test('b'));
assertFalse(/[abc]f/i.test('x'));
assertFalse(/[abc]f/i.test('xa'));
assertFalse(/[abc]</i.test('x'));
assertFalse(/[abc]</i.test('xa'));
assertFalse(/f[abc]/.test('x'));
assertFalse(/f[abc]/.test('xa'));
assertFalse(/<[abc]/.test('x'));
assertFalse(/<[abc]/.test('xa'));
assertFalse(/f[abc]/i.test('x'));
assertFalse(/f[abc]/i.test('xa'));
assertFalse(/<[abc]/i.test('x'));
assertFalse(/<[abc]/i.test('xa'));
399 400 401 402 403 404 405 406 407 408

// Test that merging of quick test masks gets it right.
assertFalse(/x([0-7]%%x|[0-6]%%y)/.test('x7%%y'), 'qt');
assertFalse(/()x\1(y([0-7]%%%x|[0-6]%%%y)|dkjasldkas)/.test('xy7%%%y'), 'qt2');
assertFalse(/()x\1(y([0-7]%%%x|[0-6]%%%y)|dkjasldkas)/.test('xy%%%y'), 'qt3');
assertFalse(/()x\1y([0-7]%%%x|[0-6]%%%y)/.test('xy7%%%y'), 'qt4');
assertFalse(/()x\1(y([0-7]%%%x|[0-6]%%%y)|dkjasldkas)/.test('xy%%%y'), 'qt5');
assertFalse(/()x\1y([0-7]%%%x|[0-6]%%%y)/.test('xy7%%%y'), 'qt6');
assertFalse(/xy([0-7]%%%x|[0-6]%%%y)/.test('xy7%%%y'), 'qt7');
assertFalse(/x([0-7]%%%x|[0-6]%%%y)/.test('x7%%%y'), 'qt8');
409 410 411 412


// Don't hang on this one.
/[^\xfe-\xff]*/.test("");
413 414 415 416 417 418 419 420 421 422 423 424 425


var long = "a";
for (var i = 0; i < 100000; i++) {
  long = "a?" + long;
}
// Don't crash on this one, but maybe throw an exception.
try {
  RegExp(long).exec("a");
} catch (e) {
  assertTrue(String(e).indexOf("Stack overflow") >= 0, "overflow");
}

426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444

// Test that compile works on modified objects
var re = /re+/;
assertEquals("re+", re.source);
assertFalse(re.global);
assertFalse(re.ignoreCase);
assertFalse(re.multiline);
assertEquals(0, re.lastIndex);

re.compile("ro+", "gim");
assertEquals("ro+", re.source);
assertTrue(re.global);
assertTrue(re.ignoreCase);
assertTrue(re.multiline);
assertEquals(0, re.lastIndex);

re.lastIndex = 42;
re.someOtherProperty = 42;
re.someDeletableProperty = 42;
445 446
re[37] = 37;
re[42] = 42;
447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473

re.compile("ra+", "i");
assertEquals("ra+", re.source);
assertFalse(re.global);
assertTrue(re.ignoreCase);
assertFalse(re.multiline);
assertEquals(0, re.lastIndex);

assertEquals(42, re.someOtherProperty);
assertEquals(42, re.someDeletableProperty);
assertEquals(37, re[37]);
assertEquals(42, re[42]);

re.lastIndex = -1;
re.someOtherProperty = 37;
re[42] = 37;
assertTrue(delete re[37]);
assertTrue(delete re.someDeletableProperty);
re.compile("ri+", "gm");

assertEquals("ri+", re.source);
assertTrue(re.global);
assertFalse(re.ignoreCase);
assertTrue(re.multiline);
assertEquals(0, re.lastIndex);
assertEquals(37, re.someOtherProperty);
assertEquals(37, re[42]);
474 475

