Port and integrate the irregexp engine from V8

tests/language/regex/regexp_test.dart

+// Copyright (c) 2014, the Dart project authors.  Please see the AUTHORS file
+// for details. All rights reserved. Use of this source code is governed by a
+// BSD-style license that can be found in the LICENSE file.
+
+import "package:expect/expect.dart";
+
+void testEscape(str, regex) {
+  assertEquals("foo:bar:baz", str.split(regex).join(":"));
+}
+
+void assertEquals(actual, expected, [message]) => 
+    Expect.equals(actual, expected, message);
+void assertTrue(actual, [message]) => Expect.isTrue(actual, message);
+void assertFalse(actual, [message]) => Expect.isFalse(actual, message);
+void assertThrows(fn) => Expect.throws(fn);
+
+void main() {
+  // Ensure we hit the optimizer.
+  for (int i = 0; i < 50; i++) {
+    testEscape("foo\nbar\nbaz", new RegExp(r"\n"));
+    testEscape("foo bar baz", new RegExp(r"\s"));
+    testEscape("foo\tbar\tbaz", new RegExp(r"\s"));
+    testEscape("foo-bar-baz", new RegExp(r"\u002D"));
+
+    // Test containing null char in regexp.
+    var s = '[' + new String.fromCharCode(0) + ']';
+    var re = new RegExp(s);
+    assertEquals(re.allMatches(s).length, 1);
+    assertEquals(re.stringMatch(s), new String.fromCharCode(0));
+
+    
+    final _vmFrame = new RegExp(
+      r'^#\d+\s+(\S.*) \((.+?):(\d+)(?::(\d+))?\)$');
+    final _traceLine =
+      "#0      Trace.Trace.parse (package:stack_trace/src/trace.dart:130:7)";
+    Expect.equals(_vmFrame.firstMatch(_traceLine).group(0), _traceLine);
+
+    // Test the UTF16 case insensitive comparison.
+    re = new RegExp(r"x(a)\1x", caseSensitive: false);
+    Expect.equals(re.firstMatch("xaAx\u1234").group(0), "xaAx");
+
+    // Test strings containing all line separators
+    s = 'aA\nbB\rcC\r\ndD\u2028eE\u2029fF';
+    // any non-newline character at the beginning of a line
+    re = new RegExp(r"^.", multiLine: true);
+    var result = re.allMatches(s).toList();
+    assertEquals(result.length, 6);
+    assertEquals(result[0][0], 'a');
+    assertEquals(result[1][0], 'b');
+    assertEquals(result[2][0], 'c');
+    assertEquals(result[3][0], 'd');
+    assertEquals(result[4][0], 'e');
+    assertEquals(result[5][0], 'f');
+
+    // any non-newline character at the end of a line
+    re = new RegExp(r".$", multiLine: true);
+    result = re.allMatches(s).toList();
+    assertEquals(result.length, 6);
+    assertEquals(result[0][0], 'A');
+    assertEquals(result[1][0], 'B');
+    assertEquals(result[2][0], 'C');
+    assertEquals(result[3][0], 'D');
+    assertEquals(result[4][0], 'E');
+    assertEquals(result[5][0], 'F');
+
+    // *any* character at the beginning of a line
+    re = new RegExp(r"^[^]", multiLine: true);
+    result = re.allMatches(s).toList();
+    assertEquals(result.length, 7);
+    assertEquals(result[0][0], 'a');
+    assertEquals(result[1][0], 'b');
+    assertEquals(result[2][0], 'c');
+    assertEquals(result[3][0], '\n');
+    assertEquals(result[4][0], 'd');
+    assertEquals(result[5][0], 'e');
+    assertEquals(result[6][0], 'f');
+
+    // *any* character at the end of a line
+    re = new RegExp(r"[^]$", multiLine: true);
+    result = re.allMatches(s).toList();
+    assertEquals(result.length, 7);
+    assertEquals(result[0][0], 'A');
+    assertEquals(result[1][0], 'B');
+    assertEquals(result[2][0], 'C');
+    assertEquals(result[3][0], '\r');
+    assertEquals(result[4][0], 'D');
+    assertEquals(result[5][0], 'E');
+    assertEquals(result[6][0], 'F');
+
+    // Some tests from the Mozilla tests, where our behavior used to differ from
+    // SpiderMonkey.
