From 40736c5763bf61337c8c14e16d8587db021a87d4 Mon Sep 17 00:00:00 2001 From: Simon Hausmann Date: Fri, 6 Jan 2012 14:44:00 +0100 Subject: Imported WebKit commit 2ea9d364d0f6efa8fa64acf19f451504c59be0e4 (http://svn.webkit.org/repository/webkit/trunk@104285) --- Source/JavaScriptCore/parser/ASTBuilder.h | 979 ++++++++++++ Source/JavaScriptCore/parser/Keywords.table | 58 + Source/JavaScriptCore/parser/Lexer.cpp | 1543 ++++++++++++++++++ Source/JavaScriptCore/parser/Lexer.h | 297 ++++ Source/JavaScriptCore/parser/NodeConstructors.h | 898 +++++++++++ Source/JavaScriptCore/parser/NodeInfo.h | 62 + Source/JavaScriptCore/parser/Nodes.cpp | 196 +++ Source/JavaScriptCore/parser/Nodes.h | 1616 +++++++++++++++++++ Source/JavaScriptCore/parser/Parser.cpp | 1672 ++++++++++++++++++++ Source/JavaScriptCore/parser/Parser.h | 1046 ++++++++++++ Source/JavaScriptCore/parser/ParserArena.cpp | 122 ++ Source/JavaScriptCore/parser/ParserArena.h | 189 +++ Source/JavaScriptCore/parser/ParserTokens.h | 162 ++ Source/JavaScriptCore/parser/ResultType.h | 182 +++ Source/JavaScriptCore/parser/SourceCode.h | 105 ++ Source/JavaScriptCore/parser/SourceProvider.h | 108 ++ .../JavaScriptCore/parser/SourceProviderCache.cpp | 56 + Source/JavaScriptCore/parser/SourceProviderCache.h | 53 + .../parser/SourceProviderCacheItem.h | 73 + Source/JavaScriptCore/parser/SyntaxChecker.h | 256 +++ 20 files changed, 9673 insertions(+) create mode 100644 Source/JavaScriptCore/parser/ASTBuilder.h create mode 100644 Source/JavaScriptCore/parser/Keywords.table create mode 100644 Source/JavaScriptCore/parser/Lexer.cpp create mode 100644 Source/JavaScriptCore/parser/Lexer.h create mode 100644 Source/JavaScriptCore/parser/NodeConstructors.h create mode 100644 Source/JavaScriptCore/parser/NodeInfo.h create mode 100644 Source/JavaScriptCore/parser/Nodes.cpp create mode 100644 Source/JavaScriptCore/parser/Nodes.h create mode 100644 Source/JavaScriptCore/parser/Parser.cpp create mode 100644 Source/JavaScriptCore/parser/Parser.h create mode 100644 Source/JavaScriptCore/parser/ParserArena.cpp create mode 100644 Source/JavaScriptCore/parser/ParserArena.h create mode 100644 Source/JavaScriptCore/parser/ParserTokens.h create mode 100644 Source/JavaScriptCore/parser/ResultType.h create mode 100644 Source/JavaScriptCore/parser/SourceCode.h create mode 100644 Source/JavaScriptCore/parser/SourceProvider.h create mode 100644 Source/JavaScriptCore/parser/SourceProviderCache.cpp create mode 100644 Source/JavaScriptCore/parser/SourceProviderCache.h create mode 100644 Source/JavaScriptCore/parser/SourceProviderCacheItem.h create mode 100644 Source/JavaScriptCore/parser/SyntaxChecker.h (limited to 'Source/JavaScriptCore/parser') diff --git a/Source/JavaScriptCore/parser/ASTBuilder.h b/Source/JavaScriptCore/parser/ASTBuilder.h new file mode 100644 index 000000000..2e81c2546 --- /dev/null +++ b/Source/JavaScriptCore/parser/ASTBuilder.h @@ -0,0 +1,979 @@ +/* + * Copyright (C) 2010 Apple Inc. All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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. + * + * THIS SOFTWARE IS PROVIDED BY APPLE INC. AND ITS 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 APPLE INC. OR ITS 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 ASTBuilder_h +#define ASTBuilder_h + +#include "NodeConstructors.h" +#include "SyntaxChecker.h" +#include + +namespace JSC { + +class ASTBuilder { + struct BinaryOpInfo { + BinaryOpInfo() {} + BinaryOpInfo(int s, int d, int e, bool r) + : start(s) + , divot(d) + , end(e) + , hasAssignment(r) + { + } + BinaryOpInfo(const BinaryOpInfo& lhs, const BinaryOpInfo& rhs) + : start(lhs.start) + , divot(rhs.start) + , end(rhs.end) + , hasAssignment(lhs.hasAssignment || rhs.hasAssignment) + { + } + int start; + int divot; + int end; + bool hasAssignment; + }; + + + struct AssignmentInfo { + AssignmentInfo() {} + AssignmentInfo(ExpressionNode* node, int start, int divot, int initAssignments, Operator op) + : m_node(node) + , m_start(start) + , m_divot(divot) + , m_initAssignments(initAssignments) + , m_op(op) + { + } + ExpressionNode* m_node; + int m_start; + int m_divot; + int m_initAssignments; + Operator m_op; + }; +public: + ASTBuilder(JSGlobalData* globalData, SourceCode* sourceCode) + : m_globalData(globalData) + , m_sourceCode(sourceCode) + , m_scope(globalData) + , m_evalCount(0) + { + } + + struct BinaryExprContext { + BinaryExprContext(ASTBuilder&) {} + }; + struct UnaryExprContext { + UnaryExprContext(ASTBuilder&) {} + }; + + typedef SyntaxChecker FunctionBodyBuilder; + + typedef ExpressionNode* Expression; + typedef JSC::SourceElements* SourceElements; + typedef ArgumentsNode* Arguments; + typedef CommaNode* Comma; + typedef PropertyNode* Property; + typedef PropertyListNode* PropertyList; + typedef ElementNode* ElementList; + typedef ArgumentListNode* ArgumentsList; + typedef ParameterNode* FormalParameterList; + typedef FunctionBodyNode* FunctionBody; + typedef StatementNode* Statement; + typedef ClauseListNode* ClauseList; + typedef CaseClauseNode* Clause; + typedef ConstDeclNode* ConstDeclList; + typedef std::pair BinaryOperand; + + static const bool CreatesAST = true; + static const bool NeedsFreeVariableInfo = true; + static const bool CanUseFunctionCache = true; + static const int DontBuildKeywords = 0; + static const int DontBuildStrings = 0; + + ExpressionNode* makeBinaryNode(int lineNumber, int token, std::pair, std::pair); + ExpressionNode* makeFunctionCallNode(int lineNumber, ExpressionNode* func, ArgumentsNode* args, int start, int divot, int end); + + JSC::SourceElements* createSourceElements() { return new (m_globalData) JSC::SourceElements(); } + + ParserArenaData* varDeclarations() { return m_scope.m_varDeclarations; } + ParserArenaData* funcDeclarations() { return m_scope.m_funcDeclarations; } + int features() const { return m_scope.m_features; } + int numConstants() const { return m_scope.m_numConstants; } + + void appendToComma(CommaNode* commaNode, ExpressionNode* expr) { commaNode->append(expr); } + + CommaNode* createCommaExpr(int lineNumber, ExpressionNode* lhs, ExpressionNode* rhs) { return new (m_globalData) CommaNode(lineNumber, lhs, rhs); } + + ExpressionNode* makeAssignNode(int lineNumber, ExpressionNode* left, Operator, ExpressionNode* right, bool leftHasAssignments, bool rightHasAssignments, int start, int divot, int end); + ExpressionNode* makePrefixNode(int lineNumber, ExpressionNode*, Operator, int start, int divot, int end); + ExpressionNode* makePostfixNode(int lineNumber, ExpressionNode*, Operator, int start, int divot, int end); + ExpressionNode* makeTypeOfNode(int lineNumber, ExpressionNode*); + ExpressionNode* makeDeleteNode(int lineNumber, ExpressionNode*, int start, int divot, int end); + ExpressionNode* makeNegateNode(int lineNumber, ExpressionNode*); + ExpressionNode* makeBitwiseNotNode(int lineNumber, ExpressionNode*); + ExpressionNode* makeMultNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeDivNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeModNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeAddNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeSubNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeBitXOrNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeBitAndNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeBitOrNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeLeftShiftNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeRightShiftNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + ExpressionNode* makeURightShiftNode(int lineNumber, ExpressionNode* left, ExpressionNode* right, bool rightHasAssignments); + + ExpressionNode* createLogicalNot(int lineNumber, ExpressionNode* expr) { return new (m_globalData) LogicalNotNode(lineNumber, expr); } + ExpressionNode* createUnaryPlus(int lineNumber, ExpressionNode* expr) { return new (m_globalData) UnaryPlusNode(lineNumber, expr); } + ExpressionNode* createVoid(int lineNumber, ExpressionNode* expr) + { + incConstants(); + return new (m_globalData) VoidNode(lineNumber, expr); + } + ExpressionNode* thisExpr(int lineNumber) + { + usesThis(); + return new (m_globalData) ThisNode(lineNumber); + } + ExpressionNode* createResolve(int lineNumber, const Identifier* ident, int start) + { + if (m_globalData->propertyNames->arguments == *ident) + usesArguments(); + return new (m_globalData) ResolveNode(lineNumber, *ident, start); + } + ExpressionNode* createObjectLiteral(int lineNumber) { return new (m_globalData) ObjectLiteralNode(lineNumber); } + ExpressionNode* createObjectLiteral(int lineNumber, PropertyListNode* properties) { return new (m_globalData) ObjectLiteralNode(lineNumber, properties); } + + ExpressionNode* createArray(int lineNumber, int elisions) + { + if (elisions) + incConstants(); + return new (m_globalData) ArrayNode(lineNumber, elisions); + } + + ExpressionNode* createArray(int lineNumber, ElementNode* elems) { return