Generate irregexp character tables with make_unicode.py.

4 files changed, 271 insertions, 109 deletions

diff --git a/js/src/irregexp/RegExpCharacters-inl.h b/js/src/irregexp/RegExpCharacters-inl.h
new file mode 100644
index 0000000000..d001819fcb
--- /dev/null
+++ b/js/src/irregexp/RegExpCharacters-inl.h
@@ -0,0 +1,40 @@
+/* Generated by make_unicode.py DO NOT MODIFY */
+/* Unicode version: 9.0.0 */
+#ifndef V8_JSREGEXPCHARACTERS_INL_H_
+#define V8_JSREGEXPCHARACTERS_INL_H_
+
+namespace js {
+
+namespace irregexp {
+
+static inline bool
+RangeContainsLatin1Equivalents(CharacterRange range, bool unicode)
+{
+    if (unicode) {
+        // "LATIN SMALL LETTER LONG S" case folds to "LATIN SMALL LETTER S".
+        if (range.Contains(0x017F))
+            return true;
+        // "LATIN CAPITAL LETTER SHARP S" case folds to "LATIN SMALL LETTER SHARP S".
+        if (range.Contains(0x1E9E))
+            return true;
+        // "KELVIN SIGN" case folds to "LATIN SMALL LETTER K".
+        if (range.Contains(0x212A))
+            return true;
+        // "ANGSTROM SIGN" case folds to "LATIN SMALL LETTER A WITH RING ABOVE".
+        if (range.Contains(0x212B))
+            return true;
+    }
+
+    // "GREEK CAPITAL LETTER MU" case maps to "MICRO SIGN".
+    // "GREEK SMALL LETTER MU" case maps to "MICRO SIGN".
+    if (range.Contains(0x039C) || range.Contains(0x03BC))
+        return true;
+    // "LATIN CAPITAL LETTER Y WITH DIAERESIS" case maps to "LATIN SMALL LETTER Y WITH DIAERESIS".
+    if (range.Contains(0x0178))
+        return true;
+    return false;
+}
+
+} } // namespace js::irregexp
+
+#endif // V8_JSREGEXPCHARACTERS_INL_H_
diff --git a/js/src/irregexp/RegExpCharacters.cpp b/js/src/irregexp/RegExpCharacters.cpp
new file mode 100644
index 0000000000..096c027605
--- /dev/null
+++ b/js/src/irregexp/RegExpCharacters.cpp
@@ -0,0 +1,135 @@
+/* Generated by make_unicode.py DO NOT MODIFY */
+/* Unicode version: 9.0.0 */
+#include "irregexp/RegExpCharacters.h"
+
+#include "mozilla/Assertions.h"
+
+char16_t
+js::irregexp::ConvertNonLatin1ToLatin1(char16_t c, bool unicode)
+{
+    MOZ_ASSERT(c > 0xFF, "Character mustn't be Latin1");
+    if (unicode) {
+        // "LATIN SMALL LETTER LONG S" case folds to "LATIN SMALL LETTER S".
+        if (c == 0x017F)
+            return 0x73;
+        // "LATIN CAPITAL LETTER SHARP S" case folds to "LATIN SMALL LETTER SHARP S".
+        if (c == 0x1E9E)
+            return 0xDF;
+        // "KELVIN SIGN" case folds to "LATIN SMALL LETTER K".
+        if (c == 0x212A)
+            return 0x6B;
+        // "ANGSTROM SIGN" case folds to "LATIN SMALL LETTER A WITH RING ABOVE".
+        if (c == 0x212B)
+            return 0xE5;
+    }
+
+    // "GREEK CAPITAL LETTER MU" case maps to "MICRO SIGN".
+    // "GREEK SMALL LETTER MU" case maps to "MICRO SIGN".
+    if (c == 0x039C || c == 0x03BC)
+        return 0xB5;
+    // "LATIN CAPITAL LETTER Y WITH DIAERESIS" case maps to "LATIN SMALL LETTER Y WITH DIAERESIS".
+    if (c == 0x0178)
+        return 0xFF;
+    return 0;
+}
+
+const int js::irregexp::kSpaceRanges[] = {
+    0x0009, 0x000D + 1, // CHARACTER TABULATION..CARRIAGE RETURN (CR)
+    0x0020, 0x0020 + 1, // SPACE
+    0x00A0, 0x00A0 + 1, // NO-BREAK SPACE
+    0x1680, 0x1680 + 1, // OGHAM SPACE MARK