+    // From ecma_3/RegExp/regress-334158.js
+    assertTrue( "\x01" .contains(new RegExp(r"\ca")));
+    assertFalse( "\\ca" .contains(new RegExp(r"\ca")));
+    assertFalse( "ca" .contains(new RegExp(r"\ca")));
+    assertTrue( "\\ca" .contains(new RegExp(r"\c[a/]")));
+    assertTrue( "\\c/" .contains(new RegExp(r"\c[a/]")));
+
+    // Test \c in character class
+    re = r"^[\cM]$";
+    assertTrue("\r".contains(new RegExp(re)));
+    assertFalse("M".contains(new RegExp(re)));
+    assertFalse("c".contains(new RegExp(re)));
+    assertFalse("\\".contains(new RegExp(re)));
+    assertFalse("\x03".contains(new RegExp(re)));  // I.e., read as \cc
+
+    re = r"^[\c]]$";
+    assertTrue("c]".contains(new RegExp(re)));
+    assertTrue("\\]".contains(new RegExp(re)));
+    assertFalse("\x1d".contains(new RegExp(re)));  // ']' & 0x1f
+    assertFalse("\x03]".contains(new RegExp(re)));  // I.e., read as \cc
+
+    re = r"^[\c1]$";  // Digit control characters are masked in character classes.
+    assertTrue("\x11".contains(new RegExp(re)));
+    assertFalse("\\".contains(new RegExp(re)));
+    assertFalse("c".contains(new RegExp(re)));
+    assertFalse("1".contains(new RegExp(re)));
+
+    // Underscore control character is masked in character classes.
+    re = r"^[\c_]$";
+    assertTrue("\x1f".contains(new RegExp(re)));
+    assertFalse("\\".contains(new RegExp(re)));
+    assertFalse("c".contains(new RegExp(re)));
+    assertFalse("_".contains(new RegExp(re)));
+
+    re = r"^[\c$]$";  // Other characters are interpreted literally.
+    assertFalse("\x04".contains(new RegExp(re)));
+    assertTrue("\\".contains(new RegExp(re)));
+    assertTrue("c".contains(new RegExp(re)));
+    assertTrue(r"$".contains(new RegExp(re)));
+
+    assertTrue("Z[\\cde".contains(new RegExp(r"^[Z-\c-e]*$")));
+
+    // Test that we handle \s and \S correctly on special Unicode characters.
+    re = r"\s";
+    assertTrue("\u2028".contains(new RegExp(re)));
+    assertTrue("\u2029".contains(new RegExp(re)));
+    assertTrue("\uFEFF".contains(new RegExp(re)));
+
+    re = r"\S";
+    assertFalse("\u2028".contains(new RegExp(re)));
+    assertFalse("\u2029".contains(new RegExp(re)));
+    assertFalse("\uFEFF".contains(new RegExp(re)));
+
+    // Test that we handle \s and \S correctly inside some bizarre
+    // character classes.
+    re = r"[\s-:]";
+    assertTrue('-'.contains(new RegExp(re)));
+    assertTrue(':'.contains(new RegExp(re)));
+    assertTrue(' '.contains(new RegExp(re)));
+    assertTrue('\t'.contains(new RegExp(re)));
+    assertTrue('\n'.contains(new RegExp(re)));
+    assertFalse('a'.contains(new RegExp(re)));
+    assertFalse('Z'.contains(new RegExp(re)));
+
+    re = r"[\S-:]";
+    assertTrue('-'.contains(new RegExp(re)));
+    assertTrue(':'.contains(new RegExp(re)));
+    assertFalse(' '.contains(new RegExp(re)));
+    assertFalse('\t'.contains(new RegExp(re)));
+    assertFalse('\n'.contains(new RegExp(re)));
+    assertTrue('a'.contains(new RegExp(re)));
+    assertTrue('Z'.contains(new RegExp(re)));
+
+    re = r"[^\s-:]";
+    assertFalse('-'.contains(new RegExp(re)));
+    assertFalse(':'.contains(new RegExp(re)));
+    assertFalse(' '.contains(new RegExp(re)));
+    assertFalse('\t'.contains(new RegExp(re)));
+    assertFalse('\n'.contains(new RegExp(re)));
+    assertTrue('a'.contains(new RegExp(re)));
+    assertTrue('Z'.contains(new RegExp(re)));