new (m_globalData) ArrayNode(lineNumber, elems); } + ExpressionNode* createArray(int lineNumber, int elisions, ElementNode* elems) + { + if (elisions) + incConstants(); + return new (m_globalData) ArrayNode(lineNumber, elisions, elems); + } + ExpressionNode* createNumberExpr(int lineNumber, double d) + { + incConstants(); + return new (m_globalData) NumberNode(lineNumber, d); + } + + ExpressionNode* createString(int lineNumber, const Identifier* string) + { + incConstants(); + return new (m_globalData) StringNode(lineNumber, *string); + } + + ExpressionNode* createBoolean(int lineNumber, bool b) + { + incConstants(); + return new (m_globalData) BooleanNode(lineNumber, b); + } + + ExpressionNode* createNull(int lineNumber) + { + incConstants(); + return new (m_globalData) NullNode(lineNumber); + } + + ExpressionNode* createBracketAccess(int lineNumber, ExpressionNode* base, ExpressionNode* property, bool propertyHasAssignments, int start, int divot, int end) + { + BracketAccessorNode* node = new (m_globalData) BracketAccessorNode(lineNumber, base, property, propertyHasAssignments); + setExceptionLocation(node, start, divot, end); + return node; + } + + ExpressionNode* createDotAccess(int lineNumber, ExpressionNode* base, const Identifier* property, int start, int divot, int end) + { + DotAccessorNode* node = new (m_globalData) DotAccessorNode(lineNumber, base, *property); + setExceptionLocation(node, start, divot, end); + return node; + } + + ExpressionNode* createRegExp(int lineNumber, const Identifier& pattern, const Identifier& flags, int start) + { + if (Yarr::checkSyntax(pattern.ustring())) + return 0; + RegExpNode* node = new (m_globalData) RegExpNode(lineNumber, pattern, flags); + int size = pattern.length() + 2; // + 2 for the two /'s + setExceptionLocation(node, start, start + size, start + size); + return node; + } + + ExpressionNode* createNewExpr(int lineNumber, ExpressionNode* expr, ArgumentsNode* arguments, int start, int divot, int end) + { + NewExprNode* node = new (m_globalData) NewExprNode(lineNumber, expr, arguments); + setExceptionLocation(node, start, divot, end); + return node; + } + + ExpressionNode* createNewExpr(int lineNumber, ExpressionNode* expr, int start, int end) + { + NewExprNode* node = new (m_globalData) NewExprNode(lineNumber, expr); + setExceptionLocation(node, start, end, end); + return node; + } + + ExpressionNode* createConditionalExpr(int lineNumber, ExpressionNode* condition, ExpressionNode* lhs, ExpressionNode* rhs) + { + return new (m_globalData) ConditionalNode(lineNumber, condition, lhs, rhs); + } + + ExpressionNode* createAssignResolve(int lineNumber, const Identifier& ident, ExpressionNode* rhs, bool rhsHasAssignment, int start, int divot, int end) + { + AssignResolveNode* node = new (m_globalData) AssignResolveNode(lineNumber, ident, rhs, rhsHasAssignment); + setExceptionLocation(node, start, divot, end); + return node; + } + + ExpressionNode* createFunctionExpr(int lineNumber, const Identifier* name, FunctionBodyNode* body, ParameterNode* parameters, int openBracePos, int closeBracePos, int bodyStartLine, int bodyEndLine) + { + FuncExprNode* result = new (m_globalData) FuncExprNode(lineNumber, *name, body, m_sourceCode->subExpression(openBracePos, closeBracePos, bodyStartLine), parameters); + body->setLoc(bodyStartLine, bodyEndLine); + return result; + } + + FunctionBodyNode* createFunctionBody(int lineNumber, bool inStrictContext) + { + usesClosures(); + return FunctionBodyNode::create(m_globalData, lineNumber, inStrictContext); + } + + template PropertyNode* createGetterOrSetterProperty(int lineNumber, PropertyNode::Type type, const Identifier* name, ParameterNode* params, FunctionBodyNode* body, int openBracePos, int closeBracePos, int bodyStartLine, int bodyEndLine) + { + ASSERT(name); + body->setLoc(bodyStartLine, bodyEndLine); + return new (m_globalData) PropertyNode(m_globalData, *name, new (m_globalData) FuncExprNode(lineNumber, m_globalData->propertyNames->nullIdentifier, body, m_sourceCode->subExpression(openBracePos, closeBracePos, bodyStartLine), params), type); + } + + + ArgumentsNode* createArguments() { return new (m_globalData) ArgumentsNode(); } + ArgumentsNode* createArguments(ArgumentListNode* args) { return new (m_globalData) ArgumentsNode(args); } + ArgumentListNode* createArgumentsList(int lineNumber, ExpressionNode* arg) { return new (m_globalData) ArgumentListNode(lineNumber, arg); } + ArgumentListNode* createArgumentsList(int lineNumber, ArgumentListNode* args, ExpressionNode* arg) { return new (m_globalData) ArgumentListNode(lineNumber, args, arg); } + + template PropertyNode* createProperty(const Identifier* propertyName, ExpressionNode* node, PropertyNode::Type type) { return new (m_globalData) PropertyNode(m_globalData, *propertyName, node, type); } + template PropertyNode* createProperty(JSGlobalData*, double propertyName, ExpressionNode* node, PropertyNode::Type type) { return new (m_globalData) PropertyNode(m_globalData, propertyName, node, type); } + PropertyListNode* createPropertyList(int lineNumber, PropertyNode* property) { return new (m_globalData) PropertyListNode(lineNumber, property); } + PropertyListNode* createPropertyList(int lineNumber, PropertyNode* property, PropertyListNode* tail) { return new (m_globalData) PropertyListNode(lineNumber, property, tail); } + + ElementNode* createElementList(int elisions, ExpressionNode* expr) { return new (m_globalData) ElementNode(elisions, expr); } + ElementNode* createElementList(ElementNode* elems, int elisions, ExpressionNode* expr) { return new (m_globalData) ElementNode(elems, elisions, expr); } + + ParameterNode* createFormalParameterList(const Identifier& ident) { return new (m_globalData) ParameterNode(ident); } + ParameterNode* createFormalParameterList(ParameterNode* list, const Identifier& ident) { return new (m_globalData) ParameterNode(list, ident); } + + CaseClauseNode* createClause(ExpressionNode* expr, JSC::SourceElements* statements) { return new (m_globalData) CaseClauseNode(expr, statements); } + ClauseListNode* createClauseList(CaseClauseNode* clause) { return new (m_globalData) ClauseListNode(clause); } + ClauseListNode* createClauseList(ClauseListNode* tail, CaseClauseNode* clause) { return new (m_globalData) ClauseListNode(tail, clause); } + + void setUsesArguments(FunctionBodyNode* node) { node->setUsesArguments(); } + + StatementNode* createFuncDeclStatement(int lineNumber, const Identifier* name, FunctionBodyNode* body, ParameterNode* parameters, int openBracePos, int closeBracePos, int bodyStartLine, int bodyEndLine) + { + FuncDeclNode* decl = new (m_globalData) FuncDeclNode(lineNumber, *name, body, m_sourceCode->subExpression(openBracePos, closeBracePos, bodyStartLine), parameters); + if (*name == m_globalData->propertyNames->arguments) + usesArguments(); + m_scope.m_funcDeclarations->data.append(decl->body()); + body->setLoc(bodyStartLine, bodyEndLine); + return decl; + } + + StatementNode* createBlockStatement(int lineNumber, JSC::SourceElements* elements, int startLine, int endLine) + { + BlockNode* block = new (m_globalData) BlockNode(lineNumber, elements); + block->setLoc(startLine, endLine); + return block; + } + + StatementNode* createExprStatement(int lineNumber, ExpressionNode* expr, int start, int end) + { + ExprStatementNode* result = new (m_globalData) ExprStatementNode(lineNumber, expr); + result->setLoc(start, end); + return result; + } + + StatementNode* createIfStatement(int lineNumber, ExpressionNode* condition, StatementNode* trueBlock, int start, int end) + { + IfNode* result = new (m_globalData) IfNode(lineNumber, condition, trueBlock); + result->setLoc(start, end); + return result; + } + + StatementNode* createIfStatement(int lineNumber, ExpressionNode* condition, StatementNode* trueBlock, StatementNode* falseBlock, int start, int end) + { + IfNode* result = new (m_globalData) IfElseNode(lineNumber, condition, trueBlock, falseBlock); + result->setLoc(start, end); + return result; + } + + StatementNode* createForLoop(int lineNumber, ExpressionNode* initializer, ExpressionNode* condition, ExpressionNode* iter, StatementNode* statements, bool b, int start, int end) + { + ForNode* result = new (m_globalData) ForNode(lineNumber, initializer, condition, iter, statements, b); + result->setLoc(start, end); + return result; + } + + StatementNode* createForInLoop(int lineNumber, const Identifier* ident, ExpressionNode* initializer, ExpressionNode* iter, StatementNode* statements, int start, int divot, int end, int initStart, int initEnd, int startLine, int endLine) + { + ForInNode* result = new (m_globalData) ForInNode(m_globalData, lineNumber, *ident, initializer, iter, statements, initStart, initStart - start, initEnd - initStart); + result->setLoc(startLine, endLine); + setExceptionLocation(result, start, divot + 1, end); + return result; + } + + StatementNode* createForInLoop(int