- Revision
- 114844
- Author
- da...@apple.com
- Date
- 2012-04-21 12:46:02 -0700 (Sat, 21 Apr 2012)
Log Message
Change _javascript_ lexer to use 0 instead of -1 for sentinel, eliminating the need to put characters into ints
https://bugs.webkit.org/show_bug.cgi?id=84523
Reviewed by Oliver Hunt.
Separate preparation step of copyright dates, renaming, and other small tweaks.
* parser/Lexer.cpp:
(JSC::Lexer::invalidCharacterMessage): Removed "get" from name to match WebKit naming conventions.
(JSC::Lexer::peek): Removed meaningless comment.
(JSC::Lexer::parseFourDigitUnicodeHex): Renamed from getUnicodeCharacter to be more precise about
what this function does.
(JSC::Lexer::shiftLineTerminator): Renamed local variable that had a data-member-style name.
(JSC::Lexer::parseStringSlowCase): Updated for new name of parseFourDigitUnicodeHex.
(JSC::Lexer::lex): Updated for new name of invalidCharacterMessage.
* parser/Lexer.h: Removed an unneeded forward declaration of the RegExp class.
Renamed getInvalidCharMessage to invalidCharacterMessage and made it const. Renamed
getUnicodeCharacter to parseFourDigitUnicodeHex.
Modified Paths
Diff
Modified: trunk/Source/_javascript_Core/ChangeLog (114843 => 114844)
--- trunk/Source/_javascript_Core/ChangeLog 2012-04-21 16:43:25 UTC (rev 114843)
+++ trunk/Source/_javascript_Core/ChangeLog 2012-04-21 19:46:02 UTC (rev 114844)
@@ -1,3 +1,25 @@
+2012-04-21 Darin Adler <da...@apple.com>
+
+ Change _javascript_ lexer to use 0 instead of -1 for sentinel, eliminating the need to put characters into ints
+ https://bugs.webkit.org/show_bug.cgi?id=84523
+
+ Reviewed by Oliver Hunt.
+
+ Separate preparation step of copyright dates, renaming, and other small tweaks.
+
+ * parser/Lexer.cpp:
+ (JSC::Lexer::invalidCharacterMessage): Removed "get" from name to match WebKit naming conventions.
+ (JSC::Lexer::peek): Removed meaningless comment.
+ (JSC::Lexer::parseFourDigitUnicodeHex): Renamed from getUnicodeCharacter to be more precise about
+ what this function does.
+ (JSC::Lexer::shiftLineTerminator): Renamed local variable that had a data-member-style name.
+ (JSC::Lexer::parseStringSlowCase): Updated for new name of parseFourDigitUnicodeHex.
+ (JSC::Lexer::lex): Updated for new name of invalidCharacterMessage.
+
+ * parser/Lexer.h: Removed an unneeded forward declaration of the RegExp class.
+ Renamed getInvalidCharMessage to invalidCharacterMessage and made it const. Renamed
+ getUnicodeCharacter to parseFourDigitUnicodeHex.
+
2012-04-20 Filip Pizlo <fpi...@apple.com>
DFG should optimize int8 and int16 arrays on ARMv7
Modified: trunk/Source/_javascript_Core/parser/Lexer.cpp (114843 => 114844)
--- trunk/Source/_javascript_Core/parser/Lexer.cpp 2012-04-21 16:43:25 UTC (rev 114843)
+++ trunk/Source/_javascript_Core/parser/Lexer.cpp 2012-04-21 19:46:02 UTC (rev 114844)
@@ -1,6 +1,6 @@
/*
* Copyright (C) 1999-2000 Harri Porten (por...@kde.org)
- * Copyright (C) 2006, 2007, 2008, 2009 Apple Inc. All Rights Reserved.
+ * Copyright (C) 2006, 2007, 2008, 2009, 2011, 2012 Apple Inc. All Rights Reserved.
