Down integrate from Google internal.

28 files changed, 927 insertions, 540 deletions

diff --git a/python/google/protobuf/descriptor.py b/python/google/protobuf/descriptor.py
index 5f613c88..3209b34d 100755
--- a/python/google/protobuf/descriptor.py
+++ b/python/google/protobuf/descriptor.py
@@ -783,6 +783,8 @@ class FileDescriptor(DescriptorBase):
   serialized_pb: (str) Byte string of serialized
     descriptor_pb2.FileDescriptorProto.
   dependencies: List of other FileDescriptors this FileDescriptor depends on.
+  public_dependencies: A list of FileDescriptors, subset of the dependencies
+    above, which were declared as "public".
   message_types_by_name: Dict of message names of their descriptors.
   enum_types_by_name: Dict of enum names and their descriptors.
   extensions_by_name: Dict of extension names and their descriptors.
@@ -794,7 +796,8 @@ class FileDescriptor(DescriptorBase):
     _C_DESCRIPTOR_CLASS = _message.FileDescriptor
 
     def __new__(cls, name, package, options=None, serialized_pb=None,
-                dependencies=None, syntax=None, pool=None):
+                dependencies=None, public_dependencies=None,
+                syntax=None, pool=None):
       # FileDescriptor() is called from various places, not only from generated
       # files, to register dynamic proto files and messages.
       if serialized_pb:
@@ -805,7 +808,8 @@ class FileDescriptor(DescriptorBase):
         return super(FileDescriptor, cls).__new__(cls)
 
   def __init__(self, name, package, options=None, serialized_pb=None,
-               dependencies=None, syntax=None, pool=None):
+               dependencies=None, public_dependencies=None,
+               syntax=None, pool=None):
     """Constructor."""
     super(FileDescriptor, self).__init__(options, 'FileOptions')
 
@@ -822,6 +826,7 @@ class FileDescriptor(DescriptorBase):
     self.enum_types_by_name = {}
     self.extensions_by_name = {}
     self.dependencies = (dependencies or [])
+    self.public_dependencies = (public_dependencies or [])
 
     if (api_implementation.Type() == 'cpp' and
         self.serialized_pb is not None):
diff --git a/python/google/protobuf/descriptor_pool.py b/python/google/protobuf/descriptor_pool.py
index 3e80795c..20a33701 100644
--- a/python/google/protobuf/descriptor_pool.py
+++ b/python/google/protobuf/descriptor_pool.py
@@ -319,6 +319,7 @@ class DescriptorPool(object):
     if file_proto.name not in self._file_descriptors:
       built_deps = list(self._GetDeps(file_proto.dependency))
       direct_deps = [self.FindFileByName(n) for n in file_proto.dependency]
+      public_deps = [direct_deps[i] for i in file_proto.public_dependency]
 
       file_descriptor = descriptor.FileDescriptor(
           pool=self,
@@ -327,7 +328,8 @@ class DescriptorPool(object):
           syntax=file_proto.syntax,
           options=file_proto.options,
           serialized_pb=file_proto.SerializeToString(),
-          dependencies=direct_deps)
+          dependencies=direct_deps,
+          public_dependencies=public_deps)
       if _USE_C_DESCRIPTORS:
         # When using C++ descriptors, all objects defined in the file were added
         # to the C++ database when the FileDescriptor was built above.
diff --git a/python/google/protobuf/internal/api_implementation.py b/python/google/protobuf/internal/api_implementation.py
index ffcf7511..460a4a6c 100755
--- a/python/google/protobuf/internal/api_implementation.py
+++ b/python/google/protobuf/internal/api_implementation.py
@@ -32,6 +32,7 @@
 """
 
 import os
+import warnings
 import sys
 
 try:
@@ -78,6 +79,11 @@ _implementation_type = os.getenv('PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION',
 if _implementation_type != 'python':
   _implementation_type = 'cpp'
 
+if 'PyPy' in sys.version and _implementation_type == 'cpp':
+  warnings.warn('PyPy does not work yet with cpp protocol buffers. '
+                'Falling back to the python implementation.')
+  _implementation_type = 'python'
+
 # This environment variable can be used to switch between the two
 # 'cpp' implementations, overriding the compile-time constants in the
 # _api_implementation module. Right now only '2' is supported. Any other
diff --git a/python/google/protobuf/internal/descriptor_pool_test.py b/python/google/protobuf/internal/descriptor_pool_test.py
index 4b1811d8..6a13e0bc 100644
--- a/python/google/protobuf/internal/descriptor_pool_test.py
+++ b/python/google/protobuf/internal/descriptor_pool_test.py
@@ -51,6 +51,7 @@ from google.protobuf.internal import descriptor_pool_test1_pb2
 from google.protobuf.internal import descriptor_pool_test2_pb2
 from google.protobuf.internal import factory_test1_pb2
 from google.protobuf.internal import factory_test2_pb2
+from google.protobuf.internal import more_messages_pb2
 from google.protobuf import descriptor
 from google.protobuf import descriptor_database
 from google.protobuf import descriptor_pool
@@ -60,11 +61,8 @@ from google.protobuf import symbol_database
 
 class DescriptorPoolTest(unittest.TestCase):
 
-  def CreatePool(self):
-    return descriptor_pool.DescriptorPool()
-
   def setUp(self):
-    self.pool = self.CreatePool()
+    self.pool = descriptor_pool.DescriptorPool()
     self.factory_test1_fd = descriptor_pb2.FileDescriptorProto.FromString(
         factory_test1_pb2.DESCRIPTOR.serialized_pb)
     self.factory_test2_fd = descriptor_pb2.FileDescriptorProto.FromString(
@@ -275,10 +273,13 @@ class DescriptorPoolTest(unittest.TestCase):
     self.testFindMessageTypeByName()
 
   def testComplexNesting(self):
+    more_messages_desc = descriptor_pb2.FileDescriptorProto.FromString(
+        more_messages_pb2.DESCRIPTOR.serialized_pb)
     test1_desc = descriptor_pb2.FileDescriptorProto.FromString(
         descriptor_pool_test1_pb2.DESCRIPTOR.serialized_pb)
     test2_desc = descriptor_pb2.FileDescriptorProto.FromString(
         descriptor_pool_test2_pb2.DESCRIPTOR.serialized_pb)
+    self.pool.Add(more_messages_desc)
     self.pool.Add(test1_desc)
     self.pool.Add(test2_desc)
     TEST1_FILE.CheckFile(self, self.pool)
@@ -350,25 +351,15 @@ class DescriptorPoolTest(unittest.TestCase):
     _CheckDefaultValues(message_class())
 