// Test boundary-checks.
476
function assertRegExpTest(re, input, test) {
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521
  assertEquals(test, re.test(input), "test:" + re + ":" + input);
}

assertRegExpTest(/b\b/, "b", true);
assertRegExpTest(/b\b$/, "b", true);
assertRegExpTest(/\bb/, "b", true);
assertRegExpTest(/^\bb/, "b", true);
assertRegExpTest(/,\b/, ",", false);
assertRegExpTest(/,\b$/, ",", false);
assertRegExpTest(/\b,/, ",", false);
assertRegExpTest(/^\b,/, ",", false);

assertRegExpTest(/b\B/, "b", false);
assertRegExpTest(/b\B$/, "b", false);
assertRegExpTest(/\Bb/, "b", false);
assertRegExpTest(/^\Bb/, "b", false);
assertRegExpTest(/,\B/, ",", true);
assertRegExpTest(/,\B$/, ",", true);
assertRegExpTest(/\B,/, ",", true);
assertRegExpTest(/^\B,/, ",", true);

assertRegExpTest(/b\b/, "b,", true);
assertRegExpTest(/b\b/, "ba", false);
assertRegExpTest(/b\B/, "b,", false);
assertRegExpTest(/b\B/, "ba", true);

assertRegExpTest(/b\Bb/, "bb", true);
assertRegExpTest(/b\bb/, "bb", false);

assertRegExpTest(/b\b[,b]/, "bb", false);
assertRegExpTest(/b\B[,b]/, "bb", true);
assertRegExpTest(/b\b[,b]/, "b,", true);
assertRegExpTest(/b\B[,b]/, "b,", false);

assertRegExpTest(/[,b]\bb/, "bb", false);
assertRegExpTest(/[,b]\Bb/, "bb", true);
assertRegExpTest(/[,b]\bb/, ",b", true);
assertRegExpTest(/[,b]\Bb/, ",b", false);

assertRegExpTest(/[,b]\b[,b]/, "bb", false);
assertRegExpTest(/[,b]\B[,b]/, "bb", true);
assertRegExpTest(/[,b]\b[,b]/, ",b", true);
assertRegExpTest(/[,b]\B[,b]/, ",b", false);
assertRegExpTest(/[,b]\b[,b]/, "b,", true);
assertRegExpTest(/[,b]\B[,b]/, "b,", false);
522 523 524 525 526 527 528 529 530 531 532 533 534

// Test that caching of result doesn't share result objects.
// More iterations increases the chance of hitting a GC.
for (var i = 0; i < 100; i++) {
  var re = /x(y)z/;
  var res = re.exec("axyzb");
  assertTrue(!!res);
  assertEquals(2, res.length);
  assertEquals("xyz", res[0]);
  assertEquals("y", res[1]);
  assertEquals(1, res.index);
  assertEquals("axyzb", res.input);
  assertEquals(undefined, res.foobar);
535

536 537 538 539
  res.foobar = "Arglebargle";
  res[3] = "Glopglyf";
  assertEquals("Arglebargle", res.foobar);
}
540 541 542 543

// Test that we perform the spec required conversions in the correct order.
var log;
var string = "the string";
544 545
var fakeLastIndex = {
      valueOf: function() {
546 547
        log.push("li");
        return 0;
548
      }
549
    };
550
var fakeString = {
551 552 553
      toString: function() {
        log.push("ts");
        return string;
554 555
      },
      length: 0
556 557 558 559 560 561 562
    };

var re = /str/;
log = [];
re.lastIndex = fakeLastIndex;
var result = re.exec(fakeString);
assertEquals(["str"], result);
563
assertEquals(["ts", "li"], log);
564 565 566 567 568 569

// Again, to check if caching interferes.
log = [];
re.lastIndex = fakeLastIndex;
result = re.exec(fakeString);
assertEquals(["str"], result);
570
assertEquals(["ts", "li"], log);
571 572 573 574 575 576

// And one more time, just to be certain.
log = [];
re.lastIndex = fakeLastIndex;
result = re.exec(fakeString);
assertEquals(["str"], result);
577
assertEquals(["ts", "li"], log);
578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603

// Now with a global regexp, where lastIndex is actually used.
re = /str/g;
log = [];
re.lastIndex = fakeLastIndex;
var result = re.exec(fakeString);
assertEquals(["str"], result);
assertEquals(["ts", "li"], log);

// Again, to check if caching interferes.
log = [];
re.lastIndex = fakeLastIndex;
result = re.exec(fakeString);
assertEquals(["str"], result);
assertEquals(["ts", "li"], log);