+    0x2000, 0x200A + 1, // EN QUAD..HAIR SPACE
+    0x2028, 0x2029 + 1, // LINE SEPARATOR..PARAGRAPH SEPARATOR
+    0x202F, 0x202F + 1, // NARROW NO-BREAK SPACE
+    0x205F, 0x205F + 1, // MEDIUM MATHEMATICAL SPACE
+    0x3000, 0x3000 + 1, // IDEOGRAPHIC SPACE
+    0xFEFF, 0xFEFF + 1, // ZERO WIDTH NO-BREAK SPACE
+    0xFFFF + 1
+};
+const int js::irregexp::kSpaceRangeCount = 21;
+
+const int js::irregexp::kSpaceAndSurrogateRanges[] = {
+    0x0009, 0x000D + 1, // CHARACTER TABULATION..CARRIAGE RETURN (CR)
+    0x0020, 0x0020 + 1, // SPACE
+    0x00A0, 0x00A0 + 1, // NO-BREAK SPACE
+    0x1680, 0x1680 + 1, // OGHAM SPACE MARK
+    0x2000, 0x200A + 1, // EN QUAD..HAIR SPACE
+    0x2028, 0x2029 + 1, // LINE SEPARATOR..PARAGRAPH SEPARATOR
+    0x202F, 0x202F + 1, // NARROW NO-BREAK SPACE
+    0x205F, 0x205F + 1, // MEDIUM MATHEMATICAL SPACE
+    0x3000, 0x3000 + 1, // IDEOGRAPHIC SPACE
+    0xD800, 0xDFFF + 1, // <Lead Surrogate Min>..<Trail Surrogate Max>
+    0xFEFF, 0xFEFF + 1, // ZERO WIDTH NO-BREAK SPACE
+    0xFFFF + 1
+};
+const int js::irregexp::kSpaceAndSurrogateRangeCount = 23;
+
+const int js::irregexp::kWordRanges[] = {
+    0x0030, 0x0039 + 1, // DIGIT ZERO..DIGIT NINE
+    0x0041, 0x005A + 1, // LATIN CAPITAL LETTER A..LATIN CAPITAL LETTER Z
+    0x005F, 0x005F + 1, // LOW LINE
+    0x0061, 0x007A + 1, // LATIN SMALL LETTER A..LATIN SMALL LETTER Z
+    0xFFFF + 1
+};
+const int js::irregexp::kWordRangeCount = 9;
+
+const int js::irregexp::kIgnoreCaseWordRanges[] = {
+    0x0030, 0x0039 + 1, // DIGIT ZERO..DIGIT NINE
+    0x0041, 0x005A + 1, // LATIN CAPITAL LETTER A..LATIN CAPITAL LETTER Z
+    0x005F, 0x005F + 1, // LOW LINE
+    0x0061, 0x007A + 1, // LATIN SMALL LETTER A..LATIN SMALL LETTER Z
+    0x017F, 0x017F + 1, // LATIN SMALL LETTER LONG S
+    0x212A, 0x212A + 1, // KELVIN SIGN
+    0xFFFF + 1
+};
+const int js::irregexp::kIgnoreCaseWordRangeCount = 13;
+
+const int js::irregexp::kWordAndSurrogateRanges[] = {
+    0x0030, 0x0039 + 1, // DIGIT ZERO..DIGIT NINE
+    0x0041, 0x005A + 1, // LATIN CAPITAL LETTER A..LATIN CAPITAL LETTER Z
+    0x005F, 0x005F + 1, // LOW LINE
+    0x0061, 0x007A + 1, // LATIN SMALL LETTER A..LATIN SMALL LETTER Z
+    0xD800, 0xDFFF + 1, // <Lead Surrogate Min>..<Trail Surrogate Max>
+    0xFFFF + 1
+};
+const int js::irregexp::kWordAndSurrogateRangeCount = 11;
+
+const int js::irregexp::kNegatedIgnoreCaseWordAndSurrogateRanges[] = {
+    0x0000, 0x002F + 1, // NULL..SOLIDUS
+    0x003A, 0x0040 + 1, // COLON..COMMERCIAL AT
+    0x005B, 0x005E + 1, // LEFT SQUARE BRACKET..CIRCUMFLEX ACCENT
+    0x0060, 0x0060 + 1, // GRAVE ACCENT
+    0x007B, 0x017E + 1, // LEFT CURLY BRACKET..LATIN SMALL LETTER Z WITH CARON
+    0x0180, 0x2129 + 1, // LATIN SMALL LETTER B WITH STROKE..TURNED GREEK SMALL LETTER IOTA
+    0x212B, 0xD7FF + 1, // ANGSTROM SIGN..<Unused>
+    0xE000, 0xFFFF + 1, // Private Use..<Unused>
+    0xFFFF + 1
+};
+const int js::irregexp::kNegatedIgnoreCaseWordAndSurrogateRangeCount = 17;
+
+const int js::irregexp::kDigitRanges[] = {
+    0x0030, 0x0039 + 1, // DIGIT ZERO..DIGIT NINE
+    0xFFFF + 1
+};
+const int js::irregexp::kDigitRangeCount = 3;
+
+const int js::irregexp::kDigitAndSurrogateRanges[] = {
+    0x0030, 0x0039 + 1, // DIGIT ZERO..DIGIT NINE
+    0xD800, 0xDFFF + 1, // <Lead Surrogate Min>..<Trail Surrogate Max>