+
+    re = r"[^\S-:]";
+    assertFalse('-'.contains(new RegExp(re)));
+    assertFalse(':'.contains(new RegExp(re)));
+    assertTrue(' '.contains(new RegExp(re)));
+    assertTrue('\t'.contains(new RegExp(re)));
+    assertTrue('\n'.contains(new RegExp(re)));
+    assertFalse('a'.contains(new RegExp(re)));
+    assertFalse('Z'.contains(new RegExp(re)));
+
+    re = r"[\s]";
+    assertFalse('-'.contains(new RegExp(re)));
+    assertFalse(':'.contains(new RegExp(re)));
+    assertTrue(' '.contains(new RegExp(re)));
+    assertTrue('\t'.contains(new RegExp(re)));
+    assertTrue('\n'.contains(new RegExp(re)));
+    assertFalse('a'.contains(new RegExp(re)));
+    assertFalse('Z'.contains(new RegExp(re)));
+
+    re = r"[^\s]";
+    assertTrue('-'.contains(new RegExp(re)));
+    assertTrue(':'.contains(new RegExp(re)));
+    assertFalse(' '.contains(new RegExp(re)));
+    assertFalse('\t'.contains(new RegExp(re)));
+    assertFalse('\n'.contains(new RegExp(re)));
+    assertTrue('a'.contains(new RegExp(re)));
+    assertTrue('Z'.contains(new RegExp(re)));
+
+    re = r"[\S]";
+    assertTrue('-'.contains(new RegExp(re)));
+    assertTrue(':'.contains(new RegExp(re)));
+    assertFalse(' '.contains(new RegExp(re)));
+    assertFalse('\t'.contains(new RegExp(re)));
+    assertFalse('\n'.contains(new RegExp(re)));
+    assertTrue('a'.contains(new RegExp(re)));
+    assertTrue('Z'.contains(new RegExp(re)));
+
+    re = r"[^\S]";
+    assertFalse('-'.contains(new RegExp(re)));
+    assertFalse(':'.contains(new RegExp(re)));
+    assertTrue(' '.contains(new RegExp(re)));
+    assertTrue('\t'.contains(new RegExp(re)));
+    assertTrue('\n'.contains(new RegExp(re)));
+    assertFalse('a'.contains(new RegExp(re)));
+    assertFalse('Z'.contains(new RegExp(re)));
+
+    re = r"[\s\S]";
+    assertTrue('-'.contains(new RegExp(re)));
+    assertTrue(':'.contains(new RegExp(re)));
+    assertTrue(' '.contains(new RegExp(re)));
+    assertTrue('\t'.contains(new RegExp(re)));
+    assertTrue('\n'.contains(new RegExp(re)));
+    assertTrue('a'.contains(new RegExp(re)));
+    assertTrue('Z'.contains(new RegExp(re)));
+
+    re = r"[^\s\S]";
+    assertFalse('-'.contains(new RegExp(re)));
+    assertFalse(':'.contains(new RegExp(re)));
+    assertFalse(' '.contains(new RegExp(re)));
+    assertFalse('\t'.contains(new RegExp(re)));
+    assertFalse('\n'.contains(new RegExp(re)));
+    assertFalse('a'.contains(new RegExp(re)));
+    assertFalse('Z'.contains(new RegExp(re)));
+
+    // 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 = r"[\s-0-9]";
+    assertTrue(' '.contains(new RegExp(re)));
+    assertTrue('\xA0'.contains(new RegExp(re)));
+    assertTrue('-'.contains(new RegExp(re)));
+    assertTrue('0'.contains(new RegExp(re)));
+    assertTrue('9'.contains(new RegExp(re)));
+    assertFalse('1'.contains(new RegExp(re)));
+
+    // Test beginning and end of line assertions with or without the
+    // multiline flag.
+    re = r"^\d+";
+    assertFalse("asdf\n123".contains(new RegExp(re)));
+    re = new RegExp(r"^\d+", multiLine: true);
+    assertTrue("asdf\n123".contains(re));
+
+    re = r"\d+$";
+    assertFalse("123\nasdf".contains(new RegExp(re)));
+    re = new RegExp(r"\d+$", multiLine: true);
+    assertTrue("123\nasdf".contains(re));
+
+    // Test that empty matches are handled correctly for multiline global
+    // regexps.
