WIP IDL-IPC2

ppapi/generators/idl_lexer.py

 #!/usr/bin/env python
 # Copyright (c) 2012 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 """ Lexer for PPAPI IDL """
 
 #
 # IDL Lexer
 #
@@ skipping 37 matching lines @@
       'COMMENT',
       'DESCRIBE',
       'ENUM',
       'LABEL',
       'SYMBOL',
       'INLINE',
       'INTERFACE',
       'STRUCT',
       'TYPEDEF',
 
+    # Extra WebIDL keywords
+      'CALLBACK',
+      'DICTIONARY',
+      'OPTIONAL',
+      'STATIC',
+
+    # Invented for apps use
+      'NAMESPACE',
+
+
     # Data types
       'FLOAT',
       'OCT',
       'INT',
       'HEX',
       'STRING',
 
     # Operators
       'LSHIFT',
       'RSHIFT'
   ]
 
   # 'keywords' is a map of string to token type.  All SYMBOL tokens are
   # matched against keywords, to determine if the token is actually a keyword.
   keywords = {
     'describe' : 'DESCRIBE',
     'enum'  : 'ENUM',
     'label' : 'LABEL',
     'interface' : 'INTERFACE',
     'readonly' : 'READONLY',
     'struct' : 'STRUCT',
     'typedef' : 'TYPEDEF',
+
+    'callback' : 'CALLBACK',
+    'dictionary' : 'DICTIONARY',
+    'optional' : 'OPTIONAL',
+    'static' : 'STATIC',
+    'namespace' : 'NAMESPACE',
   }
 
   # 'literals' is a value expected by lex which specifies a list of valid
   # literal tokens, meaning the token type and token value are identical.
-  literals = '"*.(){}[],;:=+-/~|&^'
+  literals = '"*.(){}[],;:=+-/~|&^?'
 
   # Token definitions
   #
   # Lex assumes any value or function in the form of 't_<TYPE>' represents a
   # regular expression where a match will emit a token of type <TYPE>.  In the
   # case of a function, the function is called when a match is made. These
   # definitions come from WebIDL.
 
   # 't_ignore' is a special match of items to ignore
   t_ignore = ' \t'
@@ skipping 14 matching lines @@
   # We do not process escapes in the IDL strings.  Strings are exclusively
   # used for attributes, and not used as typical 'C' constants.
   def t_STRING(self, t):
     r'"[^"]*"'
     t.value = t.value[1:-1]
     self.AddLines(t.value.count('\n'))
     return t
 
   # A C or C++ style comment:  /* xxx */ or //
   def t_COMMENT(self, t):
-    r'(/\*(.|\n)*?\*/)|(//.*)'
+    r'(/\*(.|\n)*?\*/)|(//.*(\n//.*)*)'
     self.AddLines(t.value.count('\n'))
     return t
 
   # Return a "preprocessor" inline block
   def t_INLINE(self, t):
     r'\#inline (.|\n)*?\#endinl.*'
     self.AddLines(t.value.count('\n'))
     return t
 
   # A symbol or keyword.
@@ skipping 190 matching lines @@
         return -1
     return 0
 
   except lex.LexError as le:
     sys.stderr.write('%s\n' % str(le))
   return -1
 
 
 if __name__ == '__main__':
   sys.exit(Main(sys.argv[1:]))