 
-@unittest.skipIf(api_implementation.Type() != 'cpp',
-                 'explicit tests of the C++ implementation')
-class CppDescriptorPoolTest(DescriptorPoolTest):
-  # TODO(amauryfa): remove when descriptor_pool.DescriptorPool() creates true
-  # C++ descriptor pool object for C++ implementation.
-
-  def CreatePool(self):
-    # pylint: disable=g-import-not-at-top
-    from google.protobuf.pyext import _message
-    return _message.DescriptorPool()
-
-
 class ProtoFile(object):
 
-  def __init__(self, name, package, messages, dependencies=None):
+  def __init__(self, name, package, messages, dependencies=None,
+               public_dependencies=None):
     self.name = name
     self.package = package
     self.messages = messages
     self.dependencies = dependencies or []
+    self.public_dependencies = public_dependencies or []
 
   def CheckFile(self, test, pool):
     file_desc = pool.FindFileByName(self.name)
@@ -376,6 +367,8 @@ class ProtoFile(object):
     test.assertEqual(self.package, file_desc.package)
     dependencies_names = [f.name for f in file_desc.dependencies]
     test.assertEqual(self.dependencies, dependencies_names)
+    public_dependencies_names = [f.name for f in file_desc.public_dependencies]
+    test.assertEqual(self.public_dependencies, public_dependencies_names)
     for name, msg_type in self.messages.items():
       msg_type.CheckType(test, None, name, file_desc)
 
@@ -613,18 +606,9 @@ class AddDescriptorTest(unittest.TestCase):
       pool.FindFileContainingSymbol(
           'protobuf_unittest.TestAllTypes')
 
-  def _GetDescriptorPoolClass(self):
-    # Test with both implementations of descriptor pools.
-    if api_implementation.Type() == 'cpp':
-      # pylint: disable=g-import-not-at-top
-      from google.protobuf.pyext import _message
-      return _message.DescriptorPool
-    else:
-      return descriptor_pool.DescriptorPool
-
   def testEmptyDescriptorPool(self):
-    # Check that an empty DescriptorPool() contains no message.
-    pool = self._GetDescriptorPoolClass()()
+    # Check that an empty DescriptorPool() contains no messages.
+    pool = descriptor_pool.DescriptorPool()
     proto_file_name = descriptor_pb2.DESCRIPTOR.name
     self.assertRaises(KeyError, pool.FindFileByName, proto_file_name)
     # Add the above file to the pool
@@ -636,7 +620,7 @@ class AddDescriptorTest(unittest.TestCase):
 
   def testCustomDescriptorPool(self):
     # Create a new pool, and add a file descriptor.
-    pool = self._GetDescriptorPoolClass()()
+    pool = descriptor_pool.DescriptorPool()
     file_desc = descriptor_pb2.FileDescriptorProto(
         name='some/file.proto', package='package')
     file_desc.message_type.add(name='Message')
@@ -757,7 +741,9 @@ TEST2_FILE = ProtoFile(
              ExtensionField(1001, 'DescriptorPoolTest1')),
         ]),
     },
-    dependencies=['google/protobuf/internal/descriptor_pool_test1.proto'])
+    dependencies=['google/protobuf/internal/descriptor_pool_test1.proto',
+                  'google/protobuf/internal/more_messages.proto'],
+    public_dependencies=['google/protobuf/internal/more_messages.proto'])
 
 
 if __name__ == '__main__':
diff --git a/python/google/protobuf/internal/descriptor_pool_test2.proto b/python/google/protobuf/internal/descriptor_pool_test2.proto
index e3fa660c..a218eccb 100644
--- a/python/google/protobuf/internal/descriptor_pool_test2.proto
+++ b/python/google/protobuf/internal/descriptor_pool_test2.proto
@@ -33,6 +33,7 @@ syntax = "proto2";
 package google.protobuf.python.internal;
 
 import "google/protobuf/internal/descriptor_pool_test1.proto";
+import public "google/protobuf/internal/more_messages.proto";
 
 
 message DescriptorPoolTest3 {
diff --git a/python/google/protobuf/internal/message_factory_test.py b/python/google/protobuf/internal/message_factory_test.py
index 54b1f688..7bb7d1ac 100644
--- a/python/google/protobuf/internal/message_factory_test.py
+++ b/python/google/protobuf/internal/message_factory_test.py
@@ -131,6 +131,60 @@ class MessageFactoryTest(unittest.TestCase):
       self.assertEqual('test1', msg1.Extensions[ext1])
       self.assertEqual('test2', msg1.Extensions[ext2])
 
+  def testDuplicateExtensionNumber(self):
+    pool = descriptor_pool.DescriptorPool()
+    factory = message_factory.MessageFactory(pool=pool)
+
+    # Add Container message.
+    f = descriptor_pb2.FileDescriptorProto()
+    f.name = 'google/protobuf/internal/container.proto'
+    f.package = 'google.protobuf.python.internal'
+    msg = f.message_type.add()
+    msg.name = 'Container'
+    rng = msg.extension_range.add()
+    rng.start = 1
+    rng.end = 10
+    pool.Add(f)
+    msgs = factory.GetMessages([f.name])
+    self.assertIn('google.protobuf.python.internal.Container', msgs)
+
+    # Extend container.
+    f = descriptor_pb2.FileDescriptorProto()
+    f.name = 'google/protobuf/internal/extension.proto'
+    f.package = 'google.protobuf.python.internal'
+    f.dependency.append('google/protobuf/internal/container.proto')
+    msg = f.message_type.add()
+    msg.name = 'Extension'
+    ext = msg.extension.add()
+    ext.name = 'extension_field'
+    ext.number = 2
+    ext.label = descriptor_pb2.FieldDescriptorProto.LABEL_OPTIONAL
+    ext.type_name = 'Extension'
+    ext.extendee = 'Container'
+    pool.Add(f)
+    msgs = factory.GetMessages([f.name])
+    self.assertIn('google.protobuf.python.internal.Extension', msgs)
+
+    # Add Duplicate extending the same field number.
+    f = descriptor_pb2.FileDescriptorProto()
+    f.name = 'google/protobuf/internal/duplicate.proto'
+    f.package = 'google.protobuf.python.internal'
+    f.dependency.append('google/protobuf/internal/container.proto')
+    msg = f.message_type.add()
+    msg.name = 'Duplicate'
+    ext = msg.extension.add()
+    ext.name = 'extension_field'
+    ext.number = 2
+    ext.label = descriptor_pb2.FieldDescriptorProto.LABEL_OPTIONAL
+    ext.type_name = 'Duplicate'
+    ext.extendee = 'Container'
+    pool.Add(f)
+
+    with self.assertRaises(Exception) as cm:
+      factory.GetMessages([f.name])
+
+    self.assertIsInstance(cm.exception, (AssertionError, ValueError))
+
 
 if __name__ == '__main__':
   unittest.main()
diff --git a/python/google/protobuf/internal/message_test.py b/python/google/protobuf/internal/message_test.py
index 1232ccc9..4ee31d8e 100755
--- a/python/google/protobuf/internal/message_test.py
+++ b/python/google/protobuf/internal/message_test.py
@@ -57,18 +57,18 @@ try:
 except ImportError:
   import unittest
 