lineNumber, ExpressionNode* lhs, ExpressionNode* iter, StatementNode* statements, int eStart, int eDivot, int eEnd, int start, int end) + { + ForInNode* result = new (m_globalData) ForInNode(m_globalData, lineNumber, lhs, iter, statements); + result->setLoc(start, end); + setExceptionLocation(result, eStart, eDivot, eEnd); + return result; + } + + StatementNode* createEmptyStatement(int lineNumber) { return new (m_globalData) EmptyStatementNode(lineNumber); } + + StatementNode* createVarStatement(int lineNumber, ExpressionNode* expr, int start, int end) + { + StatementNode* result; + if (!expr) + result = new (m_globalData) EmptyStatementNode(lineNumber); + else + result = new (m_globalData) VarStatementNode(lineNumber, expr); + result->setLoc(start, end); + return result; + } + + StatementNode* createReturnStatement(int lineNumber, ExpressionNode* expression, int eStart, int eEnd, int startLine, int endLine) + { + ReturnNode* result = new (m_globalData) ReturnNode(lineNumber, expression); + setExceptionLocation(result, eStart, eEnd, eEnd); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createBreakStatement(int lineNumber, int eStart, int eEnd, int startLine, int endLine) + { + BreakNode* result = new (m_globalData) BreakNode(m_globalData, lineNumber); + setExceptionLocation(result, eStart, eEnd, eEnd); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createBreakStatement(int lineNumber, const Identifier* ident, int eStart, int eEnd, int startLine, int endLine) + { + BreakNode* result = new (m_globalData) BreakNode(lineNumber, *ident); + setExceptionLocation(result, eStart, eEnd, eEnd); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createContinueStatement(int lineNumber, int eStart, int eEnd, int startLine, int endLine) + { + ContinueNode* result = new (m_globalData) ContinueNode(m_globalData, lineNumber); + setExceptionLocation(result, eStart, eEnd, eEnd); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createContinueStatement(int lineNumber, const Identifier* ident, int eStart, int eEnd, int startLine, int endLine) + { + ContinueNode* result = new (m_globalData) ContinueNode(lineNumber, *ident); + setExceptionLocation(result, eStart, eEnd, eEnd); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createTryStatement(int lineNumber, StatementNode* tryBlock, const Identifier* ident, bool catchHasEval, StatementNode* catchBlock, StatementNode* finallyBlock, int startLine, int endLine) + { + TryNode* result = new (m_globalData) TryNode(lineNumber, tryBlock, *ident, catchHasEval, catchBlock, finallyBlock); + if (catchBlock) + usesCatch(); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createSwitchStatement(int lineNumber, ExpressionNode* expr, ClauseListNode* firstClauses, CaseClauseNode* defaultClause, ClauseListNode* secondClauses, int startLine, int endLine) + { + CaseBlockNode* cases = new (m_globalData) CaseBlockNode(firstClauses, defaultClause, secondClauses); + SwitchNode* result = new (m_globalData) SwitchNode(lineNumber, expr, cases); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createWhileStatement(int lineNumber, ExpressionNode* expr, StatementNode* statement, int startLine, int endLine) + { + WhileNode* result = new (m_globalData) WhileNode(lineNumber, expr, statement); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createDoWhileStatement(int lineNumber, StatementNode* statement, ExpressionNode* expr, int startLine, int endLine) + { + DoWhileNode* result = new (m_globalData) DoWhileNode(lineNumber, statement, expr); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createLabelStatement(int lineNumber, const Identifier* ident, StatementNode* statement, int start, int end) + { + LabelNode* result = new (m_globalData) LabelNode(lineNumber, *ident, statement); + setExceptionLocation(result, start, end, end); + return result; + } + + StatementNode* createWithStatement(int lineNumber, ExpressionNode* expr, StatementNode* statement, int start, int end, int startLine, int endLine) + { + usesWith(); + WithNode* result = new (m_globalData) WithNode(lineNumber, expr, statement, end, end - start); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createThrowStatement(int lineNumber, ExpressionNode* expr, int start, int end, int startLine, int endLine) + { + ThrowNode* result = new (m_globalData) ThrowNode(lineNumber, expr); + result->setLoc(startLine, endLine); + setExceptionLocation(result, start, end, end); + return result; + } + + StatementNode* createDebugger(int lineNumber, int startLine, int endLine) + { + DebuggerStatementNode* result = new (m_globalData) DebuggerStatementNode(lineNumber); + result->setLoc(startLine, endLine); + return result; + } + + StatementNode* createConstStatement(int lineNumber, ConstDeclNode* decls, int startLine, int endLine) + { + ConstStatementNode* result = new (m_globalData) ConstStatementNode(lineNumber, decls); + result->setLoc(startLine, endLine); + return result; + } + + ConstDeclNode* appendConstDecl(int lineNumber, ConstDeclNode* tail, const Identifier* name, ExpressionNode* initializer) + { + ConstDeclNode* result = new (m_globalData) ConstDeclNode(lineNumber, *name, initializer); + if (tail) + tail->m_next = result; + return result; + } + + void appendStatement(JSC::SourceElements* elements, JSC::StatementNode* statement) + { + elements->append(statement); + } + + void addVar(const Identifier* ident, int attrs) + { + if (m_globalData->propertyNames->arguments == *ident) + usesArguments(); + m_scope.m_varDeclarations->data.append(std::make_pair(ident, attrs)); + } + + ExpressionNode* combineCommaNodes(int lineNumber, ExpressionNode* list, ExpressionNode* init) + { + if (!list) + return init; + if (list->isCommaNode()) { + static_cast(list)->append(init); + return list; + } + return new (m_globalData) CommaNode(lineNumber, list, init); + } + + int evalCount() const { return m_evalCount; } + + void appendBinaryExpressionInfo(int& operandStackDepth, ExpressionNode* current, int exprStart, int lhs, int rhs, bool hasAssignments) + { + operandStackDepth++; + m_binaryOperandStack.append(std::make_pair(current, BinaryOpInfo(exprStart, lhs, rhs, hasAssignments))); + } + + // Logic to handle datastructures used during parsing of binary expressions + void operatorStackPop(int& operatorStackDepth) + { + operatorStackDepth--; + m_binaryOperatorStack.removeLast(); + } + bool operatorStackHasHigherPrecedence(int&, int precedence) + { + return precedence <= m_binaryOperatorStack.last().second; + } + const BinaryOperand& getFromOperandStack(int i) { return m_binaryOperandStack[m_binaryOperandStack.size() + i]; } + void shrinkOperandStackBy(int& operandStackDepth, int amount) + { + operandStackDepth -= amount; + ASSERT(operandStackDepth >= 0); + m_binaryOperandStack.resize(m_binaryOperandStack.size() - amount); + } + void appendBinaryOperation(int lineNumber, int& operandStackDepth, int&, const BinaryOperand& lhs, const BinaryOperand& rhs) + { + operandStackDepth++; + m_binaryOperandStack.append(std::make_pair(makeBinaryNode(lineNumber, m_binaryOperatorStack.last().first, lhs, rhs), BinaryOpInfo(lhs.second, rhs.second))); + } + void operatorStackAppend(int& operatorStackDepth, int op, int precedence) + { + operatorStackDepth++; + m_binaryOperatorStack.append(std::make_pair(op, precedence)); + } + ExpressionNode* popOperandStack(int&) + { + ExpressionNode* result = m_binaryOperandStack.last().first; + m_binaryOperandStack.removeLast(); + return result; + } + + void appendUnaryToken(int& tokenStackDepth, int type, int start) + { + tokenStackDepth++; + m_unaryTokenStack.append(std::make_pair(type, start)); + } + + int unaryTokenStackLastType(int&) + { + return m_unaryTokenStack.last().first; + } + + int unaryTokenStackLastStart(int&) + { + return m_unaryTokenStack.last().second; + } + + void unaryTokenStackRemoveLast(int& tokenStackDepth) + { + tokenStackDepth--; + m_unaryTokenStack.removeLast(); + } + + void assignmentStackAppend(int& assignmentStackDepth, ExpressionNode* node, int start, int divot, int assignmentCount, Operator op) + { + assignmentStackDepth++; + m_assignmentInfoStack.append(AssignmentInfo(node, start, divot, assignmentCount, op)); + } + + ExpressionNode* createAssignment(int lineNumber, int& assignmentStackDepth, ExpressionNode* rhs, int initialAssignmentCount, int currentAssignmentCount, int lastTokenEnd) + { + ExpressionNode* result = makeAssignNode(lineNumber, m_assignmentInfoStack.last().m_node, m_assignmentInfoStack.last().m_op, rhs, m_assignmentInfoStack.last().m_initAssignments != initialAssignmentCount, m_assignmentInfoStack.last().m_initAssignments != currentAssignmentCount, m_assignmentInfoStack.last().m_start, m_assignmentInfoStack.last().m_divot + 1, lastTokenEnd); + m_assignmentInfoStack.removeLast(); + assignmentStackDepth--; + return result; + } + + const Identifier& getName(Property property) const { return property->name(); } + PropertyNode::Type