+    re = new RegExp(r"^(.*)", multiLine: true);
+    assertEquals(3, re.allMatches("a\n\rb").length);
+    assertEquals("*a\n*b\r*c\n*\r*d\r*\n*e",
+        "a\nb\rc\n\rd\r\ne".replaceAllMapped(re, (Match m) => "*${m.group(1)}"));
+
+    // Test that empty matches advance one character
+    re = new RegExp("");
+    assertEquals("xAx", "A".replaceAll(re, "x"));
+    assertEquals(3, new String.fromCharCode(161).replaceAll(re, "x").length);
+
+    // Check for lazy RegExp literal creation
+    void lazyLiteral(doit) {
+      if (doit) return "".replaceAll(
+          new RegExp(r"foo(", caseSensitive: false), "");
+      return true;
+    }
+
+    assertTrue(lazyLiteral(false));
+    assertThrows(() => lazyLiteral(true));
+
+    // Check $01 and $10
+    re = new RegExp("(.)(.)(.)(.)(.)(.)(.)(.)(.)(.)");
+    assertEquals("t", "123456789t".replaceAllMapped(
+          re, (Match m) => m.group(10)));
+    assertEquals("15", "123456789t".replaceAllMapped(
+          re, (Match m) => "${m.group(1)}5"));
+    assertEquals("1", "123456789t".replaceAllMapped(
+          re, (Match m) => m.group(1)));
+
+    assertFalse("football".contains(new RegExp(r"()foo$\1")), "football1");
+    assertFalse("football".contains(new RegExp(r"foo$(?=ball)")), "football2");
+    assertFalse("football".contains(new RegExp(r"foo$(?!bar)")), "football3");
+    assertTrue("foo".contains(new RegExp(r"()foo$\1")), "football4");
+    assertTrue("foo".contains(new RegExp(r"foo$(?=(ball)?)")), "football5");
+    assertTrue("foo".contains(new RegExp(r"()foo$(?!bar)")), "football6");
+    assertFalse("football".contains(new RegExp(r"(x?)foo$\1")), "football7");
+    assertFalse("football".contains(new RegExp(r"foo$(?=ball)")), "football8");
+    assertFalse("football".contains(new RegExp(r"foo$(?!bar)")), "football9");
+    assertTrue("foo".contains(new RegExp(r"(x?)foo$\1")), "football10");
+    assertTrue("foo".contains(new RegExp(r"foo$(?=(ball)?)")), "football11");
+    assertTrue("foo".contains(new RegExp(r"foo$(?!bar)")), "football12");
+
+    // Check that the back reference has two successors.  See
+    // BackReferenceNode::PropagateForward.
+    assertFalse('foo'.contains(new RegExp(r"f(o)\b\1")));
+    assertTrue('foo'.contains(new RegExp(r"f(o)\B\1")));
+
+    // Back-reference, ignore case:
+    // ASCII
+    assertEquals("a", new RegExp(r"x(a)\1x", caseSensitive: false)
+        .firstMatch("xaAx").group(1), "backref-ASCII");
+    assertFalse("xaaaaa".contains(new RegExp(r"x(...)\1", caseSensitive: false)),
+        "backref-ASCII-short");
+    assertTrue("xx".contains(new RegExp(r"x((?:))\1\1x", caseSensitive: false)),
+        "backref-ASCII-empty");
+    assertTrue("xabcx".contains(
+          new RegExp(r"x(?:...|(...))\1x", caseSensitive: false)),
+        "backref-ASCII-uncaptured");
+    assertTrue("xabcABCx".contains(
+          new RegExp(r"x(?:...|(...))\1x", caseSensitive: false)),
+        "backref-ASCII-backtrack");
+    assertEquals("aBc", new RegExp(r"x(...)\1\1x", caseSensitive: false)
+        .firstMatch("xaBcAbCABCx").group(1), "backref-ASCII-twice");
+
+    for (var i = 0; i < 128; i++) {
+      var testName = "backref-ASCII-char-$i,,${i^0x20}";
+      var test = new String.fromCharCodes([i, i ^ 0x20])
+        .contains(new RegExp(r"^(.)\1$", caseSensitive: false));
+      if (('A'.codeUnitAt(0) <= i && i <= 'Z'.codeUnitAt(0)) ||
+          ('a'.codeUnitAt(0) <= i && i <= 'z'.codeUnitAt(0))) {
+        assertTrue(test, testName);
+      } else {
+        assertFalse(test, testName);
+      }
+    }
+
+    assertFalse('foo'.contains(new RegExp(r"f(o)$\1")), "backref detects at_end");
+
+    // Check decimal escapes doesn't overflow.