-from google.protobuf.internal import _parameterized
+from google.protobuf import map_unittest_pb2
+from google.protobuf import unittest_pb2
+from google.protobuf import unittest_proto3_arena_pb2
 from google.protobuf import descriptor_pb2
 from google.protobuf import descriptor_pool
-from google.protobuf import map_unittest_pb2
 from google.protobuf import message_factory
 from google.protobuf import text_format
-from google.protobuf import unittest_pb2
-from google.protobuf import unittest_proto3_arena_pb2
 from google.protobuf.internal import api_implementation
 from google.protobuf.internal import packed_field_test_pb2
 from google.protobuf.internal import test_util
 from google.protobuf import message
+from google.protobuf.internal import _parameterized
 
 if six.PY3:
   long = int
@@ -1265,7 +1265,10 @@ class Proto3Test(unittest.TestCase):
     self.assertFalse(-2**33 in msg.map_int64_int64)
     self.assertFalse(123 in msg.map_uint32_uint32)
     self.assertFalse(2**33 in msg.map_uint64_uint64)
+    self.assertFalse(123 in msg.map_int32_double)
+    self.assertFalse(False in msg.map_bool_bool)
     self.assertFalse('abc' in msg.map_string_string)
+    self.assertFalse(111 in msg.map_int32_bytes)
     self.assertFalse(888 in msg.map_int32_enum)
 
     # Accessing an unset key returns the default.
@@ -1273,7 +1276,12 @@ class Proto3Test(unittest.TestCase):
     self.assertEqual(0, msg.map_int64_int64[-2**33])
     self.assertEqual(0, msg.map_uint32_uint32[123])
     self.assertEqual(0, msg.map_uint64_uint64[2**33])
+    self.assertEqual(0.0, msg.map_int32_double[123])
+    self.assertTrue(isinstance(msg.map_int32_double[123], float))
+    self.assertEqual(False, msg.map_bool_bool[False])
+    self.assertTrue(isinstance(msg.map_bool_bool[False], bool))
     self.assertEqual('', msg.map_string_string['abc'])
+    self.assertEqual(b'', msg.map_int32_bytes[111])
     self.assertEqual(0, msg.map_int32_enum[888])
 
     # It also sets the value in the map
@@ -1281,7 +1289,10 @@ class Proto3Test(unittest.TestCase):
     self.assertTrue(-2**33 in msg.map_int64_int64)
     self.assertTrue(123 in msg.map_uint32_uint32)
     self.assertTrue(2**33 in msg.map_uint64_uint64)
+    self.assertTrue(123 in msg.map_int32_double)
+    self.assertTrue(False in msg.map_bool_bool)
     self.assertTrue('abc' in msg.map_string_string)
+    self.assertTrue(111 in msg.map_int32_bytes)
     self.assertTrue(888 in msg.map_int32_enum)
 
     self.assertIsInstance(msg.map_string_string['abc'], six.text_type)
@@ -1587,6 +1598,21 @@ class Proto3Test(unittest.TestCase):
     matching_dict = {2: 4, 3: 6, 4: 8}
     self.assertMapIterEquals(msg.map_int32_int32.items(), matching_dict)
 
+  def testMapItems(self):
+    # Map items used to have strange behaviors when use c extension. Because
+    # [] may reorder the map and invalidate any exsting iterators.
+    # TODO(jieluo): Check if [] reordering the map is a bug or intended
+    # behavior.
+    msg = map_unittest_pb2.TestMap()
+    msg.map_string_string['local_init_op'] = ''
+    msg.map_string_string['trainable_variables'] = ''
+    msg.map_string_string['variables'] = ''
+    msg.map_string_string['init_op'] = ''
+    msg.map_string_string['summaries'] = ''
+    items1 = msg.map_string_string.items()
+    items2 = msg.map_string_string.items()
+    self.assertEqual(items1, items2)
+
   def testMapIterationClearMessage(self):
     # Iterator needs to work even if message and map are deleted.
     msg = map_unittest_pb2.TestMap()
diff --git a/python/google/protobuf/internal/proto_builder_test.py b/python/google/protobuf/internal/proto_builder_test.py
index 822ad895..36dfbfde 100644
--- a/python/google/protobuf/internal/proto_builder_test.py
+++ b/python/google/protobuf/internal/proto_builder_test.py
@@ -40,6 +40,7 @@ try:
   import unittest2 as unittest
 except ImportError:
   import unittest
+
 from google.protobuf import descriptor_pb2
 from google.protobuf import descriptor_pool
 from google.protobuf import proto_builder
diff --git a/python/google/protobuf/internal/python_message.py b/python/google/protobuf/internal/python_message.py
index 87f60666..f8f73dd2 100755
--- a/python/google/protobuf/internal/python_message.py
+++ b/python/google/protobuf/internal/python_message.py
@@ -56,7 +56,14 @@ import struct
 import weakref
 
 import six
-import six.moves.copyreg as copyreg
+try:
+  import six.moves.copyreg as copyreg
+except ImportError:
+  # On some platforms, for example gMac, we run native Python because there is
+  # nothing like hermetic Python. This means lesser control on the system and
+  # the six.moves package may be missing (is missing on 20150321 on gMac). Be
+  # extra conservative and try to load the old replacement if it fails.
+  import copy_reg as copyreg
 
 # We use "as" to avoid name collisions with variables.
 from google.protobuf.internal import containers
@@ -490,6 +497,9 @@ def _AddInitMethod(message_descriptor, cls):
       if field is None:
         raise TypeError("%s() got an unexpected keyword argument '%s'" %
                         (message_descriptor.name, field_name))
+      if field_value is None:
+        # field=None is the same as no field at all.
+        continue
       if field.label == _FieldDescriptor.LABEL_REPEATED:
         copy = field._default_constructor(self)
         if field.cpp_type == _FieldDescriptor.CPPTYPE_MESSAGE:  # Composite
@@ -889,17 +899,6 @@ def _AddClearExtensionMethod(cls):
   cls.ClearExtension = ClearExtension
 