* Copyright (C) 2007 Cameron Zwarich (cwzwar...@uwaterloo.ca)
* Copyright (C) 2010 Zoltan Herczeg (zherc...@inf.u-szeged.hu)
* Copyright (C) 2012 Mathias Bynens (math...@qiwi.be)
@@ -368,7 +368,7 @@
}
template <typename T>
-UString Lexer<T>::getInvalidCharMessage()
+UString Lexer<T>::invalidCharacterMessage() const
{
switch (m_current) {
case 0:
@@ -450,14 +450,13 @@
template <typename T>
ALWAYS_INLINE int Lexer<T>::peek(int offset)
{
- // Only use if necessary
ASSERT(offset > 0 && offset < 5);
const T* code = m_code + offset;
return (code < m_codeEnd) ? *code : -1;
}
template <typename T>
-int Lexer<T>::getUnicodeCharacter()
+int Lexer<T>::parseFourDigitUnicodeHex()
{
int char1 = peek(1);
int char2 = peek(2);
@@ -479,11 +478,11 @@
{
ASSERT(isLineTerminator(static_cast<T>(m_current)));
- int m_prev = m_current;
+ int prev = m_current;
shift();
// Allow both CRLF and LFCR.
- if (m_prev + m_current == '\n' + '\r')
+ if (prev + m_current == '\n' + '\r')
shift();
++m_lineNumber;
@@ -633,7 +632,7 @@
}
template <>
- template <bool shouldCreateIdentifier> ALWAYS_INLINE JSTokenType Lexer<LChar>::parseIdentifier(JSTokenData* tokenData, unsigned lexerFlags, bool strictMode)
+template <bool shouldCreateIdentifier> ALWAYS_INLINE JSTokenType Lexer<LChar>::parseIdentifier(JSTokenData* tokenData, unsigned lexerFlags, bool strictMode)
{
const ptrdiff_t remaining = m_codeEnd - m_code;
if ((remaining >= maxTokenLength) && !(lexerFlags & LexerFlagsIgnoreReservedWords)) {
@@ -691,6 +690,7 @@
return keyword == RESERVED_IF_STRICT && !strictMode ? IDENT : keyword;
}
}
+
const UChar* identifierStart = currentCharacter();
UChar orAllChars = 0;
@@ -762,7 +762,7 @@
if (UNLIKELY(m_current != 'u'))
return ERRORTOK;
shift();
- int character = getUnicodeCharacter();
+ int character = parseFourDigitUnicodeHex();
if (UNLIKELY(character == -1))
return ERRORTOK;
UChar ucharacter = static_cast<UChar>(character);
@@ -910,7 +910,7 @@
shift();
} else if (m_current == 'u') {
shift();
- int character = getUnicodeCharacter();
+ int character = parseFourDigitUnicodeHex();
if (character != -1) {
if (shouldBuildStrings)
record16(character);
@@ -1508,7 +1508,7 @@
m_terminator = true;
goto start;
case CharacterInvalid:
- m_lexErrorMessage = getInvalidCharMessage();
+ m_lexErrorMessage = invalidCharacterMessage();
goto returnError;
default:
ASSERT_NOT_REACHED();
Modified: trunk/Source/_javascript_Core/parser/Lexer.h (114843 => 114844)
--- trunk/Source/_javascript_Core/parser/Lexer.h 2012-04-21 16:43:25 UTC (rev 114843)
+++ trunk/Source/_javascript_Core/parser/Lexer.h 2012-04-21 19:46:02 UTC (rev 114844)
@@ -1,6 +1,6 @@
/*
* Copyright (C) 1999-2000 Harri Porten (por...@kde.org)
- * Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 Apple Inc. All rights reserved.
+ * Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2011, 2012 Apple Inc. All rights reserved.
* Copyright (C) 2010 Zoltan Herczeg (zherc...@inf.u-szeged.hu)
*
* This library is free software; you can redistribute it and/or
@@ -67,8 +67,6 @@
LexexFlagsDontBuildKeywords = 4
};
-class RegExp;
-
template <typename T>
class Lexer {
WTF_MAKE_NONCOPYABLE(Lexer);
@@ -134,10 +132,10 @@
ALWAYS_INLINE void shift();
ALWAYS_INLINE int peek(int offset);
- int getUnicodeCharacter();
+ int parseFourDigitUnicodeHex();
void shiftLineTerminator();
- UString getInvalidCharMessage();
+ UString invalidCharacterMessage() const;
ALWAYS_INLINE const T* currentCharacter() const;
ALWAYS_INLINE int currentOffset() const { return m_code - m_codeStart; }
ALWAYS_INLINE void setOffsetFromCharOffset(const T* charOffset) { setOffset(charOffset - m_codeStart); }