getType(Property property) const { return property->type(); } + + bool isResolve(ExpressionNode* expr) const { return expr->isResolveNode(); } + +private: + struct Scope { + Scope(JSGlobalData* globalData) + : m_varDeclarations(new (globalData) ParserArenaData) + , m_funcDeclarations(new (globalData) ParserArenaData) + , m_features(0) + , m_numConstants(0) + { + } + ParserArenaData* m_varDeclarations; + ParserArenaData* m_funcDeclarations; + int m_features; + int m_numConstants; + }; + + static void setExceptionLocation(ThrowableExpressionData* node, unsigned start, unsigned divot, unsigned end) + { + node->setExceptionSourceCode(divot, divot - start, end - divot); + } + + void incConstants() { m_scope.m_numConstants++; } + void usesThis() { m_scope.m_features |= ThisFeature; } + void usesCatch() { m_scope.m_features |= CatchFeature; } + void usesClosures() { m_scope.m_features |= ClosureFeature; } + void usesArguments() { m_scope.m_features |= ArgumentsFeature; } + void usesAssignment() { m_scope.m_features |= AssignFeature; } + void usesWith() { m_scope.m_features |= WithFeature; } + void usesEval() + { + m_evalCount++; + m_scope.m_features |= EvalFeature; + } + ExpressionNode* createNumber(int lineNumber, double d) + { + return new (m_globalData) NumberNode(lineNumber, d); + } + + JSGlobalData* m_globalData; + SourceCode* m_sourceCode; + Scope m_scope; + Vector m_binaryOperandStack; + Vector m_assignmentInfoStack; + Vector, 10> m_binaryOperatorStack; + Vector, 10> m_unaryTokenStack; + int m_evalCount; +}; + +ExpressionNode* ASTBuilder::makeTypeOfNode(int lineNumber, ExpressionNode* expr) +{ + if (expr->isResolveNode()) { + ResolveNode* resolve = static_cast(expr); + return new (m_globalData) TypeOfResolveNode(lineNumber, resolve->identifier()); + } + return new (m_globalData) TypeOfValueNode(lineNumber, expr); +} + +ExpressionNode* ASTBuilder::makeDeleteNode(int lineNumber, ExpressionNode* expr, int start, int divot, int end) +{ + if (!expr->isLocation()) + return new (m_globalData) DeleteValueNode(lineNumber, expr); + if (expr->isResolveNode()) { + ResolveNode* resolve = static_cast(expr); + return new (m_globalData) DeleteResolveNode(lineNumber, resolve->identifier(), divot, divot - start, end - divot); + } + if (expr->isBracketAccessorNode()) { + BracketAccessorNode* bracket = static_cast(expr); + return new (m_globalData) DeleteBracketNode(lineNumber, bracket->base(), bracket->subscript(), divot, divot - start, end - divot); + } + ASSERT(expr->isDotAccessorNode()); + DotAccessorNode* dot = static_cast(expr); + return new (m_globalData) DeleteDotNode(lineNumber, dot->base(), dot->identifier(), divot, divot - start, end - divot); +} + +ExpressionNode* ASTBuilder::makeNegateNode(int lineNumber, ExpressionNode* n) +{ + if (n->isNumber()) { + NumberNode* numberNode = static_cast(n); + numberNode->setValue(-numberNode->value()); + return numberNode; + } + + return new (m_globalData) NegateNode(lineNumber, n); +} + +ExpressionNode* ASTBuilder::makeBitwiseNotNode(int lineNumber, ExpressionNode* expr) +{ + if (expr->isNumber()) + return createNumber(lineNumber, ~toInt32(static_cast(expr)->value())); + return new (m_globalData) BitwiseNotNode(lineNumber, expr); +} + +ExpressionNode* ASTBuilder::makeMultNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + expr1 = expr1->stripUnaryPlus(); + expr2 = expr2->stripUnaryPlus(); + + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, static_cast(expr1)->value() * static_cast(expr2)->value()); + + if (expr1->isNumber() && static_cast(expr1)->value() == 1) + return new (m_globalData) UnaryPlusNode(lineNumber, expr2); + + if (expr2->isNumber() && static_cast(expr2)->value() == 1) + return new (m_globalData) UnaryPlusNode(lineNumber, expr1); + + return new (m_globalData) MultNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeDivNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + expr1 = expr1->stripUnaryPlus(); + expr2 = expr2->stripUnaryPlus(); + + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, static_cast(expr1)->value() / static_cast(expr2)->value()); + return new (m_globalData) DivNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeModNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + expr1 = expr1->stripUnaryPlus(); + expr2 = expr2->stripUnaryPlus(); + + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, fmod(static_cast(expr1)->value(), static_cast(expr2)->value())); + return new (m_globalData) ModNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeAddNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, static_cast(expr1)->value() + static_cast(expr2)->value()); + return new (m_globalData) AddNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeSubNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + expr1 = expr1->stripUnaryPlus(); + expr2 = expr2->stripUnaryPlus(); + + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, static_cast(expr1)->value() - static_cast(expr2)->value()); + return new (m_globalData) SubNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeLeftShiftNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, toInt32(static_cast(expr1)->value()) << (toUInt32(static_cast(expr2)->value()) & 0x1f)); + return new (m_globalData) LeftShiftNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeRightShiftNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, toInt32(static_cast(expr1)->value()) >> (toUInt32(static_cast(expr2)->value()) & 0x1f)); + return new (m_globalData) RightShiftNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeURightShiftNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, toUInt32(static_cast(expr1)->value()) >> (toUInt32(static_cast(expr2)->value()) & 0x1f)); + return new (m_globalData) UnsignedRightShiftNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeBitOrNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, toInt32(static_cast(expr1)->value()) | toInt32(static_cast(expr2)->value())); + return new (m_globalData) BitOrNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeBitAndNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, toInt32(static_cast(expr1)->value()) & toInt32(static_cast(expr2)->value())); + return new (m_globalData) BitAndNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeBitXOrNode(int lineNumber, ExpressionNode* expr1, ExpressionNode* expr2, bool rightHasAssignments) +{ + if (expr1->isNumber() && expr2->isNumber()) + return createNumber(lineNumber, toInt32(static_cast(expr1)->value()) ^ toInt32(static_cast(expr2)->value())); + return new (m_globalData) BitXOrNode(lineNumber, expr1, expr2, rightHasAssignments); +} + +ExpressionNode* ASTBuilder::makeFunctionCallNode(int lineNumber, ExpressionNode* func, ArgumentsNode* args, int start, int divot, int end) +{ + if (!func->isLocation()) + return new (m_globalData) FunctionCallValueNode(lineNumber, func, args, divot, divot - start, end - divot); + if (func->isResolveNode()) { + ResolveNode* resolve = static_cast(func); + const Identifier& identifier = resolve->identifier(); + if (identifier == m_globalData->propertyNames->eval) { + usesEval(); + return new (m_globalData) EvalFunctionCallNode(lineNumber, args, divot, divot - start, end - divot); + } + return new (m_globalData) FunctionCallResolveNode(lineNumber, identifier, args, divot, divot - start, end - divot); + } + if (func->isBracketAccessorNode()) { + BracketAccessorNode* bracket = static_cast(func); + FunctionCallBracketNode* node = new (m_globalData) FunctionCallBracketNode(lineNumber, bracket->base(), bracket->subscript(), args, divot, divot - start, end - divot); + node->setSubexpressionInfo(bracket->divot(), bracket->endOffset()); + return node; + } + ASSERT(func->isDotAccessorNode()); + DotAccessorNode* dot = static_cast(func); + FunctionCallDotNode* node; + if (dot->identifier() == m_globalData->propertyNames->call) + node = new (m_globalData) CallFunctionCallDotNode(lineNumber, dot->base(), dot->identifier(), args, divot, divot - start, end - divot); + else if (dot->identifier() == m_globalData->propertyNames->apply) + node = new (m_globalData) ApplyFunctionCallDotNode(lineNumber, dot->base(), dot->identifier(), args, divot, divot - start, end - divot); + else + node = new (m_globalData) FunctionCallDotNode(lineNumber, dot->base(), dot->identifier(), args, divot, divot - start, end - divot); + node->setSubexpressionInfo(dot->divot(), dot->endOffset()); + return node; +} + +ExpressionNode* ASTBuilder::makeBinaryNode(int lineNumber, int token, pair lhs, pair rhs) +{ + switch (token) { + case OR: + return new (m_globalData) LogicalOpNode(lineNumber, lhs.first, rhs.first, OpLogicalOr); + + case AND: + return new (m_globalData) LogicalOpNode(lineNumber, lhs.first, rhs.first, OpLogicalAnd); + + case BITOR: + return makeBitOrNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case BITXOR: + return makeBitXOrNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case