+    // (Note: \214 is interpreted as octal).
+    assertEquals("\x8c7483648", new RegExp(r"\2147483648")
+        .firstMatch("\x8c7483648").group(0), "Overflow decimal escape");
+
+    // Check numbers in quantifiers doesn't overflow and doesn't throw on
+    // too large numbers.
+    assertFalse('b'.contains(
+          new RegExp(r"a{111111111111111111111111111111111111111111111}")),
+                "overlarge1");
+    assertFalse('b'.contains(
+          new RegExp(r"a{999999999999999999999999999999999999999999999}")),
+                "overlarge2");
+    assertFalse('b'.contains(
+          new RegExp(r"a{1,111111111111111111111111111111111111111111111}")),
+                "overlarge3");
+    assertFalse('b'.contains(
+          new RegExp(r"a{1,999999999999999999999999999999999999999999999}")),
+                "overlarge4");
+    assertFalse('b'.contains(new RegExp(r"a{2147483648}")),
+                "overlarge5");
+    assertFalse('b'.contains(new RegExp(r"a{21474836471}")),
+                "overlarge6");
+    assertFalse('b'.contains(new RegExp(r"a{1,2147483648}")),
+                "overlarge7");
+    assertFalse('b'.contains(new RegExp(r"a{1,21474836471}")),
+                "overlarge8");
+    assertFalse('b'.contains(new RegExp(r"a{2147483648,2147483648}")),
+                "overlarge9");
+    assertFalse('b'.contains(new RegExp(r"a{21474836471,21474836471}")),
+                "overlarge10");
+    assertFalse('b'.contains(new RegExp(r"a{2147483647}")),
+                "overlarge11");
+    assertFalse('b'.contains(new RegExp(r"a{1,2147483647}")),
+                "overlarge12");
+    assertTrue('a'.contains(new RegExp(r"a{1,2147483647}")),
+                "overlarge13");
+    assertFalse('a'.contains(new RegExp(r"a{2147483647,2147483647}")),
+                "overlarge14");
+
+    // Check that we don't read past the end of the string.
+    assertFalse('b'.contains(new RegExp(r"f")));
+    assertFalse('x'.contains(new RegExp(r"[abc]f")));
+    assertFalse('xa'.contains(new RegExp(r"[abc]f")));
+    assertFalse('x'.contains(new RegExp(r"[abc]<")));
+    assertFalse('xa'.contains(new RegExp(r"[abc]<")));
+    assertFalse('b'.contains(new RegExp(r"f", caseSensitive: false)));
+    assertFalse('x'.contains(new RegExp(r"[abc]f", caseSensitive: false)));
+    assertFalse('xa'.contains(new RegExp(r"[abc]f", caseSensitive: false)));
+    assertFalse('x'.contains(new RegExp(r"[abc]<", caseSensitive: false)));
+    assertFalse('xa'.contains(new RegExp(r"[abc]<", caseSensitive: false)));
+    assertFalse('x'.contains(new RegExp(r"f[abc]")));
+    assertFalse('xa'.contains(new RegExp(r"f[abc]")));
+    assertFalse('x'.contains(new RegExp(r"<[abc]")));
+    assertFalse('xa'.contains(new RegExp(r"<[abc]")));
+    assertFalse('x'.contains(new RegExp(r"f[abc]", caseSensitive: false)));
+    assertFalse('xa'.contains(new RegExp(r"f[abc]", caseSensitive: false)));
+    assertFalse('x'.contains(new RegExp(r"<[abc]", caseSensitive: false)));
+    assertFalse('xa'.contains(new RegExp(r"<[abc]", caseSensitive: false)));
+
+    // Test that merging of quick test masks gets it right.