 
-def _AddClearMethod(message_descriptor, cls):
-  """Helper for _AddMessageMethods()."""
-  def Clear(self):
-    # Clear fields.
-    self._fields = {}
-    self._unknown_fields = ()
-    self._oneofs = {}
-    self._Modified()
-  cls.Clear = Clear
-
-
 def _AddHasExtensionMethod(cls):
   """Helper for _AddMessageMethods()."""
   def HasExtension(self, extension_handle):
@@ -999,16 +998,6 @@ def _AddUnicodeMethod(unused_message_descriptor, cls):
   cls.__unicode__ = __unicode__
 
 
-def _AddSetListenerMethod(cls):
-  """Helper for _AddMessageMethods()."""
-  def SetListener(self, listener):
-    if listener is None:
-      self._listener = message_listener_mod.NullMessageListener()
-    else:
-      self._listener = listener
-  cls._SetListener = SetListener
-
-
 def _BytesForNonRepeatedElement(value, field_number, field_type):
   """Returns the number of bytes needed to serialize a non-repeated element.
   The returned byte count includes space for tag information and any
@@ -1288,6 +1277,32 @@ def _AddWhichOneofMethod(message_descriptor, cls):
   cls.WhichOneof = WhichOneof
 
 
+def _Clear(self):
+  # Clear fields.
+  self._fields = {}
+  self._unknown_fields = ()
+  self._oneofs = {}
+  self._Modified()
+
+
+def _DiscardUnknownFields(self):
+  self._unknown_fields = []
+  for field, value in self.ListFields():
+    if field.cpp_type == _FieldDescriptor.CPPTYPE_MESSAGE:
+      if field.label == _FieldDescriptor.LABEL_REPEATED:
+        for sub_message in value:
+          sub_message.DiscardUnknownFields()
+      else:
+        value.DiscardUnknownFields()
+
+
+def _SetListener(self, listener):
+  if listener is None:
+    self._listener = message_listener_mod.NullMessageListener()
+  else:
+    self._listener = listener
+
+
 def _AddMessageMethods(message_descriptor, cls):
   """Adds implementations of all Message methods to cls."""
   _AddListFieldsMethod(message_descriptor, cls)
@@ -1296,12 +1311,10 @@ def _AddMessageMethods(message_descriptor, cls):
   if message_descriptor.is_extendable:
     _AddClearExtensionMethod(cls)
     _AddHasExtensionMethod(cls)
-  _AddClearMethod(message_descriptor, cls)
   _AddEqualsMethod(message_descriptor, cls)
   _AddStrMethod(message_descriptor, cls)
   _AddReprMethod(message_descriptor, cls)
   _AddUnicodeMethod(message_descriptor, cls)
-  _AddSetListenerMethod(cls)
   _AddByteSizeMethod(message_descriptor, cls)
   _AddSerializeToStringMethod(message_descriptor, cls)
   _AddSerializePartialToStringMethod(message_descriptor, cls)
@@ -1309,6 +1322,10 @@ def _AddMessageMethods(message_descriptor, cls):
   _AddIsInitializedMethod(message_descriptor, cls)
   _AddMergeFromMethod(cls)
   _AddWhichOneofMethod(message_descriptor, cls)
+  # Adds methods which do not depend on cls.
+  cls.Clear = _Clear
+  cls.DiscardUnknownFields = _DiscardUnknownFields
+  cls._SetListener = _SetListener
 
 
 def _AddPrivateHelperMethods(message_descriptor, cls):
@@ -1518,3 +1535,14 @@ class _ExtensionDict(object):
       Extension field descriptor.
     """
     return self._extended_message._extensions_by_name.get(name, None)
+
+  def _FindExtensionByNumber(self, number):
+    """Tries to find a known extension with the field number.
+
+    Args:
+      number: Extension field number.
+
+    Returns:
+      Extension field descriptor.
+    """
+    return self._extended_message._extensions_by_number.get(number, None)
diff --git a/python/google/protobuf/internal/reflection_test.py b/python/google/protobuf/internal/reflection_test.py
index 9e61ea0e..6dc2fffe 100755
--- a/python/google/protobuf/internal/reflection_test.py
+++ b/python/google/protobuf/internal/reflection_test.py
@@ -120,11 +120,13 @@ class ReflectionTest(unittest.TestCase):
     proto = unittest_pb2.TestAllTypes(
         optional_int32=24,
         optional_double=54.321,
-        optional_string='optional_string')
+        optional_string='optional_string',
+        optional_float=None)
 
     self.assertEqual(24, proto.optional_int32)
     self.assertEqual(54.321, proto.optional_double)
     self.assertEqual('optional_string', proto.optional_string)
+    self.assertFalse(proto.HasField("optional_float"))
 
   def testRepeatedScalarConstructor(self):
     # Constructor with only repeated scalar types should succeed.
@@ -132,12 +134,14 @@ class ReflectionTest(unittest.TestCase):
         repeated_int32=[1, 2, 3, 4],
         repeated_double=[1.23, 54.321],
         repeated_bool=[True, False, False],
-        repeated_string=["optional_string"])
+        repeated_string=["optional_string"],
+        repeated_float=None)
 
     self.assertEqual([1, 2, 3, 4], list(proto.repeated_int32))
     self.assertEqual([1.23, 54.321], list(proto.repeated_double))
     self.assertEqual([True, False, False], list(proto.repeated_bool))
     self.assertEqual(["optional_string"], list(proto.repeated_string))
+    self.assertEqual([], list(proto.repeated_float))
 
   def testRepeatedCompositeConstructor(self):
     # Constructor with only repeated composite types should succeed.
@@ -188,7 +192,8 @@ class ReflectionTest(unittest.TestCase):
         repeated_foreign_message=[
             unittest_pb2.ForeignMessage(c=-43),
             unittest_pb2.ForeignMessage(c=45324),
-            unittest_pb2.ForeignMessage(c=12)])
+            unittest_pb2.ForeignMessage(c=12)],
+        optional_nested_message=None)
 
     self.assertEqual(24, proto.optional_int32)
     self.assertEqual('optional_string', proto.optional_string)
@@ -205,6 +210,7 @@ class ReflectionTest(unittest.TestCase):
          unittest_pb2.ForeignMessage(c=45324),
          unittest_pb2.ForeignMessage(c=12)],
         list(proto.repeated_foreign_message))
+    self.assertFalse(proto.HasField("optional_nested_message"))
 
   def testConstructorTypeError(self):
     self.assertRaises(
diff --git a/python/google/protobuf/internal/text_format_test.py b/python/google/protobuf/internal/text_format_test.py
index 8ce0a44f..ab2bf05b 100755
--- a/python/google/protobuf/internal/text_format_test.py
+++ b/python/google/protobuf/internal/text_format_test.py
@@ -250,6 +250,36 @@ class TextFormatTest(TextFormatBase):
     message.c = 123
     self.assertEqual('c: 123\n', str(message))
 