BITAND: + return makeBitAndNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case EQEQ: + return new (m_globalData) EqualNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case NE: + return new (m_globalData) NotEqualNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case STREQ: + return new (m_globalData) StrictEqualNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case STRNEQ: + return new (m_globalData) NotStrictEqualNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case LT: + return new (m_globalData) LessNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case GT: + return new (m_globalData) GreaterNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case LE: + return new (m_globalData) LessEqNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case GE: + return new (m_globalData) GreaterEqNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case INSTANCEOF: { + InstanceOfNode* node = new (m_globalData) InstanceOfNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + setExceptionLocation(node, lhs.second.start, rhs.second.start, rhs.second.end); + return node; + } + + case INTOKEN: { + InNode* node = new (m_globalData) InNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + setExceptionLocation(node, lhs.second.start, rhs.second.start, rhs.second.end); + return node; + } + + case LSHIFT: + return makeLeftShiftNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case RSHIFT: + return makeRightShiftNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case URSHIFT: + return makeURightShiftNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case PLUS: + return makeAddNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case MINUS: + return makeSubNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case TIMES: + return makeMultNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case DIVIDE: + return makeDivNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + + case MOD: + return makeModNode(lineNumber, lhs.first, rhs.first, rhs.second.hasAssignment); + } + CRASH(); + return 0; +} + +ExpressionNode* ASTBuilder::makeAssignNode(int lineNumber, ExpressionNode* loc, Operator op, ExpressionNode* expr, bool locHasAssignments, bool exprHasAssignments, int start, int divot, int end) +{ + usesAssignment(); + if (!loc->isLocation()) + return new (m_globalData) AssignErrorNode(lineNumber, loc, op, expr, divot, divot - start, end - divot); + + if (loc->isResolveNode()) { + ResolveNode* resolve = static_cast(loc); + if (op == OpEqual) { + AssignResolveNode* node = new (m_globalData) AssignResolveNode(lineNumber, resolve->identifier(), expr, exprHasAssignments); + setExceptionLocation(node, start, divot, end); + return node; + } + return new (m_globalData) ReadModifyResolveNode(lineNumber, resolve->identifier(), op, expr, exprHasAssignments, divot, divot - start, end - divot); + } + if (loc->isBracketAccessorNode()) { + BracketAccessorNode* bracket = static_cast(loc); + if (op == OpEqual) + return new (m_globalData) AssignBracketNode(lineNumber, bracket->base(), bracket->subscript(), expr, locHasAssignments, exprHasAssignments, bracket->divot(), bracket->divot() - start, end - bracket->divot()); + ReadModifyBracketNode* node = new (m_globalData) ReadModifyBracketNode(lineNumber, bracket->base(), bracket->subscript(), op, expr, locHasAssignments, exprHasAssignments, divot, divot - start, end - divot); + node->setSubexpressionInfo(bracket->divot(), bracket->endOffset()); + return node; + } + ASSERT(loc->isDotAccessorNode()); + DotAccessorNode* dot = static_cast(loc); + if (op == OpEqual) + return new (m_globalData) AssignDotNode(lineNumber, dot->base(), dot->identifier(), expr, exprHasAssignments, dot->divot(), dot->divot() - start, end - dot->divot()); + + ReadModifyDotNode* node = new (m_globalData) ReadModifyDotNode(lineNumber, dot->base(), dot->identifier(), op, expr, exprHasAssignments, divot, divot - start, end - divot); + node->setSubexpressionInfo(dot->divot(), dot->endOffset()); + return node; +} + +ExpressionNode* ASTBuilder::makePrefixNode(int lineNumber, ExpressionNode* expr, Operator op, int start, int divot, int end) +{ + usesAssignment(); + if (!expr->isLocation()) + return new (m_globalData) PrefixErrorNode(lineNumber, expr, op, divot, divot - start, end - divot); + + if (expr->isResolveNode()) { + ResolveNode* resolve = static_cast(expr); + return new (m_globalData) PrefixResolveNode(lineNumber, resolve->identifier(), op, divot, divot - start, end - divot); + } + if (expr->isBracketAccessorNode()) { + BracketAccessorNode* bracket = static_cast(expr); + PrefixBracketNode* node = new (m_globalData) PrefixBracketNode(lineNumber, bracket->base(), bracket->subscript(), op, divot, divot - start, end - divot); + node->setSubexpressionInfo(bracket->divot(), bracket->startOffset()); + return node; + } + ASSERT(expr->isDotAccessorNode()); + DotAccessorNode* dot = static_cast(expr); + PrefixDotNode* node = new (m_globalData) PrefixDotNode(lineNumber, dot->base(), dot->identifier(), op, divot, divot - start, end - divot); + node->setSubexpressionInfo(dot->divot(), dot->startOffset()); + return node; +} + +ExpressionNode* ASTBuilder::makePostfixNode(int lineNumber, ExpressionNode* expr, Operator op, int start, int divot, int end) +{ + usesAssignment(); + if (!expr->isLocation()) + return new (m_globalData) PostfixErrorNode(lineNumber, expr, op, divot, divot - start, end - divot); + + if (expr->isResolveNode()) { + ResolveNode* resolve = static_cast(expr); + return new (m_globalData) PostfixResolveNode(lineNumber, resolve->identifier(), op, divot, divot - start, end - divot); + } + if (expr->isBracketAccessorNode()) { + BracketAccessorNode* bracket = static_cast(expr); + PostfixBracketNode* node = new (m_globalData) PostfixBracketNode(lineNumber, bracket->base(), bracket->subscript(), op, divot, divot - start, end - divot); + node->setSubexpressionInfo(bracket->divot(), bracket->endOffset()); + return node; + + } + ASSERT(expr->isDotAccessorNode()); + DotAccessorNode* dot = static_cast(expr); + PostfixDotNode* node = new (m_globalData) PostfixDotNode(lineNumber, dot->base(), dot->identifier(), op, divot, divot - start, end - divot); + node->setSubexpressionInfo(dot->divot(), dot->endOffset()); + return node; +} + +} + +#endif diff --git a/Source/JavaScriptCore/parser/Keywords.table b/Source/JavaScriptCore/parser/Keywords.table new file mode 100644 index 000000000..fab0e23cc --- /dev/null +++ b/Source/JavaScriptCore/parser/Keywords.table @@ -0,0 +1,58 @@ +# main keywords +@begin mainTable 47 + +# types +null NULLTOKEN +true TRUETOKEN +false FALSETOKEN + +# keywords +break BREAK +case CASE +catch CATCH +const CONSTTOKEN +default DEFAULT +finally FINALLY +for FOR +instanceof INSTANCEOF +new NEW +var VAR +continue CONTINUE +function FUNCTION +return RETURN +void VOIDTOKEN +delete DELETETOKEN +if IF +this THISTOKEN +do DO +while WHILE +else ELSE +in INTOKEN +switch SWITCH +throw THROW +try TRY +typeof TYPEOF +with WITH +debugger DEBUGGER + +# reserved for future use +class RESERVED +enum RESERVED +export RESERVED +extends RESERVED +import RESERVED +super RESERVED + +# reserved for future use in strict code +implements RESERVED_IF_STRICT +interface RESERVED_IF_STRICT +let RESERVED_IF_STRICT +package RESERVED_IF_STRICT +private RESERVED_IF_STRICT +protected RESERVED_IF_STRICT +public RESERVED_IF_STRICT +static RESERVED_IF_STRICT +yield RESERVED_IF_STRICT + +@end + diff --git a/Source/JavaScriptCore/parser/Lexer.cpp b/Source/JavaScriptCore/parser/Lexer.cpp new file mode 100644 index 000000000..e38b52480 --- /dev/null +++ b/Source/JavaScriptCore/parser/Lexer.cpp @@ -0,0 +1,1543 @@ +/* + * Copyright (C) 1999-2000 Harri Porten (porten@kde.org) + * Copyright (C) 2006, 2007, 2008, 2009 Apple Inc. All Rights Reserved. + * Copyright (C) 2007 Cameron Zwarich (cwzwarich@uwaterloo.ca) + * Copyright (C) 2010 Zoltan Herczeg (zherczeg@inf.u-szeged.hu) + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Library General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Library General Public License for more details. + * + * You should have received a copy of the GNU Library General Public License + * along with this library; see the file COPYING.LIB. If not, write to + * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, + * Boston, MA 02110-1301, USA. + * + */ + +#include "config.h" +#include "Lexer.h" + +#include "JSFunction.h" + +#include "JSGlobalObjectFunctions.h" +#include "Identifier.h" +#include "NodeInfo.h" +#include "Nodes.h" +#include "dtoa.h" +#include +#include +#include +#include + +using namespace WTF; +using namespace Unicode; + +#include "KeywordLookup.h" +#include "Lexer.lut.h" +#include "Parser.h" + +namespace JSC { + +Keywords::Keywords(JSGlobalData* globalData) + : m_globalData(globalData) + , m_keywordTable(JSC::mainTable) +{ +} + +enum CharacterType { + // Types for the main switch + + // The first three types are