+    assertFalse('x7%%y'.contains(new RegExp(r"x([0-7]%%x|[0-6]%%y)")), 'qt');
+    assertFalse('xy7%%%y'.contains(
+          new RegExp(r"()x\1(y([0-7]%%%x|[0-6]%%%y)|dkjasldkas)")), 'qt2');
+    assertFalse('xy%%%y'.contains(
+          new RegExp(r"()x\1(y([0-7]%%%x|[0-6]%%%y)|dkjasldkas)")), 'qt3');
+    assertFalse('xy7%%%y'.contains(
+          new RegExp(r"()x\1y([0-7]%%%x|[0-6]%%%y)")), 'qt4');
+    assertFalse('xy%%%y'.contains(
+          new RegExp(r"()x\1(y([0-7]%%%x|[0-6]%%%y)|dkjasldkas)")), 'qt5');
+    assertFalse('xy7%%%y'.contains(
+          new RegExp(r"()x\1y([0-7]%%%x|[0-6]%%%y)")), 'qt6');
+    assertFalse('xy7%%%y'.contains(
+          new RegExp(r"xy([0-7]%%%x|[0-6]%%%y)")), 'qt7');
+    assertFalse('x7%%%y'.contains(new RegExp(r"x([0-7]%%%x|[0-6]%%%y)")), 'qt8');
+
+
+    // Don't hang on this one.
+    "".contains(new RegExp(r"[^\xfe-\xff]*"));
+
+    var longbuffer = new StringBuffer("a");
+    for (var i = 0; i < 100000; i++) {
+      longbuffer.write("a?");
+    }
+    var long = longbuffer.toString();
+
+    // Don't crash on this one, but maybe throw an exception.
+    try {
+      new RegExp(long).allMatches("a");
+    } catch (e) {
+      assertTrue(String(e).indexOf("Stack overflow") >= 0, "overflow");
+    }
+
+
+    // Test boundary-checks.
+    void assertRegExpTest(re, input, test) {
+      assertEquals(test, input.contains(
+            new RegExp(re)), "test:" + re + ":" + input);
+    }
+
+    assertRegExpTest(r"b\b", "b", true);
+    assertRegExpTest(r"b\b$", "b", true);
+    assertRegExpTest(r"\bb", "b", true);
+    assertRegExpTest(r"^\bb", "b", true);
+    assertRegExpTest(r",\b", ",", false);
+    assertRegExpTest(r",\b$", ",", false);
+    assertRegExpTest(r"\b,", ",", false);
+    assertRegExpTest(r"^\b,", ",", false);
+
+    assertRegExpTest(r"b\B", "b", false);
+    assertRegExpTest(r"b\B$", "b", false);
+    assertRegExpTest(r"\Bb", "b", false);
+    assertRegExpTest(r"^\Bb", "b", false);
+    assertRegExpTest(r",\B", ",", true);
+    assertRegExpTest(r",\B$", ",", true);
+    assertRegExpTest(r"\B,", ",", true);
+    assertRegExpTest(r"^\B,", ",", true);
+
+    assertRegExpTest(r"b\b", "b,", true);
+    assertRegExpTest(r"b\b", "ba", false);
+    assertRegExpTest(r"b\B", "b,", false);
+    assertRegExpTest(r"b\B", "ba", true);
+
+    assertRegExpTest(r"b\Bb", "bb", true);
+    assertRegExpTest(r"b\bb", "bb", false);
+
+    assertRegExpTest(r"b\b[,b]", "bb", false);
+    assertRegExpTest(r"b\B[,b]", "bb", true);
+    assertRegExpTest(r"b\b[,b]", "b,", true);
+    assertRegExpTest(r"b\B[,b]", "b,", false);
+
+    assertRegExpTest(r"[,b]\bb", "bb", false);
+    assertRegExpTest(r"[,b]\Bb", "bb", true);
+    assertRegExpTest(r"[,b]\bb", ",b", true);
+    assertRegExpTest(r"[,b]\Bb", ",b", false);
+
+    assertRegExpTest(r"[,b]\b[,b]", "bb", false);
+    assertRegExpTest(r"[,b]\B[,b]", "bb", true);
+    assertRegExpTest(r"[,b]\b[,b]", ",b", true);
+    assertRegExpTest(r"[,b]\B[,b]", ",b", false);
+    assertRegExpTest(r"[,b]\b[,b]", "b,", true);
+    assertRegExpTest(r"[,b]\B[,b]", "b,", false);
+
+    // Skipped tests from V8:
+
+    // Test that caching of result doesn't share result objects.