// And one more time, just to be certain.
log = [];
re.lastIndex = fakeLastIndex;
result = re.exec(fakeString);
assertEquals(["str"], result);
assertEquals(["ts", "li"], log);


// Check that properties of RegExp have the correct permissions.
var re = /x/g;
604 605 606
var desc = Object.getOwnPropertyDescriptor(re.__proto__, "global");
assertInstanceof(desc.get, Function);
assertEquals(true, desc.configurable);
607
assertEquals(false, desc.enumerable);
608

609 610 611
desc = Object.getOwnPropertyDescriptor(re.__proto__, "multiline");
assertInstanceof(desc.get, Function);
assertEquals(true, desc.configurable);
612
assertEquals(false, desc.enumerable);
613

614 615 616
desc = Object.getOwnPropertyDescriptor(re.__proto__, "ignoreCase");
assertInstanceof(desc.get, Function);
assertEquals(true, desc.configurable);
617
assertEquals(false, desc.enumerable);
618 619 620 621 622 623 624 625 626

desc = Object.getOwnPropertyDescriptor(re, "global");
assertEquals(undefined, desc);

desc = Object.getOwnPropertyDescriptor(re, "multiline");
assertEquals(undefined, desc);

desc = Object.getOwnPropertyDescriptor(re, "ignoreCase");
assertEquals(undefined, desc);
627 628 629 630 631 632

desc = Object.getOwnPropertyDescriptor(re, "lastIndex");
assertEquals(0, desc.value);
assertEquals(false, desc.configurable);
assertEquals(false, desc.enumerable);
assertEquals(true, desc.writable);
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690


// Check that end-anchored regexps are optimized correctly.
var re = /(?:a|bc)g$/;
assertTrue(re.test("ag"));
assertTrue(re.test("bcg"));
assertTrue(re.test("abcg"));
assertTrue(re.test("zimbag"));
assertTrue(re.test("zimbcg"));

assertFalse(re.test("g"));
assertFalse(re.test(""));

// Global regexp (non-zero start).
var re = /(?:a|bc)g$/g;
assertTrue(re.test("ag"));
re.lastIndex = 1;  // Near start of string.
assertTrue(re.test("zimbag"));
re.lastIndex = 6;  // At end of string.
assertFalse(re.test("zimbag"));
re.lastIndex = 5;  // Near end of string.
assertFalse(re.test("zimbag"));
re.lastIndex = 4;
assertTrue(re.test("zimbag"));

// Anchored at both ends.
var re = /^(?:a|bc)g$/g;
assertTrue(re.test("ag"));
re.lastIndex = 1;
assertFalse(re.test("ag"));
re.lastIndex = 1;
assertFalse(re.test("zag"));

// Long max_length of RegExp.
var re = /VeryLongRegExp!{1,1000}$/;
assertTrue(re.test("BahoolaVeryLongRegExp!!!!!!"));
assertFalse(re.test("VeryLongRegExp"));
assertFalse(re.test("!"));

// End anchor inside disjunction.
var re = /(?:a$|bc$)/;
assertTrue(re.test("a"));
assertTrue(re.test("bc"));
assertTrue(re.test("abc"));
assertTrue(re.test("zimzamzumba"));
assertTrue(re.test("zimzamzumbc"));
assertFalse(re.test("c"));
assertFalse(re.test(""));

// Only partially anchored.
var re = /(?:a|bc$)/;
assertTrue(re.test("a"));
assertTrue(re.test("bc"));
assertEquals(["a"], re.exec("abc"));
assertEquals(4, re.exec("zimzamzumba").index);
assertEquals(["bc"], re.exec("zimzomzumbc"));
assertFalse(re.test("c"));
assertFalse(re.test(""));
691

692 693 694 695 696 697 698 699 700 701 702 703 704 705
// Valid syntax in ES5.
re = RegExp("(?:x)*");
re = RegExp("(x)*");

// Syntax extension relative to ES5, for matching JSC (and ES3).
// Shouldn't throw.
re = RegExp("(?=x)*");
re = RegExp("(?!x)*");

// Should throw. Shouldn't hit asserts in debug mode.
assertThrows("RegExp('(*)')");
assertThrows("RegExp('(?:*)')");
assertThrows("RegExp('(?=*)')");
assertThrows("RegExp('(?!*)')");
706 707 708 709

