Irregexp: Added derived knowledge of whether we are at the start of input.

src/parser.cc

 // Copyright 2006-2008 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 514 matching lines @@
 
   CharacterRange ParseClassAtom(uc16* char_class);
   RegExpTree* ReportError(Vector<const char> message);
   void Advance();
   void Advance(int dist);
   void Reset(int pos);
 
   // Reports whether the pattern might be used as a literal search string.
   // Only use if the result of the parse is a single atom node.
   bool simple();
-
+  bool contains_anchor() { return contains_anchor_; }
+  void set_contains_anchor() { contains_anchor_ = true; }
   int captures_started() { return captures_ == NULL ? 0 : captures_->length(); }
   int position() { return next_pos_ - 1; }
   bool failed() { return failed_; }
 
   static const int kMaxCaptures = 1 << 16;
   static const uc32 kEndMarker = (1 << 21);
  private:
 
   uc32 current() { return current_; }
   bool has_more() { return has_more_; }
   bool has_next() { return next_pos_ < in()->length(); }
   uc32 Next();
   FlatStringReader* in() { return in_; }
   void ScanForCaptures();
   bool CaptureAvailable(int index);
   uc32 current_;
   bool has_more_;
   bool multiline_;
   int next_pos_;
   FlatStringReader* in_;
   Handle<String>* error_;
   bool simple_;
+  bool contains_anchor_;
   ZoneList<RegExpCapture*>* captures_;
   bool is_scanned_for_captures_;
   // The capture count is only valid after we have scanned for captures.
   int capture_count_;
   bool failed_;
 };
 
 
 // A temporary scope stores information during parsing, just like
 // a plain scope.  However, temporary scopes are not kept around
@@ skipping 2911 matching lines @@
 RegExpParser::RegExpParser(FlatStringReader* in,
                            Handle<String>* error,
                            bool multiline)
   : current_(kEndMarker),
     has_more_(true),
     multiline_(multiline),
     next_pos_(0),
     in_(in),
     error_(error),
     simple_(true),
+    contains_anchor_(false),
     captures_(NULL),
     is_scanned_for_captures_(false),
     capture_count_(0),
     failed_(false) {
   Advance(1);
 }
 
 
 uc32 RegExpParser::Next() {
   if (has_next()) {
@@ skipping 97 matching lines @@
       }
       capture_start_index = capture_new_alt_start_index;
       continue;
     }
     case '*':
     case '+':
     case '?':
       ReportError(CStrVector("Nothing to repeat") CHECK_FAILED);
     case '^': {
       Advance();
-      RegExpAssertion::Type type =
-          multiline_ ? RegExpAssertion::START_OF_LINE :
-                       RegExpAssertion::START_OF_INPUT;
-      builder.AddAssertion(new RegExpAssertion(type));
+      if (multiline_) {
+        builder.AddAssertion(
+            new RegExpAssertion(RegExpAssertion::START_OF_LINE));
+      } else {
+        builder.AddAssertion(
+            new RegExpAssertion(RegExpAssertion::START_OF_INPUT));
+        set_contains_anchor();
+      }
       continue;
     }
     case '$': {
       Advance();
       RegExpAssertion::Type type =
           multiline_ ? RegExpAssertion::END_OF_LINE :
                        RegExpAssertion::END_OF_INPUT;
       builder.AddAssertion(new RegExpAssertion(type));
       continue;
     }
@@ skipping 685 matching lines @@
   RegExpTree* tree = parser.ParsePattern();
   if (parser.failed()) {
     ASSERT(tree == NULL);
     ASSERT(!result->error.is_null());
   } else {
     ASSERT(tree != NULL);
     ASSERT(result->error.is_null());
     result->tree = tree;
     int capture_count = parser.captures_started();
     result->simple = tree->IsAtom() && parser.simple() && capture_count == 0;
+    result->contains_anchor = parser.contains_anchor();
     result->capture_count = capture_count;
   }
   return !parser.failed();
 }
 
 
 FunctionLiteral* MakeAST(bool compile_in_global_context,
                          Handle<Script> script,
                          v8::Extension* extension,
                          ScriptDataImpl* pre_data) {
@@ skipping 38 matching lines @@
                        start_position,
                        is_expression);
   return result;
 }
 
 
 #undef NEW
 
 
 } }  // namespace v8::internal