+    // More iterations increases the chance of hitting a GC.
+    
+    // Test that we perform the spec required conversions in the correct order.
+
+    // Check that properties of RegExp have the correct permissions.
+
+    // Check that end-anchored regexps are optimized correctly.
+    re = r"(?:a|bc)g$";
+    assertTrue("ag".contains(new RegExp(re)));
+    assertTrue("bcg".contains(new RegExp(re)));
+    assertTrue("abcg".contains(new RegExp(re)));
+    assertTrue("zimbag".contains(new RegExp(re)));
+    assertTrue("zimbcg".contains(new RegExp(re)));
+
+    assertFalse("g".contains(new RegExp(re)));
+    assertFalse("".contains(new RegExp(re)));
+
+    // Global regexp (non-zero start).
+    re = r"(?:a|bc)g$";
+    assertTrue("ag".contains(new RegExp(re)));
+    // Near start of string.
+    assertTrue(new RegExp(re).allMatches("zimbag", 1).isNotEmpty);
+    // At end of string.
+    assertTrue(new RegExp(re).allMatches("zimbag", 6).isEmpty);
+    5;  // Near end of string.
+    assertTrue(new RegExp(re).allMatches("zimbag", 5).isEmpty);
+    assertTrue(new RegExp(re).allMatches("zimbag", 4).isNotEmpty);
+
+    // Anchored at both ends.
+    re = r"^(?:a|bc)g$";
+    assertTrue("ag".contains(new RegExp(re)));
+    assertTrue(new RegExp(re).allMatches("ag", 1).isEmpty);
+    assertTrue(new RegExp(re).allMatches("zag", 1).isEmpty);
+
+    // Long max_length of RegExp.
+    re = r"VeryLongRegExp!{1,1000}$";
+    assertTrue("BahoolaVeryLongRegExp!!!!!!".contains(new RegExp(re)));
+    assertFalse("VeryLongRegExp".contains(new RegExp(re)));
+    assertFalse("!".contains(new RegExp(re)));
+
+    // End anchor inside disjunction.
+    re = r"(?:a$|bc$)";
+    assertTrue("a".contains(new RegExp(re)));
+    assertTrue("bc".contains(new RegExp(re)));
+    assertTrue("abc".contains(new RegExp(re)));
+    assertTrue("zimzamzumba".contains(new RegExp(re)));
+    assertTrue("zimzamzumbc".contains(new RegExp(re)));
+    assertFalse("c".contains(new RegExp(re)));
+    assertFalse("".contains(new RegExp(re)));
+
+    // Only partially anchored.
+    re = r"(?:a|bc$)";
+    assertTrue("a".contains(new RegExp(re)));
+    assertTrue("bc".contains(new RegExp(re)));
+    assertEquals("a", new RegExp(re).firstMatch("abc").group(0));
+    assertEquals(4, new RegExp(re).firstMatch("zimzamzumba").start);
+    assertEquals("bc", new RegExp(re).firstMatch("zimzomzumbc").group(0));
+    assertFalse("c".contains(new RegExp(re)));
+    assertFalse("".contains(new RegExp(re)));
+
+    // Valid syntax in ES5.
+    re = new RegExp("(?:x)*");
+    re = new RegExp("(x)*");
+
+    // Syntax extension relative to ES5, for matching JSC (and ES3).
+    // Shouldn't throw.
+    re = new RegExp("(?=x)*");
+    re = new RegExp("(?!x)*");
+
+    // Should throw. Shouldn't hit asserts in debug mode.
+    assertThrows(() => new RegExp('(*)'));
+    assertThrows(() => new RegExp('(?:*)'));
+    assertThrows(() => new RegExp('(?=*)'));
+    assertThrows(() => new RegExp('(?!*)'));
+
+    // Test trimmed regular expression for RegExp.test().
+    assertTrue("abc".contains(new RegExp(r".*abc")));
+    assertFalse("q".contains(new RegExp(r".*\d+")));
+
+    // Tests skipped from V8:
+    // Test that RegExp.prototype.toString() throws TypeError for
+    // incompatible receivers (ES5 section 15.10.6 and 15.10.6.4).
+  }
+}