Preparser extracted into separate files that can be compiled to a library.

src/preparser.h

 // Copyright 2010 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
@@ skipping 28 matching lines @@
 // and collects some information about the program along the way.
 // The grammar check is only performed in order to understand the program
 // sufficiently to deduce some information about it, that can be used
 // to speed up later parsing. Finding errors is not the goal of pre-parsing,
 // rather it is to speed up properly written and correct programs.
 // That means that contextual checks (like a label being declared where
 // it is used) are generally omitted.
 
 namespace i = v8::internal;
 
-enum StatementType {
-  kUnknownStatement
-};
-
-enum ExpressionType {
-  kUnknownExpression,
-  kIdentifierExpression,  // Used to detect labels.
-  kThisExpression,
-  kThisPropertyExpression
-};
-
-enum IdentifierType {
-  kUnknownIdentifier
-};
-
-enum SourceElementTypes {
-  kUnknownSourceElements
-};
-
-
-typedef int SourceElements;
-typedef int Expression;
-typedef int Statement;
-typedef int Identifier;
-typedef int Arguments;
-
-
 class PreParser {
  public:
-  PreParser() : scope_(NULL), allow_lazy_(true) { }
+  enum PreParseResult {
+    kPreParseStackOverflow,
+    kPreParseSuccess
+  };
+
   ~PreParser() { }
 
   // Pre-parse the program from the character stream; returns true on
   // success (even if parsing failed, the pre-parse data successfully
   // captured the syntax error), and false if a stack-overflow happened
   // during parsing.
-  bool PreParseProgram(i::JavaScriptScanner* scanner,
-                       i::ParserRecorder* log,
-                       bool allow_lazy) {
-    allow_lazy_ = allow_lazy;
-    scanner_ = scanner;
-    log_ = log;
-    Scope top_scope(&scope_, kTopLevelScope);
-    bool ok = true;
-    ParseSourceElements(i::Token::EOS, &ok);
-    bool stack_overflow = scanner_->stack_overflow();
-    if (!ok && !stack_overflow) {
-      ReportUnexpectedToken(scanner_->current_token());
-    }
-    return !stack_overflow;
+  static PreParseResult PreParseProgram(i::JavaScriptScanner* scanner,
+                                        i::ParserRecorder* log,
+                                        bool allow_lazy,
+                                        uintptr_t stack_limit) {
+    return PreParser(scanner, log, stack_limit, allow_lazy).PreParse();
   }
 
  private:
   enum ScopeType {
     kTopLevelScope,
     kFunctionScope
   };
 
+  // Types that allow us to recognize simple this-property assignments.
+  // A simple this-property assignment is a statement on the form
+  // "this.propertyName = {primitive constant or function parameter name);"
+  // where propertyName isn't "__proto__".
+  // The result is only relevant if the function body contains only
+  // simple this-property assignments.
+
+  enum StatementType {
+    kUnknownStatement
+  };
+
+  enum ExpressionType {
+    kUnknownExpression,
+    kIdentifierExpression,  // Used to detect labels.
+    kThisExpression,
+    kThisPropertyExpression
+  };
+
+  enum IdentifierType {
+    kUnknownIdentifier
+  };
+
+  enum SourceElementTypes {
+    kUnknownSourceElements
+  };
+
+  typedef int SourceElements;
+  typedef int Expression;
+  typedef int Statement;
+  typedef int Identifier;
+  typedef int Arguments;
+
   class Scope {
    public:
     Scope(Scope** variable, ScopeType type)
         : variable_(variable),
           prev_(*variable),
           type_(type),
           materialized_literal_count_(0),
           expected_properties_(0),
           with_nesting_count_(0) {
       *variable = this;
@@ skipping 10 matching lines @@
 
    private:
     Scope** const variable_;
     Scope* const prev_;
     const ScopeType type_;
     int materialized_literal_count_;
     int expected_properties_;
     int with_nesting_count_;
   };
 
-  // Types that allow us to recognize simple this-property assignments.
-  // A simple this-property assignment is a statement on the form
-  // "this.propertyName = {primitive constant or function parameter name);"
-  // where propertyName isn't "__proto__".
-  // The result is only relevant if the function body contains only
-  // simple this-property assignments.
+  // Private constructor only used in PreParseProgram.
+  PreParser(i::JavaScriptScanner* scanner,
+            i::ParserRecorder* log,
+            uintptr_t stack_limit,
+            bool allow_lazy)
+      : scanner_(scanner),
+        log_(log),
+        scope_(NULL),
+        stack_limit_(stack_limit),
+        stack_overflow_(false),
+        allow_lazy_(true) { }
+
+  // Preparse the program. Only called in PreParseProgram after creating
+  // the instance.
+  PreParseResult PreParse() {
+    Scope top_scope(&scope_, kTopLevelScope);
+    bool ok = true;
+    ParseSourceElements(i::Token::EOS, &ok);
+    if (stack_overflow_) return kPreParseStackOverflow;
+    if (!ok) {
+      ReportUnexpectedToken(scanner_->current_token());
+    }
+    return kPreParseSuccess;
+  }
 
   // Report syntax error
   void ReportUnexpectedToken(i::Token::Value token);
   void ReportMessageAt(int start_pos,
                        int end_pos,
                        const char* type,
                        const char* name_opt) {
     log_->LogMessage(start_pos, end_pos, type, name_opt);
   }
 
@@ skipping 42 matching lines @@
   Expression ParseFunctionLiteral(bool* ok);
 
   Identifier ParseIdentifier(bool* ok);
   Identifier ParseIdentifierName(bool* ok);
   Identifier ParseIdentifierOrGetOrSet(bool* is_get, bool* is_set, bool* ok);
 
   Identifier GetIdentifierSymbol();
   unsigned int HexDigitValue(char digit);
   Expression GetStringSymbol();
 
-
-  i::Token::Value peek() { return scanner_->peek(); }
-  i::Token::Value Next() {
-    i::Token::Value next = scanner_->Next();
-    return next;
+  i::Token::Value peek() {
+    if (stack_overflow_) return i::Token::ILLEGAL;
+    return scanner_->peek();
   }
 
-  void Consume(i::Token::Value token) {
-    Next();
+  i::Token::Value Next() {
+    if (stack_overflow_) return i::Token::ILLEGAL;
+    {
+      int marker;
+      if (reinterpret_cast<uintptr_t>(&marker) < stack_limit_) {
+        // Further calls to peek/Next will return illegal token.
+        // The current one will still be returned. It might already
+        // have been seen using peek.
+        stack_overflow_ = true;
+      }
+    }
+    return scanner_->Next();
   }
 
+  void Consume(i::Token::Value token) { Next(); }
+
   void Expect(i::Token::Value token, bool* ok) {
     if (Next() != token) {
       *ok = false;
     }
   }
 
   bool Check(i::Token::Value token) {
     i::Token::Value next = peek();
     if (next == token) {
       Consume(next);
       return true;
     }
     return false;
   }
   void ExpectSemicolon(bool* ok);
 
   static int Precedence(i::Token::Value tok, bool accept_IN);
 
   i::JavaScriptScanner* scanner_;
   i::ParserRecorder* log_;
   Scope* scope_;
+  uintptr_t stack_limit_;
+  bool stack_overflow_;
   bool allow_lazy_;
 };
 } }  // v8::preparser
 
 #endif  // V8_PREPARSER_H