+  def testPrintField(self, message_module):
+    message = message_module.TestAllTypes()
+    field = message.DESCRIPTOR.fields_by_name['optional_float']
+    value = message.optional_float
+    out = text_format.TextWriter(False)
+    text_format.PrintField(field, value, out)
+    self.assertEqual('optional_float: 0.0\n', out.getvalue())
+    out.close()
+    # Test Printer
+    out = text_format.TextWriter(False)
+    printer = text_format._Printer(out)
+    printer.PrintField(field, value)
+    self.assertEqual('optional_float: 0.0\n', out.getvalue())
+    out.close()
+
+  def testPrintFieldValue(self, message_module):
+    message = message_module.TestAllTypes()
+    field = message.DESCRIPTOR.fields_by_name['optional_float']
+    value = message.optional_float
+    out = text_format.TextWriter(False)
+    text_format.PrintFieldValue(field, value, out)
+    self.assertEqual('0.0', out.getvalue())
+    out.close()
+    # Test Printer
+    out = text_format.TextWriter(False)
+    printer = text_format._Printer(out)
+    printer.PrintFieldValue(field, value)
+    self.assertEqual('0.0', out.getvalue())
+    out.close()
+
   def testParseAllFields(self, message_module):
     message = message_module.TestAllTypes()
     test_util.SetAllFields(message)
@@ -616,6 +646,26 @@ class Proto2Tests(TextFormatBase):
         '  text: \"bar\"\n'
         '}\n')
 
+  def testPrintMessageSetByFieldNumber(self):
+    out = text_format.TextWriter(False)
+    message = unittest_mset_pb2.TestMessageSetContainer()
+    ext1 = unittest_mset_pb2.TestMessageSetExtension1.message_set_extension
+    ext2 = unittest_mset_pb2.TestMessageSetExtension2.message_set_extension
+    message.message_set.Extensions[ext1].i = 23
+    message.message_set.Extensions[ext2].str = 'foo'
+    text_format.PrintMessage(message, out, use_field_number=True)
+    self.CompareToGoldenText(
+        out.getvalue(),
+        '1 {\n'
+        '  1545008 {\n'
+        '    15: 23\n'
+        '  }\n'
+        '  1547769 {\n'
+        '    25: \"foo\"\n'
+        '  }\n'
+        '}\n')
+    out.close()
+
   def testPrintMessageSetAsOneLine(self):
     message = unittest_mset_pb2.TestMessageSetContainer()
     ext1 = unittest_mset_pb2.TestMessageSetExtension1.message_set_extension
@@ -656,6 +706,48 @@ class Proto2Tests(TextFormatBase):
     self.assertEqual(23, message.message_set.Extensions[ext1].i)
     self.assertEqual('foo', message.message_set.Extensions[ext2].str)
 
+  def testParseMessageByFieldNumber(self):
+    message = unittest_pb2.TestAllTypes()
+    text = ('34: 1\n'
+            'repeated_uint64: 2\n')
+    text_format.Parse(text, message, allow_field_number=True)
+    self.assertEqual(1, message.repeated_uint64[0])
+    self.assertEqual(2, message.repeated_uint64[1])
+
+    message = unittest_mset_pb2.TestMessageSetContainer()
+    text = ('1 {\n'
+            '  1545008 {\n'
+            '    15: 23\n'
+            '  }\n'
+            '  1547769 {\n'
+            '    25: \"foo\"\n'
+            '  }\n'
+            '}\n')
+    text_format.Parse(text, message, allow_field_number=True)
+    ext1 = unittest_mset_pb2.TestMessageSetExtension1.message_set_extension
+    ext2 = unittest_mset_pb2.TestMessageSetExtension2.message_set_extension
+    self.assertEqual(23, message.message_set.Extensions[ext1].i)
+    self.assertEqual('foo', message.message_set.Extensions[ext2].str)
+
+    # Can't parse field number without set allow_field_number=True.
+    message = unittest_pb2.TestAllTypes()
+    text = '34:1\n'
+    six.assertRaisesRegex(
+        self,
+        text_format.ParseError,
+        (r'1:1 : Message type "\w+.TestAllTypes" has no field named '
+         r'"34".'),
+        text_format.Parse, text, message)
+
+    # Can't parse if field number is not found.
+    text = '1234:1\n'
+    six.assertRaisesRegex(
+        self,
+        text_format.ParseError,
+        (r'1:1 : Message type "\w+.TestAllTypes" has no field named '
+         r'"1234".'),
+        text_format.Parse, text, message, allow_field_number=True)
+
   def testPrintAllExtensions(self):
     message = unittest_pb2.TestAllExtensions()
     test_util.SetAllExtensions(message)
@@ -696,6 +788,7 @@ class Proto2Tests(TextFormatBase):
     text = ('message_set {\n'
             '  [unknown_extension] {\n'
             '    i: 23\n'
+            '    bin: "\xe0"'
             '    [nested_unknown_ext]: {\n'
             '      i: 23\n'
             '      test: "test_string"\n'
diff --git a/python/google/protobuf/internal/type_checkers.py b/python/google/protobuf/internal/type_checkers.py
index f30ca6a8..1be3ad9a 100755
--- a/python/google/protobuf/internal/type_checkers.py
+++ b/python/google/protobuf/internal/type_checkers.py
@@ -109,6 +109,16 @@ class TypeChecker(object):
     return proposed_value
 