// Test trimmed regular expression for RegExp.test().
assertTrue(/.*abc/.test("abc"));
assertFalse(/.*\d+/.test("q"));
710 711 712 713 714 715 716 717

// Test that RegExp.prototype.toString() throws TypeError for
// incompatible receivers (ES5 section 15.10.6 and 15.10.6.4).
assertThrows("RegExp.prototype.toString.call(null)", TypeError);
assertThrows("RegExp.prototype.toString.call(0)", TypeError);
assertThrows("RegExp.prototype.toString.call('')", TypeError);
assertThrows("RegExp.prototype.toString.call(false)", TypeError);
assertThrows("RegExp.prototype.toString.call(true)", TypeError);
718 719 720 721 722 723 724

// Test mutually recursive capture and backreferences.
assertEquals(["b", "", ""], /(\2)b(\1)/.exec("aba"));
assertEquals(["a", "", ""], /(\2).(\1)/.exec("aba"));
assertEquals(["aba", "a", "a"], /(.\2).(\1)/.exec("aba"));
assertEquals(["acbc", "c", "c"], /a(.\2)b(\1)$/.exec("acbc"));
assertEquals(["acbc", "c", "c"], /a(.\2)b(\1)/.exec("aabcacbc"));
725 726 727 728 729

// Test surrogate pair detection in split.
// \u{daff}\u{e000} is not a surrogate pair, while \u{daff}\u{dfff} is.
assertEquals(["\u{daff}", "\u{e000}"], "\u{daff}\u{e000}".split(/[a-z]{0,1}/u));
assertEquals(["\u{daff}\u{dfff}"], "\u{daff}\u{dfff}".split(/[a-z]{0,1}/u));
730 731 732 733 734 735 736 737

// Test that changing a property on RegExp.prototype results in us taking the
// slow path, which executes RegExp.prototype.exec instead of our
// RegExpExecStub.
const RegExpPrototypeExec = RegExp.prototype.exec;
RegExp.prototype.exec = function() { throw new Error(); }
assertThrows(() => "abc".replace(/./, ""));
RegExp.prototype.exec = RegExpPrototypeExec;
738 739

// Test the code path in RE.proto[@@search] when previousLastIndex is a receiver
740 741 742
// but can't be converted to a primitive. This exposed a crash in an older
// C++ implementation of @@search which a) still relied on Object::Equals,
// and b) incorrectly returned isolate->pending_exception() on error.
743 744 745

var re = /./;
re.lastIndex = { [Symbol.toPrimitive]: 42 };
746
try { "abc".search(re); } catch (_) {}  // Ensure we don't crash.
747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762

// Test lastIndex values of -0.0 and NaN (since @@search uses SameValue).

var re = /./;
re.exec = function(str) { assertEquals(0, re.lastIndex); return []; }
re.lastIndex = -0.0;
assertEquals(-0, re.lastIndex);
"abc".search(re);
assertEquals(-0, re.lastIndex);

var re = /./;
re.exec = function(str) { assertEquals(0, re.lastIndex); return []; }
re.lastIndex = NaN;
assertEquals(NaN, re.lastIndex);
"abc".search(re);
assertEquals(NaN, re.lastIndex);
763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801