+    0xFFFF + 1
+};
+const int js::irregexp::kDigitAndSurrogateRangeCount = 5;
+
+const int js::irregexp::kSurrogateRanges[] = {
+    0xD800, 0xDFFF + 1, // <Lead Surrogate Min>..<Trail Surrogate Max>
+    0xFFFF + 1
+};
+const int js::irregexp::kSurrogateRangeCount = 3;
+
+const int js::irregexp::kLineTerminatorRanges[] = {
+    0x000A, 0x000A + 1, // LINE FEED (LF)
+    0x000D, 0x000D + 1, // CARRIAGE RETURN (CR)
+    0x2028, 0x2029 + 1, // LINE SEPARATOR..PARAGRAPH SEPARATOR
+    0xFFFF + 1
+};
+const int js::irregexp::kLineTerminatorRangeCount = 7;
diff --git a/js/src/irregexp/RegExpCharacters.h b/js/src/irregexp/RegExpCharacters.h
new file mode 100644
index 0000000000..0d3cf096fa
--- /dev/null
+++ b/js/src/irregexp/RegExpCharacters.h
@@ -0,0 +1,90 @@
+/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 4 -*-
+ * vim: set ts=8 sts=4 et sw=4 tw=99: */
+
+// Copyright 2012 the V8 project authors. All rights reserved.
+// 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.
+
+#ifndef V8_JSREGEXPCHARACTERS_H_
+#define V8_JSREGEXPCHARACTERS_H_
+
+namespace js {
+
+namespace irregexp {
+
+char16_t
+ConvertNonLatin1ToLatin1(char16_t c, bool unicode);
+
+// -------------------------------------------------------------------
+// CharacterRange
+
+// The ranges have inclusive from and exclusive to.
+
+// This covers \s as defined in ES2016, 21.2.2.12 CharacterClassEscape,
+// which includes WhiteSpace (11.2) and LineTerminator (11.3) values.
+extern const int kSpaceRanges[];
+extern const int kSpaceRangeCount;
+
+// Characters in \s and additionally all surrogate characters.
+extern const int kSpaceAndSurrogateRanges[];
+extern const int kSpaceAndSurrogateRangeCount;
+
+// This covers \w as defined in ES2016, 21.2.2.12 CharacterClassEscape.
+extern const int kWordRanges[];
+extern const int kWordRangeCount;
+
+// Characters which case-fold to characters in \w.
+extern const int kIgnoreCaseWordRanges[];
+extern const int kIgnoreCaseWordRangeCount;
+
+// Characters in \w and additionally all surrogate characters.
+extern const int kWordAndSurrogateRanges[];
+extern const int kWordAndSurrogateRangeCount;
+
+// All characters excluding those which case-fold to \w and excluding all
+// surrogate characters.
+extern const int kNegatedIgnoreCaseWordAndSurrogateRanges[];
+extern const int kNegatedIgnoreCaseWordAndSurrogateRangeCount;
+
+// This covers \d as defined in ES2016, 21.2.2.12 CharacterClassEscape.
+extern const int kDigitRanges[];
+extern const int kDigitRangeCount;
+
+// Characters in \d and additionally all surrogate characters.
+extern const int kDigitAndSurrogateRanges[];
+extern const int kDigitAndSurrogateRangeCount;
+
+// The range of all surrogate characters.
+extern const int kSurrogateRanges[];
+extern const int kSurrogateRangeCount;
+
+// Line terminators as defined in ES2016, 11.3 LineTerminator.
+extern const int kLineTerminatorRanges[];
+extern const int kLineTerminatorRangeCount;
+
+} } // namespace js::irregexp
+
+#endif // V8_JSREGEXPCHARACTERS_H_
diff --git a/js/src/irregexp/RegExpEngine.cpp b/js/src/irregexp/RegExpEngine.cpp
index 2e19065fdf..0011e976f3 100644
--- a/js/src/irregexp/RegExpEngine.cpp
+++ b/js/src/irregexp/RegExpEngine.cpp
@@ -31,10 +31,14 @@
 #include "irregexp/RegExpEngine.h"
 