 
+class TypeCheckerWithDefault(TypeChecker):
+
+  def __init__(self, default_value, *acceptable_types):
+    TypeChecker.__init__(self, acceptable_types)
+    self._default_value = default_value
+
+  def DefaultValue(self):
+    return self._default_value
+
+
 # IntValueChecker and its subclasses perform integer type-checks
 # and bounds-checks.
 class IntValueChecker(object):
@@ -212,12 +222,13 @@ _VALUE_CHECKERS = {
     _FieldDescriptor.CPPTYPE_INT64: Int64ValueChecker(),
     _FieldDescriptor.CPPTYPE_UINT32: Uint32ValueChecker(),
     _FieldDescriptor.CPPTYPE_UINT64: Uint64ValueChecker(),
-    _FieldDescriptor.CPPTYPE_DOUBLE: TypeChecker(
-        float, int, long),
-    _FieldDescriptor.CPPTYPE_FLOAT: TypeChecker(
-        float, int, long),
-    _FieldDescriptor.CPPTYPE_BOOL: TypeChecker(bool, int),
-    _FieldDescriptor.CPPTYPE_STRING: TypeChecker(bytes),
+    _FieldDescriptor.CPPTYPE_DOUBLE: TypeCheckerWithDefault(
+        0.0, float, int, long),
+    _FieldDescriptor.CPPTYPE_FLOAT: TypeCheckerWithDefault(
+        0.0, float, int, long),
+    _FieldDescriptor.CPPTYPE_BOOL: TypeCheckerWithDefault(
+        False, bool, int),
+    _FieldDescriptor.CPPTYPE_STRING: TypeCheckerWithDefault(b'', bytes),
     }
 
 
diff --git a/python/google/protobuf/internal/unknown_fields_test.py b/python/google/protobuf/internal/unknown_fields_test.py
index bb2748e4..84073f1c 100755
--- a/python/google/protobuf/internal/unknown_fields_test.py
+++ b/python/google/protobuf/internal/unknown_fields_test.py
@@ -119,6 +119,26 @@ class UnknownFieldsTest(unittest.TestCase):
     message.ParseFromString(self.all_fields.SerializeToString())
     self.assertNotEqual(self.empty_message, message)
 
+  def testDiscardUnknownFields(self):
+    self.empty_message.DiscardUnknownFields()
+    self.assertEqual(b'', self.empty_message.SerializeToString())
+    # Test message field and repeated message field.
+    message = unittest_pb2.TestAllTypes()
+    other_message = unittest_pb2.TestAllTypes()
+    other_message.optional_string = 'discard'
+    message.optional_nested_message.ParseFromString(
+        other_message.SerializeToString())
+    message.repeated_nested_message.add().ParseFromString(
+        other_message.SerializeToString())
+    self.assertNotEqual(
+        b'', message.optional_nested_message.SerializeToString())
+    self.assertNotEqual(
+        b'', message.repeated_nested_message[0].SerializeToString())
+    message.DiscardUnknownFields()
+    self.assertEqual(b'', message.optional_nested_message.SerializeToString())
+    self.assertEqual(
+        b'', message.repeated_nested_message[0].SerializeToString())
+
 
 class UnknownFieldsAccessorsTest(unittest.TestCase):
 
diff --git a/python/google/protobuf/internal/well_known_types.py b/python/google/protobuf/internal/well_known_types.py
index d35fcc5f..7c5dffd0 100644
--- a/python/google/protobuf/internal/well_known_types.py
+++ b/python/google/protobuf/internal/well_known_types.py
@@ -82,10 +82,14 @@ class Any(object):
     msg.ParseFromString(self.value)
     return True
 
+  def TypeName(self):
+    """Returns the protobuf type name of the inner message."""
+    # Only last part is to be used: b/25630112
+    return self.type_url.split('/')[-1]
+
   def Is(self, descriptor):
     """Checks if this Any represents the given protobuf type."""
-    # Only last part is to be used: b/25630112
-    return self.type_url.split('/')[-1] == descriptor.full_name
+    return self.TypeName() == descriptor.full_name
 
 
 class Timestamp(object):
diff --git a/python/google/protobuf/internal/well_known_types_test.py b/python/google/protobuf/internal/well_known_types_test.py
index 18329205..2f32ac99 100644
--- a/python/google/protobuf/internal/well_known_types_test.py
+++ b/python/google/protobuf/internal/well_known_types_test.py
@@ -610,6 +610,14 @@ class AnyTest(unittest.TestCase):
       raise AttributeError('%s should not have Pack method.' %
                            msg_descriptor.full_name)
 
+  def testMessageName(self):
+    # Creates and sets message.
+    submessage = any_test_pb2.TestAny()
+    submessage.int_value = 12345
+    msg = any_pb2.Any()
+    msg.Pack(submessage)
+    self.assertEqual(msg.TypeName(), 'google.protobuf.internal.TestAny')
+
   def testPackWithCustomTypeUrl(self):
     submessage = any_test_pb2.TestAny()
     submessage.int_value = 12345
diff --git a/python/google/protobuf/message.py b/python/google/protobuf/message.py
index de2f5697..606f735f 100755
--- a/python/google/protobuf/message.py
+++ b/python/google/protobuf/message.py
@@ -255,6 +255,9 @@ class Message(object):
   def ClearExtension(self, extension_handle):
     raise NotImplementedError
 
+  def DiscardUnknownFields(self):
+    raise NotImplementedError
+
   def ByteSize(self):
     """Returns the serialized size of this message.
     Recursively calls ByteSize() on all contained messages.
diff --git a/python/google/protobuf/pyext/descriptor.cc b/python/google/protobuf/pyext/descriptor.cc
index 07550706..23557538 100644
--- a/python/google/protobuf/pyext/descriptor.cc
+++ b/python/google/protobuf/pyext/descriptor.cc
@@ -200,8 +200,8 @@ static PyObject* GetOrBuildOptions(const DescriptorClass *descriptor) {
   // read-only instance.
   const Message& options(descriptor->options());
   const Descriptor *message_type = options.GetDescriptor();
-  PyObject* message_class(cdescriptor_pool::GetMessageClass(
-      pool, message_type));
+  CMessageClass* message_class(
+      cdescriptor_pool::GetMessageClass(pool, message_type));
   if (message_class == NULL) {
     // The Options message was not found in the current DescriptorPool.
     // In this case, there cannot be extensions to these options, and we can
@@ -215,7 +215,8 @@ static PyObject* GetOrBuildOptions(const DescriptorClass *descriptor) {
                  message_type->full_name().c_str());
     return NULL;
   }
-  ScopedPyObjectPtr value(PyEval_CallObject(message_class, NULL));
+  ScopedPyObjectPtr value(
+      PyEval_CallObject(message_class->AsPyObject(), NULL));
   if (value == NULL) {
     return NULL;
   }
@@ -433,11 +434,11 @@ static PyObject* GetConcreteClass(PyBaseDescriptor* self, void *closure) {
   // which contains this descriptor.
   // This might not be the one you expect! For example the returned object does
   // not know about extensions defined in a custom pool.
-  PyObject* concrete_class(cdescriptor_pool::GetMessageClass(
+  CMessageClass* concrete_class(cdescriptor_pool::GetMessageClass(
       GetDescriptorPool_FromPool(_GetDescriptor(self)->file()->pool()),
       _GetDescriptor(self)));
   Py_XINCREF(concrete_class);
-  return concrete_class;
+  return concrete_class->AsPyObject();
 }
 
 static PyObject* GetFieldsByName(PyBaseDescriptor* self, void *closure) {
diff --git a/python/google/protobuf/pyext/descriptor_pool.cc b/python/google/protobuf/pyext/descriptor_pool.cc
index 0bc76bc9..1faff96b 100644
--- a/python/google/protobuf/pyext/descriptor_pool.cc
+++ b/python/google/protobuf/pyext/descriptor_pool.cc
@@ -190,8 +190,8 @@ PyObject* FindMessageByName(PyDescriptorPool* self, PyObject* arg) {
 