fixed, and also used for identifying + // ASCII alpha and alphanumeric characters (see isIdentStart and isIdentPart). + CharacterIdentifierStart, + CharacterZero, + CharacterNumber, + + CharacterInvalid, + CharacterLineTerminator, + CharacterExclamationMark, + CharacterOpenParen, + CharacterCloseParen, + CharacterOpenBracket, + CharacterCloseBracket, + CharacterComma, + CharacterColon, + CharacterQuestion, + CharacterTilde, + CharacterQuote, + CharacterDot, + CharacterSlash, + CharacterBackSlash, + CharacterSemicolon, + CharacterOpenBrace, + CharacterCloseBrace, + + CharacterAdd, + CharacterSub, + CharacterMultiply, + CharacterModulo, + CharacterAnd, + CharacterXor, + CharacterOr, + CharacterLess, + CharacterGreater, + CharacterEqual, + + // Other types (only one so far) + CharacterWhiteSpace, +}; + +// 128 ASCII codes +static const unsigned short typesOfASCIICharacters[128] = { +/* 0 - Null */ CharacterInvalid, +/* 1 - Start of Heading */ CharacterInvalid, +/* 2 - Start of Text */ CharacterInvalid, +/* 3 - End of Text */ CharacterInvalid, +/* 4 - End of Transm. */ CharacterInvalid, +/* 5 - Enquiry */ CharacterInvalid, +/* 6 - Acknowledgment */ CharacterInvalid, +/* 7 - Bell */ CharacterInvalid, +/* 8 - Back Space */ CharacterInvalid, +/* 9 - Horizontal Tab */ CharacterWhiteSpace, +/* 10 - Line Feed */ CharacterLineTerminator, +/* 11 - Vertical Tab */ CharacterWhiteSpace, +/* 12 - Form Feed */ CharacterWhiteSpace, +/* 13 - Carriage Return */ CharacterLineTerminator, +/* 14 - Shift Out */ CharacterInvalid, +/* 15 - Shift In */ CharacterInvalid, +/* 16 - Data Line Escape */ CharacterInvalid, +/* 17 - Device Control 1 */ CharacterInvalid, +/* 18 - Device Control 2 */ CharacterInvalid, +/* 19 - Device Control 3 */ CharacterInvalid, +/* 20 - Device Control 4 */ CharacterInvalid, +/* 21 - Negative Ack. */ CharacterInvalid, +/* 22 - Synchronous Idle */ CharacterInvalid, +/* 23 - End of Transmit */ CharacterInvalid, +/* 24 - Cancel */ CharacterInvalid, +/* 25 - End of Medium */ CharacterInvalid, +/* 26 - Substitute */ CharacterInvalid, +/* 27 - Escape */ CharacterInvalid, +/* 28 - File Separator */ CharacterInvalid, +/* 29 - Group Separator */ CharacterInvalid, +/* 30 - Record Separator */ CharacterInvalid, +/* 31 - Unit Separator */ CharacterInvalid, +/* 32 - Space */ CharacterWhiteSpace, +/* 33 - ! */ CharacterExclamationMark, +/* 34 - " */ CharacterQuote, +/* 35 - # */ CharacterInvalid, +/* 36 - $ */ CharacterIdentifierStart, +/* 37 - % */ CharacterModulo, +/* 38 - & */ CharacterAnd, +/* 39 - ' */ CharacterQuote, +/* 40 - ( */ CharacterOpenParen, +/* 41 - ) */ CharacterCloseParen, +/* 42 - * */ CharacterMultiply, +/* 43 - + */ CharacterAdd, +/* 44 - , */ CharacterComma, +/* 45 - - */ CharacterSub, +/* 46 - . */ CharacterDot, +/* 47 - / */ CharacterSlash, +/* 48 - 0 */ CharacterZero, +/* 49 - 1 */ CharacterNumber, +/* 50 - 2 */ CharacterNumber, +/* 51 - 3 */ CharacterNumber, +/* 52 - 4 */ CharacterNumber, +/* 53 - 5 */ CharacterNumber, +/* 54 - 6 */ CharacterNumber, +/* 55 - 7 */ CharacterNumber, +/* 56 - 8 */ CharacterNumber, +/* 57 - 9 */ CharacterNumber, +/* 58 - : */ CharacterColon, +/* 59 - ; */ CharacterSemicolon, +/* 60 - < */ CharacterLess, +/* 61 - = */ CharacterEqual, +/* 62 - > */ CharacterGreater, +/* 63 - ? */ CharacterQuestion, +/* 64 - @ */ CharacterInvalid, +/* 65 - A */ CharacterIdentifierStart, +/* 66 - B */ CharacterIdentifierStart, +/* 67 - C */ CharacterIdentifierStart, +/* 68 - D */ CharacterIdentifierStart, +/* 69 - E */ CharacterIdentifierStart, +/* 70 - F */ CharacterIdentifierStart, +/* 71 - G */ CharacterIdentifierStart, +/* 72 - H */ CharacterIdentifierStart, +/* 73 - I */ CharacterIdentifierStart, +/* 74 - J */ CharacterIdentifierStart, +/* 75 - K */ CharacterIdentifierStart, +/* 76 - L */ CharacterIdentifierStart, +/* 77 - M */ CharacterIdentifierStart, +/* 78 - N */ CharacterIdentifierStart, +/* 79 - O */ CharacterIdentifierStart, +/* 80 - P */ CharacterIdentifierStart, +/* 81 - Q */ CharacterIdentifierStart, +/* 82 - R */ CharacterIdentifierStart, +/* 83 - S */ CharacterIdentifierStart, +/* 84 - T */ CharacterIdentifierStart, +/* 85 - U */ CharacterIdentifierStart, +/* 86 - V */ CharacterIdentifierStart, +/* 87 - W */ CharacterIdentifierStart, +/* 88 - X */ CharacterIdentifierStart, +/* 89 - Y */ CharacterIdentifierStart, +/* 90 - Z */ CharacterIdentifierStart, +/* 91 - [ */ CharacterOpenBracket, +/* 92 - \ */ CharacterBackSlash, +/* 93 - ] */ CharacterCloseBracket, +/* 94 - ^ */ CharacterXor, +/* 95 - _ */ CharacterIdentifierStart, +/* 96 - ` */ CharacterInvalid, +/* 97 - a */ CharacterIdentifierStart, +/* 98 - b */ CharacterIdentifierStart, +/* 99 - c */ CharacterIdentifierStart, +/* 100 - d */ CharacterIdentifierStart, +/* 101 - e */ CharacterIdentifierStart, +/* 102 - f */ CharacterIdentifierStart, +/* 103 - g */ CharacterIdentifierStart, +/* 104 - h */ CharacterIdentifierStart, +/* 105 - i */ CharacterIdentifierStart, +/* 106 - j */ CharacterIdentifierStart, +/* 107 - k */ CharacterIdentifierStart, +/* 108 - l */ CharacterIdentifierStart, +/* 109 - m */ CharacterIdentifierStart, +/* 110 - n */ CharacterIdentifierStart, +/* 111 - o */ CharacterIdentifierStart, +/* 112 - p */ CharacterIdentifierStart, +/* 113 - q */ CharacterIdentifierStart, +/* 114 - r */ CharacterIdentifierStart, +/* 115 - s */ CharacterIdentifierStart, +/* 116 - t */ CharacterIdentifierStart, +/* 117 - u */ CharacterIdentifierStart, +/* 118 - v */ CharacterIdentifierStart, +/* 119 - w */ CharacterIdentifierStart, +/* 120 - x */ CharacterIdentifierStart, +/* 121 - y */ CharacterIdentifierStart, +/* 122 - z */ CharacterIdentifierStart, +/* 123 - { */ CharacterOpenBrace, +/* 124 - | */ CharacterOr, +/* 125 - } */ CharacterCloseBrace, +/* 126 - ~ */ CharacterTilde, +/* 127 - Delete */ CharacterInvalid, +}; + +template +Lexer::Lexer(JSGlobalData* globalData) + : m_isReparsing(false) + , m_globalData(globalData) +{ +} + +template +Lexer::~Lexer() +{ +} + +template +UString Lexer::getInvalidCharMessage() +{ + switch (m_current) { + case 0: + return "Invalid character: '\\0'"; + case 10: + return "Invalid character: '\\n'"; + case 11: + return "Invalid character: '\\v'"; + case 13: + return "Invalid character: '\\r'"; + case 35: + return "Invalid character: '#'"; + case 64: + return "Invalid character: '@'"; + case 96: + return "Invalid character: '`'"; + default: + return String::format("Invalid character '\\u%04u'", m_current).impl(); + } +} + +template +ALWAYS_INLINE const T* Lexer::currentCharacter() const +{ + ASSERT(m_code <= m_codeEnd); + return m_code; +} + +template +void Lexer::setCode(const SourceCode& source, ParserArena* arena) +{ + m_arena = &arena->identifierArena(); + + m_lineNumber = source.firstLine(); + m_delimited = false; + m_lastToken = -1; + + const StringImpl* sourceString = source.provider()->data(); + + if (sourceString) + setCodeStart(sourceString); + else + m_codeStart = 0; + + m_source = &source; + m_code = m_codeStart + source.startOffset(); + m_codeEnd = m_codeStart + source.endOffset(); + m_error = false; + m_atLineStart = true; + m_lexErrorMessage = UString(); + + m_buffer8.reserveInitialCapacity(initialReadBufferCapacity); + m_buffer16.reserveInitialCapacity((m_codeEnd - m_code) / 2); + + if (LIKELY(m_code < m_codeEnd)) + m_current = *m_code; + else + m_current = -1; + ASSERT(currentOffset() == source.startOffset()); +} + +template +template ALWAYS_INLINE void Lexer::internalShift() +{ + m_code += shiftAmount; + m_current = *m_code; +} + +template +ALWAYS_INLINE void Lexer::shift() +{ + // Faster than an if-else sequence + ASSERT(m_current != -1); + m_current = -1; + m_code++; + if (LIKELY(m_code < m_codeEnd)) + m_current = *m_code; +} + +template +ALWAYS_INLINE int Lexer::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 +int Lexer::getUnicodeCharacter() +{ + int char1 = peek(1); + int char2 = peek(2); + int char3 = peek(3); + + if (UNLIKELY(!isASCIIHexDigit(m_current) || !isASCIIHexDigit(char1) || !isASCIIHexDigit(char2) || !isASCIIHexDigit(char3))) + return -1; + + int result = convertUnicode(m_current, char1, char2, char3); + shift(); + shift(); + shift(); + shift(); + return result; +} + +template +void Lexer::shiftLineTerminator() +{ + ASSERT(isLineTerminator(m_current)); + + int m_prev = m_current; + shift(); + + // Allow both CRLF and LFCR. + if (m_prev + m_current == '\n' + '\r') + shift(); + + ++m_lineNumber; +} + +template +ALWAYS_INLINE bool Lexer::lastTokenWasRestrKeyword() const +{ + return m_lastToken == CONTINUE || m_lastToken == BREAK || m_lastToken == RETURN || m_lastToken == THROW; +} + +static NEVER_INLINE bool isNonASCIIIdentStart(int c) +{ + return category(c) & (Letter_Uppercase | Letter_Lowercase | Letter_Titlecase | Letter_Modifier | Letter_Other); +} + +static inline bool isIdentStart(int c) +{ + return isASCII(c) ? typesOfASCIICharacters[c] == CharacterIdentifierStart : isNonASCIIIdentStart(c); +} + +static