 #include "irregexp/NativeRegExpMacroAssembler.h"
+#include "irregexp/RegExpCharacters.h" 
 #include "irregexp/RegExpMacroAssembler.h"
 #include "jit/ExecutableAllocator.h"
 #include "jit/JitCommon.h"
 
+// Generated table
+#include "irregexp/RegExpCharacters-inl.h"
+
 using namespace js;
 using namespace js::irregexp;
 
@@ -61,61 +65,6 @@ RegExpNode::RegExpNode(LifoAlloc* alloc)
     bm_info_[0] = bm_info_[1] = nullptr;
 }
 
-// -------------------------------------------------------------------
-// CharacterRange
-
-// The '2' variant has inclusive from and exclusive to.
-// This covers \s as defined in ECMA-262 5.1, 15.10.2.12,
-// which include WhiteSpace (7.2) or LineTerminator (7.3) values.
-static const int kSpaceRanges[] = { '\t', '\r' + 1, ' ', ' ' + 1,
-    0x00A0, 0x00A1, 0x1680, 0x1681, 0x180E, 0x180F, 0x2000, 0x200B,
-    0x2028, 0x202A, 0x202F, 0x2030, 0x205F, 0x2060, 0x3000, 0x3001,
-    0xFEFF, 0xFF00, 0x10000 };
-static const int kSpaceRangeCount = ArrayLength(kSpaceRanges);
-
-static const int kSpaceAndSurrogateRanges[] = { '\t', '\r' + 1, ' ', ' ' + 1,
-    0x00A0, 0x00A1, 0x1680, 0x1681, 0x180E, 0x180F, 0x2000, 0x200B,
-    0x2028, 0x202A, 0x202F, 0x2030, 0x205F, 0x2060, 0x3000, 0x3001,
-    unicode::LeadSurrogateMin, unicode::TrailSurrogateMax + 1,
-    0xFEFF, 0xFF00, 0x10000 };
-static const int kSpaceAndSurrogateRangeCount = ArrayLength(kSpaceAndSurrogateRanges);
-static const int kWordRanges[] = {
-    '0', '9' + 1, 'A', 'Z' + 1, '_', '_' + 1, 'a', 'z' + 1, 0x10000 };
-static const int kWordRangeCount = ArrayLength(kWordRanges);
-static const int kIgnoreCaseWordRanges[] = {
-    '0', '9' + 1, 'A', 'Z' + 1, '_', '_' + 1, 'a', 'z' + 1,
-    0x017F, 0x017F + 1, 0x212A, 0x212A + 1,
-    0x10000 };
-static const int kIgnoreCaseWordCount = ArrayLength(kIgnoreCaseWordRanges);
-static const int kWordAndSurrogateRanges[] = {
-    '0', '9' + 1, 'A', 'Z' + 1, '_', '_' + 1, 'a', 'z' + 1,
-    unicode::LeadSurrogateMin, unicode::TrailSurrogateMax + 1,
-    0x10000 };
-static const int kWordAndSurrogateRangeCount = ArrayLength(kWordAndSurrogateRanges);
-static const int kNegatedIgnoreCaseWordAndSurrogateRanges[] = {
-    0, '0', '9' + 1, 'A',
-    'Z' + 1, '_', '_' + 1, 'a',
-    'z' + 1, 0x017F,
-    0x017F + 1, 0x212A,
-    0x212A + 1, unicode::LeadSurrogateMin,
-    unicode::TrailSurrogateMax + 1, 0x10000,
-    0x10000 };
-static const int kNegatedIgnoreCaseWordAndSurrogateRangeCount =
-    ArrayLength(kNegatedIgnoreCaseWordAndSurrogateRanges);
-static const int kDigitRanges[] = { '0', '9' + 1, 0x10000 };
-static const int kDigitRangeCount = ArrayLength(kDigitRanges);
-static const int kDigitAndSurrogateRanges[] = {
-    '0', '9' + 1,
-    unicode::LeadSurrogateMin, unicode::TrailSurrogateMax + 1,
-    0x10000 };
-static const int kDigitAndSurrogateRangeCount = ArrayLength(kDigitAndSurrogateRanges);
-static const int kSurrogateRanges[] = {
-    unicode::LeadSurrogateMin, unicode::TrailSurrogateMax + 1,
-    0x10000 };
-static const int kSurrogateRangeCount = ArrayLength(kSurrogateRanges);
-static const int kLineTerminatorRanges[] = { 0x000A, 0x000B, 0x000D, 0x000E,
-    0x2028, 0x202A, 0x10000 };
-static const int kLineTerminatorRangeCount = ArrayLength(kLineTerminatorRanges);
 static const int kMaxOneByteCharCode = 0xff;
 static const int kMaxUtf16CodeUnit = 0xffff;
 