 // Add a message class to our database.
 int RegisterMessageClass(PyDescriptorPool* self,
-                         const Descriptor *message_descriptor,
-                         PyObject *message_class) {
+                         const Descriptor* message_descriptor,
+                         CMessageClass* message_class) {
   Py_INCREF(message_class);
   typedef PyDescriptorPool::ClassesByMessageMap::iterator iterator;
   std::pair<iterator, bool> ret = self->classes_by_descriptor->insert(
@@ -205,8 +205,8 @@ int RegisterMessageClass(PyDescriptorPool* self,
 }
 
 // Retrieve the message class added to our database.
-PyObject *GetMessageClass(PyDescriptorPool* self,
-                          const Descriptor *message_descriptor) {
+CMessageClass* GetMessageClass(PyDescriptorPool* self,
+                               const Descriptor* message_descriptor) {
   typedef PyDescriptorPool::ClassesByMessageMap::iterator iterator;
   iterator ret = self->classes_by_descriptor->find(message_descriptor);
   if (ret == self->classes_by_descriptor->end()) {
diff --git a/python/google/protobuf/pyext/descriptor_pool.h b/python/google/protobuf/pyext/descriptor_pool.h
index 16bc910c..2a42c112 100644
--- a/python/google/protobuf/pyext/descriptor_pool.h
+++ b/python/google/protobuf/pyext/descriptor_pool.h
@@ -42,6 +42,9 @@ class MessageFactory;
 
 namespace python {
 
+// The (meta) type of all Messages classes.
+struct CMessageClass;
+
 // Wraps operations to the global DescriptorPool which contains information
 // about all messages and fields.
 //
@@ -78,7 +81,7 @@ typedef struct PyDescriptorPool {
   //
   // Descriptor pointers stored here are owned by the DescriptorPool above.
   // Python references to classes are owned by this PyDescriptorPool.
-  typedef hash_map<const Descriptor*, PyObject*> ClassesByMessageMap;
+  typedef hash_map<const Descriptor*, CMessageClass*> ClassesByMessageMap;
   ClassesByMessageMap* classes_by_descriptor;
 
   // Cache the options for any kind of descriptor.
@@ -101,14 +104,14 @@ const Descriptor* FindMessageTypeByName(PyDescriptorPool* self,
 // On error, returns -1 with a Python exception set.
 int RegisterMessageClass(PyDescriptorPool* self,
                          const Descriptor* message_descriptor,
-                         PyObject* message_class);
+                         CMessageClass* message_class);
 
 // Retrieves the Python class registered with the given message descriptor.
 //
 // Returns a *borrowed* reference if found, otherwise returns NULL with an
 // exception set.
-PyObject* GetMessageClass(PyDescriptorPool* self,
-                          const Descriptor* message_descriptor);
+CMessageClass* GetMessageClass(PyDescriptorPool* self,
+                               const Descriptor* message_descriptor);
 
 // The functions below are also exposed as methods of the DescriptorPool type.
 
diff --git a/python/google/protobuf/pyext/extension_dict.cc b/python/google/protobuf/pyext/extension_dict.cc
index 555bd293..21bbb8c2 100644
--- a/python/google/protobuf/pyext/extension_dict.cc
+++ b/python/google/protobuf/pyext/extension_dict.cc
@@ -130,7 +130,7 @@ PyObject* subscript(ExtensionDict* self, PyObject* key) {
 
   if (descriptor->label() == FieldDescriptor::LABEL_REPEATED) {
     if (descriptor->cpp_type() == FieldDescriptor::CPPTYPE_MESSAGE) {
-      PyObject *message_class = cdescriptor_pool::GetMessageClass(
+      CMessageClass* message_class = cdescriptor_pool::GetMessageClass(
           cmessage::GetDescriptorPoolForMessage(self->parent),
           descriptor->message_type());
       if (message_class == NULL) {
@@ -239,6 +239,21 @@ PyObject* _FindExtensionByName(ExtensionDict* self, PyObject* name) {
   }
 }
 
+PyObject* _FindExtensionByNumber(ExtensionDict* self, PyObject* number) {
+  ScopedPyObjectPtr extensions_by_number(PyObject_GetAttrString(
+      reinterpret_cast<PyObject*>(self->parent), "_extensions_by_number"));
+  if (extensions_by_number == NULL) {
+    return NULL;
+  }
+  PyObject* result = PyDict_GetItem(extensions_by_number.get(), number);
+  if (result == NULL) {
+    Py_RETURN_NONE;
+  } else {
+    Py_INCREF(result);
+    return result;
+  }
+}
+
 ExtensionDict* NewExtensionDict(CMessage *parent) {
   ExtensionDict* self = reinterpret_cast<ExtensionDict*>(
       PyType_GenericAlloc(&ExtensionDict_Type, 0));
@@ -271,6 +286,8 @@ static PyMethodDef Methods[] = {
   EDMETHOD(HasExtension, METH_O, "Checks if the object has an extension."),
   EDMETHOD(_FindExtensionByName, METH_O,
            "Finds an extension by name."),
+  EDMETHOD(_FindExtensionByNumber, METH_O,
+           "Finds an extension by field number."),
   { NULL, NULL }
 };
 
diff --git a/python/google/protobuf/pyext/extension_dict.h b/python/google/protobuf/pyext/extension_dict.h
index 1e7f6f7b..049d2e45 100644
--- a/python/google/protobuf/pyext/extension_dict.h
+++ b/python/google/protobuf/pyext/extension_dict.h
@@ -123,6 +123,12 @@ PyObject* ClearExtension(ExtensionDict* self,
 // Returns a new reference.
 PyObject* _FindExtensionByName(ExtensionDict* self, PyObject* name);
 