// Annex B changes: https://github.com/tc39/ecma262/pull/303

assertThrows("/{1}/", SyntaxError);
assertTrue(/^{*$/.test("{{{"));
assertTrue(/^}*$/.test("}}}"));
assertTrue(/]/.test("]"));
assertTrue(/^\c%$/.test("\\c%"));   // We go into ExtendedPatternCharacter.
assertTrue(/^\d%$/.test("2%"));     // ... CharacterClassEscape.
assertTrue(/^\e%$/.test("e%"));     // ... IdentityEscape.
assertTrue(/^\ca$/.test("\u{1}"));  // ... ControlLetter.
assertTrue(/^\cA$/.test("\u{1}"));  // ... ControlLetter.
assertTrue(/^\c9$/.test("\\c9"));   // ... ExtendedPatternCharacter.
assertTrue(/^\c$/.test("\\c"));   // ... ExtendedPatternCharacter.
assertTrue(/^[\c%]*$/.test("\\c%"));  // TODO(v8:6201): Not covered by the spec.
assertTrue(/^[\c:]*$/.test("\\c:"));  // TODO(v8:6201): Not covered by the spec.
assertTrue(/^[\c0]*$/.test("\u{10}"));  // ... ClassControlLetter.
assertTrue(/^[\c1]*$/.test("\u{11}"));  // ('0' % 32 == 0x10)
assertTrue(/^[\c2]*$/.test("\u{12}"));
assertTrue(/^[\c3]*$/.test("\u{13}"));
assertTrue(/^[\c4]*$/.test("\u{14}"));
assertTrue(/^[\c5]*$/.test("\u{15}"));
assertTrue(/^[\c6]*$/.test("\u{16}"));
assertTrue(/^[\c7]*$/.test("\u{17}"));
assertTrue(/^[\c8]*$/.test("\u{18}"));
assertTrue(/^[\c9]*$/.test("\u{19}"));
assertTrue(/^[\c_]*$/.test("\u{1F}"));
assertTrue(/^[\c11]*$/.test("\u{11}1"));
assertTrue(/^[\8]*$/.test("8"));  // ... ClassEscape ~~> IdentityEscape.
assertTrue(/^[\7]*$/.test("\u{7}"));  // ... ClassEscape
                                      // ~~> LegacyOctalEscapeSequence.
assertTrue(/^[\11]*$/.test("\u{9}"));
assertTrue(/^[\111]*$/.test("\u{49}"));
assertTrue(/^[\222]*$/.test("\u{92}"));
assertTrue(/^[\333]*$/.test("\u{DB}"));
assertTrue(/^[\444]*$/.test("\u{24}4"));
assertTrue(/^[\d-X]*$/.test("234-X-432"));  // CharacterRangeOrUnion.
assertTrue(/^[\d-X-Z]*$/.test("234-XZ-432"));
assertFalse(/^[\d-X-Z]*$/.test("234-XYZ-432"));
802 803 804 805 806

// Lone leading surrogates. Just here to exercise specific parsing code-paths.

assertFalse(/\uDB88|\uDBEC|aa/.test(""));
assertFalse(/\uDB88|\uDBEC|aa/u.test(""));
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822

// EscapeRegExpPattern
assertEquals("\\n", /\n/.source);
assertEquals("\\n", new RegExp("\n").source);
assertEquals("\\n", new RegExp("\\n").source);
assertEquals("\\\\n", /\\n/.source);
assertEquals("\\r", /\r/.source);
assertEquals("\\r", new RegExp("\r").source);
assertEquals("\\r", new RegExp("\\r").source);
assertEquals("\\\\r", /\\r/.source);
assertEquals("\\u2028", /\u2028/.source);
assertEquals("\\u2028", new RegExp("\u2028").source);
assertEquals("\\u2028", new RegExp("\\u2028").source);
assertEquals("\\u2029", /\u2029/.source);
assertEquals("\\u2029", new RegExp("\u2029").source);
assertEquals("\\u2029", new RegExp("\\u2029").source);
823 824 825 826 827 828 829 830 831 832 833
assertEquals("[/]", /[/]/.source);
assertEquals("[\\/]", /[\/]/.source);
assertEquals("[\\\\/]", /[\\/]/.source);
assertEquals("[/]", new RegExp("[/]").source);
assertEquals("[/]", new RegExp("[\/]").source);
assertEquals("[\\/]", new RegExp("[\\/]").source);
assertEquals("[[/]", /[[/]/.source);
assertEquals("[/]]", /[/]]/.source);
assertEquals("[[/]]", /[[/]]/.source);
assertEquals("[[\\/]", /[[\/]/.source);
assertEquals("[[\\/]]", /[[\/]]/.source);
834 835 836 837 838 839 840 841 842 843
assertEquals("\\n", new RegExp("\\\n").source);
assertEquals("\\r", new RegExp("\\\r").source);
assertEquals("\\u2028", new RegExp("\\\u2028").source);
assertEquals("\\u2029", new RegExp("\\\u2029").source);

{
  // No escapes needed, the original string should be reused as `.source`.
  const pattern = "\\n";
  assertTrue(%ReferenceEqual(pattern, new RegExp(pattern).source));
}