NEVER_INLINE bool isNonASCIIIdentPart(int c) +{ + return category(c) & (Letter_Uppercase | Letter_Lowercase | Letter_Titlecase | Letter_Modifier | Letter_Other + | Mark_NonSpacing | Mark_SpacingCombining | Number_DecimalDigit | Punctuation_Connector); +} + +static ALWAYS_INLINE bool isIdentPart(int c) +{ + // Character types are divided into two groups depending on whether they can be part of an + // identifier or not. Those whose type value is less or equal than CharacterNumber can be + // part of an identifier. (See the CharacterType definition for more details.) + return isASCII(c) ? typesOfASCIICharacters[c] <= CharacterNumber : isNonASCIIIdentPart(c); +} + +static inline int singleEscape(int c) +{ + switch (c) { + case 'b': + return 0x08; + case 't': + return 0x09; + case 'n': + return 0x0A; + case 'v': + return 0x0B; + case 'f': + return 0x0C; + case 'r': + return 0x0D; + case '\\': + return '\\'; + case '\'': + return '\''; + case '"': + return '"'; + default: + return 0; + } +} + +template +inline void Lexer::record8(int c) +{ + ASSERT(c >= 0); + ASSERT(c <= 0xFF); + m_buffer8.append(static_cast(c)); +} + +template +inline void assertCharIsIn8BitRange(T c) +{ + UNUSED_PARAM(c); + ASSERT(c >= 0); + ASSERT(c <= 0xFF); +} + +template <> +inline void assertCharIsIn8BitRange(UChar c) +{ + UNUSED_PARAM(c); + ASSERT(c <= 0xFF); +} + +template <> +inline void assertCharIsIn8BitRange(LChar) +{ +} + +template +inline void Lexer::append8(const T* p, size_t length) +{ + size_t currentSize = m_buffer8.size(); + m_buffer8.grow(currentSize + length); + LChar* rawBuffer = m_buffer8.data() + currentSize; + + for (size_t i = 0; i < length; i++) { + T c = p[i]; + assertCharIsIn8BitRange(c); + rawBuffer[i] = c; + } +} + +template +inline void Lexer::append16(const LChar* p, size_t length) +{ + size_t currentSize = m_buffer16.size(); + m_buffer16.grow(currentSize + length); + UChar* rawBuffer = m_buffer16.data() + currentSize; + + for (size_t i = 0; i < length; i++) + rawBuffer[i] = p[i]; +} + +template +inline void Lexer::record16(T c) +{ + m_buffer16.append(c); +} + +template +inline void Lexer::record16(int c) +{ + ASSERT(c >= 0); + ASSERT(c <= USHRT_MAX); + m_buffer16.append(static_cast(c)); +} + +template <> + template ALWAYS_INLINE JSTokenType Lexer::parseIdentifier(JSTokenData* tokenData, unsigned lexerFlags, bool strictMode) +{ + const ptrdiff_t remaining = m_codeEnd - m_code; + if ((remaining >= maxTokenLength) && !(lexerFlags & LexerFlagsIgnoreReservedWords)) { + JSTokenType keyword = parseKeyword(tokenData); + if (keyword != IDENT) { + ASSERT((!shouldCreateIdentifier) || tokenData->ident); + return keyword == RESERVED_IF_STRICT && !strictMode ? IDENT : keyword; + } + } + + const LChar* identifierStart = currentCharacter(); + + while (isIdentPart(m_current)) + shift(); + + if (UNLIKELY(m_current == '\\')) { + setOffsetFromCharOffset(identifierStart); + return parseIdentifierSlowCase(tokenData, lexerFlags, strictMode); + } + + const Identifier* ident = 0; + + if (shouldCreateIdentifier) { + int identifierLength = currentCharacter() - identifierStart; + ident = makeIdentifier(identifierStart, identifierLength); + + tokenData->ident = ident; + } else + tokenData->ident = 0; + + m_delimited = false; + + if (UNLIKELY((remaining < maxTokenLength) && !(lexerFlags & LexerFlagsIgnoreReservedWords))) { + ASSERT(shouldCreateIdentifier); + if (remaining < maxTokenLength) { + const HashEntry* entry = m_globalData->keywords->getKeyword(*ident); + ASSERT((remaining < maxTokenLength) || !entry); + if (!entry) + return IDENT; + JSTokenType token = static_cast(entry->lexerValue()); + return (token != RESERVED_IF_STRICT) || strictMode ? token : IDENT; + } + return IDENT; + } + + return IDENT; +} + +template <> +template ALWAYS_INLINE JSTokenType Lexer::parseIdentifier(JSTokenData* tokenData, unsigned lexerFlags, bool strictMode) +{ + const ptrdiff_t remaining = m_codeEnd - m_code; + if ((remaining >= maxTokenLength) && !(lexerFlags & LexerFlagsIgnoreReservedWords)) { + JSTokenType keyword = parseKeyword(tokenData); + if (keyword != IDENT) { + ASSERT((!shouldCreateIdentifier) || tokenData->ident); + return keyword == RESERVED_IF_STRICT && !strictMode ? IDENT : keyword; + } + } + const UChar* identifierStart = currentCharacter(); + + UChar orAllChars = 0; + + while (isIdentPart(m_current)) { + orAllChars |= m_current; + shift(); + } + + if (UNLIKELY(m_current == '\\')) { + setOffsetFromCharOffset(identifierStart); + return parseIdentifierSlowCase(tokenData, lexerFlags, strictMode); + } + + bool isAll8Bit = false; + + if (!(orAllChars & ~0xff)) + isAll8Bit = true; + + const Identifier* ident = 0; + + if (shouldCreateIdentifier) { + int identifierLength = currentCharacter() - identifierStart; + if (isAll8Bit) + ident = makeIdentifierLCharFromUChar(identifierStart, identifierLength); + else + ident = makeIdentifier(identifierStart, identifierLength); + + tokenData->ident = ident; + } else + tokenData->ident = 0; + + m_delimited = false; + + if (UNLIKELY((remaining < maxTokenLength) && !(lexerFlags & LexerFlagsIgnoreReservedWords))) { + ASSERT(shouldCreateIdentifier); + if (remaining < maxTokenLength) { + const HashEntry* entry = m_globalData->keywords->getKeyword(*ident); + ASSERT((remaining < maxTokenLength) || !entry); + if (!entry) + return IDENT; + JSTokenType token = static_cast(entry->lexerValue()); + return (token != RESERVED_IF_STRICT) || strictMode ? token : IDENT; + } + return IDENT; + } + + return IDENT; +} + +template +template JSTokenType Lexer::parseIdentifierSlowCase(JSTokenData* tokenData, unsigned lexerFlags, bool strictMode) +{ + const ptrdiff_t remaining = m_codeEnd - m_code; + const T* identifierStart = currentCharacter(); + bool bufferRequired = false; + + while (true) { + if (LIKELY(isIdentPart(m_current))) { + shift(); + continue; + } + if (LIKELY(m_current != '\\')) + break; + + // \uXXXX unicode characters. + bufferRequired = true; + if (identifierStart != currentCharacter()) + m_buffer16.append(identifierStart, currentCharacter() - identifierStart); + shift(); + if (UNLIKELY(m_current != 'u')) + return ERRORTOK; + shift(); + int character = getUnicodeCharacter(); + if (UNLIKELY(character == -1)) + return ERRORTOK; + if (UNLIKELY(m_buffer16.size() ? !isIdentPart(character) : !isIdentStart(character))) + return ERRORTOK; + if (shouldCreateIdentifier) + record16(character); + identifierStart = currentCharacter(); + } + + int identifierLength; + const Identifier* ident = 0; + if (shouldCreateIdentifier) { + if (!bufferRequired) { + identifierLength = currentCharacter() - identifierStart; + ident = makeIdentifier(identifierStart, identifierLength); + } else { + if (identifierStart != currentCharacter()) + m_buffer16.append(identifierStart, currentCharacter() - identifierStart); + ident = makeIdentifier(m_buffer16.data(), m_buffer16.size()); + } + + tokenData->ident = ident; + } else + tokenData->ident = 0; + + m_delimited = false; + + if (LIKELY(!bufferRequired && !(lexerFlags & LexerFlagsIgnoreReservedWords))) { + ASSERT(shouldCreateIdentifier); + // Keywords must not be recognized if there was an \uXXXX in the identifier. + if (remaining < maxTokenLength) { + const HashEntry* entry = m_globalData->keywords->getKeyword(*ident); + ASSERT((remaining < maxTokenLength) || !entry); + if (!entry) + return IDENT; + JSTokenType token = static_cast(entry->lexerValue()); + return (token != RESERVED_IF_STRICT) || strictMode ? token : IDENT; + } + return IDENT; + } + + m_buffer16.resize(0); + return IDENT; +} + +template +template ALWAYS_INLINE bool Lexer::parseString(JSTokenData* tokenData, bool strictMode) +{ + int startingOffset = currentOffset(); + int startingLineNumber = lineNumber(); + int stringQuoteCharacter = m_current; + shift(); + + const T* stringStart = currentCharacter(); + + while (m_current != stringQuoteCharacter) { + if (UNLIKELY((m_current == '\\'))) { + if (stringStart != currentCharacter() && shouldBuildStrings) + append8(stringStart, currentCharacter() - stringStart); + shift(); + + int escape = singleEscape(m_current); + + // Most common escape sequences first + if (escape) { + if (shouldBuildStrings) + record8(escape); + shift(); + } else if (UNLIKELY(isLineTerminator(m_current))) + shiftLineTerminator(); + else if (m_current == 'x') { + shift(); + if (isASCIIHexDigit(m_current) && isASCIIHexDigit(peek(1))) { + int prev = m_current; + shift(); + if (shouldBuildStrings) + record8(convertHex(prev, m_current)); + shift(); + } else if (shouldBuildStrings) + record8('x'); + } else { + setOffset(startingOffset); + setLineNumber(startingLineNumber); + m_buffer8.resize(0); + return parseStringSlowCase(tokenData, strictMode); + } + stringStart = currentCharacter(); + continue; + } + + if (UNLIKELY(((m_current > 0xff) || (m_current < 0xe)))) { + setOffset(startingOffset); + setLineNumber(startingLineNumber); + m_buffer8.resize(0); + return