@@ -213,7 +162,7 @@ CharacterRange::AddClassEscapeUnicode(LifoAlloc* alloc, char16_t type,
         break;
       case 'w':
         if (ignore_case)
-            AddClass(kIgnoreCaseWordRanges, kIgnoreCaseWordCount, ranges);
+            AddClass(kIgnoreCaseWordRanges, kIgnoreCaseWordRangeCount, ranges);
         else
             AddClassEscape(alloc, type, ranges);
         break;
@@ -233,33 +182,6 @@ CharacterRange::AddClassEscapeUnicode(LifoAlloc* alloc, char16_t type,
     }
 }
 
-#define FOR_EACH_NON_ASCII_TO_ASCII_FOLDING(macro)      \
-    /* LATIN CAPITAL LETTER Y WITH DIAERESIS */         \
-    macro(0x0178, 0x00FF)                               \
-    /* LATIN SMALL LETTER LONG S */                     \
-    macro(0x017F, 0x0073)                               \
-    /* LATIN CAPITAL LETTER SHARP S */                  \
-    macro(0x1E9E, 0x00DF)                               \
-    /* KELVIN SIGN */                                   \
-    macro(0x212A, 0x006B)                               \
-    /* ANGSTROM SIGN */                                 \
-    macro(0x212B, 0x00E5)
-
-// We need to check for the following characters: 0x39c 0x3bc 0x178.
-static inline bool
-RangeContainsLatin1Equivalents(CharacterRange range, bool unicode)
-{
-    /* TODO(dcarney): this could be a lot more efficient. */
-    if (unicode) {
-#define CHECK_RANGE(C, F) \
-        if (range.Contains(C)) return true;
-FOR_EACH_NON_ASCII_TO_ASCII_FOLDING(CHECK_RANGE)
-#undef CHECK_RANGE
-    }
-
-    return range.Contains(0x39c) || range.Contains(0x3bc) || range.Contains(0x178);
-}
-
 static bool
 RangesContainLatin1Equivalents(const CharacterRangeVector& ranges, bool unicode)
 {
@@ -336,7 +258,7 @@ GetCaseIndependentLetters(char16_t character,
     // step 3.g.
     // The standard requires that non-ASCII characters cannot have ASCII
     // character codes in their equivalence class, even though this
-    // situation occurs multiple times in the unicode tables.
+    // situation occurs multiple times in the Unicode tables.
     static const unsigned kMaxAsciiCharCode = 127;
     if (upper <= kMaxAsciiCharCode) {
         if (character > kMaxAsciiCharCode) {
@@ -365,31 +287,6 @@ GetCaseIndependentLetters(char16_t character,
                                      choices, ArrayLength(choices), letters);
 }
 
-static char16_t
-ConvertNonLatin1ToLatin1(char16_t c, bool unicode)
-{
-    MOZ_ASSERT(c > kMaxOneByteCharCode);
-    if (unicode) {
-        switch (c) {
-#define CONVERT(C, F) case C: return F;
-FOR_EACH_NON_ASCII_TO_ASCII_FOLDING(CONVERT)
-#undef CONVERT
-        }
-    }
-
-    switch (c) {
-      // This are equivalent characters in unicode.
-      case 0x39c:
-      case 0x3bc:
-        return 0xb5;
-      // This is an uppercase of a Latin-1 character
-      // outside of Latin-1.
-      case 0x178:
-        return 0xff;
-    }
-    return 0;
-}
-
 void
 CharacterRange::AddCaseEquivalents(bool is_ascii, bool unicode, CharacterRangeVector* ranges)
 {