parseStringSlowCase(tokenData, strictMode); + } + + shift(); + } + + if (currentCharacter() != stringStart && shouldBuildStrings) + append8(stringStart, currentCharacter() - stringStart); + if (shouldBuildStrings) { + tokenData->ident = makeIdentifier(m_buffer8.data(), m_buffer8.size()); + m_buffer8.resize(0); + } else + tokenData->ident = 0; + + return true; +} + +template +template bool Lexer::parseStringSlowCase(JSTokenData* tokenData, bool strictMode) +{ + int stringQuoteCharacter = m_current; + shift(); + + const T* stringStart = currentCharacter(); + + while (m_current != stringQuoteCharacter) { + if (UNLIKELY(m_current == '\\')) { + if (stringStart != currentCharacter() && shouldBuildStrings) + append16(stringStart, currentCharacter() - stringStart); + shift(); + + int escape = singleEscape(m_current); + + // Most common escape sequences first + if (escape) { + if (shouldBuildStrings) + record16(escape); + shift(); + } else if (UNLIKELY(isLineTerminator(m_current))) + shiftLineTerminator(); + else if (m_current == 'x') { + shift(); + if (isASCIIHexDigit(m_current) && isASCIIHexDigit(peek(1))) { + int prev = m_current; + shift(); + if (shouldBuildStrings) + record16(convertHex(prev, m_current)); + shift(); + } else if (shouldBuildStrings) + record16('x'); + } else if (m_current == 'u') { + shift(); + int character = getUnicodeCharacter(); + if (character != -1) { + if (shouldBuildStrings) + record16(character); + } else if (m_current == stringQuoteCharacter) { + if (shouldBuildStrings) + record16('u'); + } else { + m_lexErrorMessage = "\\u can only be followed by a Unicode character sequence"; + return false; + } + } else if (strictMode && isASCIIDigit(m_current)) { + // The only valid numeric escape in strict mode is '\0', and this must not be followed by a decimal digit. + int character1 = m_current; + shift(); + if (character1 != '0' || isASCIIDigit(m_current)) { + m_lexErrorMessage = "The only valid numeric escape in strict mode is '\\0'"; + return false; + } + if (shouldBuildStrings) + record16(0); + } else if (!strictMode && isASCIIOctalDigit(m_current)) { + // Octal character sequences + int character1 = m_current; + shift(); + if (isASCIIOctalDigit(m_current)) { + // Two octal characters + int character2 = m_current; + shift(); + if (character1 >= '0' && character1 <= '3' && isASCIIOctalDigit(m_current)) { + if (shouldBuildStrings) + record16((character1 - '0') * 64 + (character2 - '0') * 8 + m_current - '0'); + shift(); + } else { + if (shouldBuildStrings) + record16((character1 - '0') * 8 + character2 - '0'); + } + } else { + if (shouldBuildStrings) + record16(character1 - '0'); + } + } else if (m_current != -1) { + if (shouldBuildStrings) + record16(m_current); + shift(); + } else { + m_lexErrorMessage = "Unterminated string constant"; + return false; + } + + stringStart = currentCharacter(); + continue; + } + // Fast check for characters that require special handling. + // Catches -1, \n, \r, 0x2028, and 0x2029 as efficiently + // as possible, and lets through all common ASCII characters. + if (UNLIKELY(((static_cast(m_current) - 0xE) & 0x2000))) { + // New-line or end of input is not allowed + if (UNLIKELY(isLineTerminator(m_current)) || UNLIKELY(m_current == -1)) { + m_lexErrorMessage = "Unexpected EOF"; + return false; + } + // Anything else is just a normal character + } + shift(); + } + + if (currentCharacter() != stringStart && shouldBuildStrings) + append16(stringStart, currentCharacter() - stringStart); + if (shouldBuildStrings) + tokenData->ident = makeIdentifier(m_buffer16.data(), m_buffer16.size()); + else + tokenData->ident = 0; + + m_buffer16.resize(0); + return true; +} + +template +ALWAYS_INLINE void Lexer::parseHex(double& returnValue) +{ + // Optimization: most hexadecimal values fit into 4 bytes. + uint32_t hexValue = 0; + int maximumDigits = 7; + + // Shift out the 'x' prefix. + shift(); + + do { + hexValue = (hexValue << 4) + toASCIIHexValue(m_current); + shift(); + --maximumDigits; + } while (isASCIIHexDigit(m_current) && maximumDigits >= 0); + + if (maximumDigits >= 0) { + returnValue = hexValue; + return; + } + + // No more place in the hexValue buffer. + // The values are shifted out and placed into the m_buffer8 vector. + for (int i = 0; i < 8; ++i) { + int digit = hexValue >> 28; + if (digit < 10) + record8(digit + '0'); + else + record8(digit - 10 + 'a'); + hexValue <<= 4; + } + + while (isASCIIHexDigit(m_current)) { + record8(m_current); + shift(); + } + + returnValue = parseIntOverflow(m_buffer8.data(), m_buffer8.size(), 16); +} + +template +ALWAYS_INLINE bool Lexer::parseOctal(double& returnValue) +{ + // Optimization: most octal values fit into 4 bytes. + uint32_t octalValue = 0; + int maximumDigits = 9; + // Temporary buffer for the digits. Makes easier + // to reconstruct the input characters when needed. + LChar digits[10]; + + do { + octalValue = octalValue * 8 + (m_current - '0'); + digits[maximumDigits] = m_current; + shift(); + --maximumDigits; + } while (isASCIIOctalDigit(m_current) && maximumDigits >= 0); + + if (!isASCIIDigit(m_current) && maximumDigits >= 0) { + returnValue = octalValue; + return true; + } + + for (int i = 9; i > maximumDigits; --i) + record8(digits[i]); + + while (isASCIIOctalDigit(m_current)) { + record8(m_current); + shift(); + } + + if (isASCIIDigit(m_current)) + return false; + + returnValue = parseIntOverflow(m_buffer8.data(), m_buffer8.size(), 8); + return true; +} + +template +ALWAYS_INLINE bool Lexer::parseDecimal(double& returnValue) +{ + // Optimization: most decimal values fit into 4 bytes. + uint32_t decimalValue = 0; + + // Since parseOctal may be executed before parseDecimal, + // the m_buffer8 may hold ascii digits. + if (!m_buffer8.size()) { + int maximumDigits = 9; + // Temporary buffer for the digits. Makes easier + // to reconstruct the input characters when needed. + LChar digits[10]; + + do { + decimalValue = decimalValue * 10 + (m_current - '0'); + digits[maximumDigits] = m_current; + shift(); + --maximumDigits; + } while (isASCIIDigit(m_current) && maximumDigits >= 0); + + if (maximumDigits >= 0 && m_current != '.' && (m_current | 0x20) != 'e') { + returnValue = decimalValue; + return true; + } + + for (int i = 9; i > maximumDigits; --i) + record8(digits[i]); + } + + while (isASCIIDigit(m_current)) { + record8(m_current); + shift(); + } + + return false; +} + +template +ALWAYS_INLINE void Lexer::parseNumberAfterDecimalPoint() +{ + record8('.'); + while (isASCIIDigit(m_current)) { + record8(m_current); + shift(); + } +} + +template +ALWAYS_INLINE bool Lexer::parseNumberAfterExponentIndicator() +{ + record8('e'); + shift(); + if (m_current == '+' || m_current == '-') { + record8(m_current); + shift(); + } + + if (!isASCIIDigit(m_current)) + return false; + + do { + record8(m_current); + shift(); + } while (isASCIIDigit(m_current)); + return true; +} + +template +ALWAYS_INLINE bool Lexer::parseMultilineComment() +{ + while (true) { + while (UNLIKELY(m_current == '*')) { + shift(); + if (m_current == '/') { + shift(); + return true; + } + } + + if (UNLIKELY(m_current == -1)) + return false; + + if (isLineTerminator(m_current)) { + shiftLineTerminator(); + m_terminator = true; + } else + shift(); + } +} + +template +bool Lexer::nextTokenIsColon() +{ + const T* code = m_code; + while (code < m_codeEnd && (isWhiteSpace(*code) || isLineTerminator(*code))) + code++; + + return code < m_codeEnd && *code == ':'; +} + +template +JSTokenType Lexer::lex(JSTokenData* tokenData, JSTokenInfo* tokenInfo, unsigned lexerFlags, bool strictMode) +{ + ASSERT(!m_error); + ASSERT(m_buffer8.isEmpty()); + ASSERT(m_buffer16.isEmpty()); + + JSTokenType token = ERRORTOK; + m_terminator = false; + +start: + while (isWhiteSpace(m_current)) + shift(); + + int startOffset = currentOffset(); + + if (UNLIKELY(m_current == -1)) + return EOFTOK; + + m_delimited = false; + + CharacterType type; + if (LIKELY(isASCII(m_current))) + type = static_cast(typesOfASCIICharacters[m_current]); + else if (isNonASCIIIdentStart(m_current)) + type = CharacterIdentifierStart; + else if (isLineTerminator(m_current)) + type = CharacterLineTerminator; + else + type = CharacterInvalid; + + switch (type) { + case CharacterGreater: + shift(); + if (m_current == '>') { + shift(); + if (m_current == '>') { + shift(); + if (m_current == '=') { + shift(); + token = URSHIFTEQUAL; + break; + } + token = URSHIFT; + break; + } + if (m_current == '=') { + shift(); + token = RSHIFTEQUAL; + break; + } + token = RSHIFT; + break; + } + if (m_current == '=') { + shift(); + token = GE; + break; + } + token = GT; + break; + case CharacterEqual: + shift(); + if (m_current == '=') { + shift(); + if (m_current == '=') { + shift(); + token = STREQ; + break; + } + token = EQEQ; + break; + } + token = EQUAL; + break; + case CharacterLess: + shift(); + if (m_current == '!' && peek(1) == '-' && peek(2) == '-') { + //