+// Gets an extension from the dict given the extension field number as
+// opposed to descriptor.
+//
+// Returns a new reference.
+PyObject* _FindExtensionByNumber(ExtensionDict* self, PyObject* number);
+
 }  // namespace extension_dict
 }  // namespace python
 }  // namespace protobuf
diff --git a/python/google/protobuf/pyext/map_container.cc b/python/google/protobuf/pyext/map_container.cc
index df9138a4..e022406d 100644
--- a/python/google/protobuf/pyext/map_container.cc
+++ b/python/google/protobuf/pyext/map_container.cc
@@ -32,6 +32,11 @@
 
 #include <google/protobuf/pyext/map_container.h>
 
+#include <memory>
+#ifndef _SHARED_PTR_H
+#include <google/protobuf/stubs/shared_ptr.h>
+#endif
+
 #include <google/protobuf/stubs/logging.h>
 #include <google/protobuf/stubs/common.h>
 #include <google/protobuf/stubs/scoped_ptr.h>
@@ -70,7 +75,7 @@ class MapReflectionFriend {
 struct MapIterator {
   PyObject_HEAD;
 
-  scoped_ptr< ::google::protobuf::MapIterator> iter;
+  google::protobuf::scoped_ptr< ::google::protobuf::MapIterator> iter;
 
   // A pointer back to the container, so we can notice changes to the version.
   // We own a ref on this.
@@ -610,8 +615,7 @@ static PyObject* GetCMessage(MessageMapContainer* self, Message* message) {
   PyObject* ret = PyDict_GetItem(self->message_dict, key.get());
 
   if (ret == NULL) {
-    CMessage* cmsg = cmessage::NewEmptyMessage(self->subclass_init,
-                                               message->GetDescriptor());
+    CMessage* cmsg = cmessage::NewEmptyMessage(self->message_class);
     ret = reinterpret_cast<PyObject*>(cmsg);
 
     if (cmsg == NULL) {
@@ -634,7 +638,7 @@ static PyObject* GetCMessage(MessageMapContainer* self, Message* message) {
 
 PyObject* NewMessageMapContainer(
     CMessage* parent, const google::protobuf::FieldDescriptor* parent_field_descriptor,
-    PyObject* concrete_class) {
+    CMessageClass* message_class) {
   if (!CheckFieldBelongsToMessage(parent_field_descriptor, parent->message)) {
     return NULL;
   }
@@ -669,8 +673,8 @@ PyObject* NewMessageMapContainer(
                         "Could not allocate message dict.");
   }
 
-  Py_INCREF(concrete_class);
-  self->subclass_init = concrete_class;
+  Py_INCREF(message_class);
+  self->message_class = message_class;
 
   if (self->key_field_descriptor == NULL ||
       self->value_field_descriptor == NULL) {
@@ -763,6 +767,7 @@ static void MessageMapDealloc(PyObject* _self) {
   MessageMapContainer* self = GetMessageMap(_self);
   self->owner.reset();
   Py_DECREF(self->message_dict);
+  Py_DECREF(self->message_class);
   Py_TYPE(_self)->tp_free(_self);
 }
 
diff --git a/python/google/protobuf/pyext/map_container.h b/python/google/protobuf/pyext/map_container.h
index 27ee6dbd..b11dfa34 100644
--- a/python/google/protobuf/pyext/map_container.h
+++ b/python/google/protobuf/pyext/map_container.h
@@ -55,6 +55,7 @@ using internal::shared_ptr;
 namespace python {
 
 struct CMessage;
+struct CMessageClass;
 
 // This struct is used directly for ScalarMap, and is the base class of
 // MessageMapContainer, which is used for MessageMap.
@@ -104,8 +105,8 @@ struct MapContainer {
 };
 
 struct MessageMapContainer : public MapContainer {
-  // A callable that is used to create new child messages.
-  PyObject* subclass_init;
+  // The type used to create new child messages.
+  CMessageClass* message_class;
 
   // A dict mapping Message* -> CMessage.
   PyObject* message_dict;
@@ -132,7 +133,7 @@ extern PyObject* NewScalarMapContainer(
 // field descriptor.
 extern PyObject* NewMessageMapContainer(
     CMessage* parent, const FieldDescriptor* parent_field_descriptor,
-    PyObject* concrete_class);
+    CMessageClass* message_class);
 
 }  // namespace python
 }  // namespace protobuf
diff --git a/python/google/protobuf/pyext/message.cc b/python/google/protobuf/pyext/message.cc
index 6d7b2b0f..83c151ff 100644
--- a/python/google/protobuf/pyext/message.cc
+++ b/python/google/protobuf/pyext/message.cc
@@ -98,31 +98,6 @@ static PyObject* PythonMessage_class;
 static PyObject* kEmptyWeakref;
 static PyObject* WKT_classes = NULL;
 
-// Defines the Metaclass of all Message classes.
-// It allows us to cache some C++ pointers in the class object itself, they are
-// faster to extract than from the type's dictionary.
-
-struct PyMessageMeta {
-  // This is how CPython subclasses C structures: the base structure must be
-  // the first member of the object.
-  PyHeapTypeObject super;
-
-  // C++ descriptor of this message.
-  const Descriptor* message_descriptor;
-
-  // Owned reference, used to keep the pointer above alive.
-  PyObject* py_message_descriptor;
-
-  // The Python DescriptorPool used to create the class. It is needed to resolve
-  // fields descriptors, including extensions fields; its C++ MessageFactory is
-  // used to instantiate submessages.
-  // This can be different from DESCRIPTOR.file.pool, in the case of a custom
-  // DescriptorPool which defines new extensions.
-  // We own the reference, because it's important to keep the descriptors and
-  // factory alive.
-  PyDescriptorPool* py_descriptor_pool;
-};
-
 namespace message_meta {
 
 static int InsertEmptyWeakref(PyTypeObject* base);
@@ -173,10 +148,6 @@ static int AddDescriptors(PyObject* cls, const Descriptor* descriptor) {
   }
 
   // For each enum set cls.<enum name> = EnumTypeWrapper(<enum descriptor>).
-  //
-  // The enum descriptor we get from
-  // <messagedescriptor>.enum_types_by_name[name]
-  // which was built previously.
   for (int i = 0; i < descriptor->enum_type_count(); ++i) {
     const EnumDescriptor* enum_descriptor = descriptor->enum_type(i);
     ScopedPyObjectPtr enum_type(
@@ -309,7 +280,7 @@ static PyObject* New(PyTypeObject* type,
   if (result == NULL) {
     return NULL;
   }
-  PyMessageMeta* newtype = reinterpret_cast<PyMessageMeta*>(result.get());
+  CMessageClass* newtype = reinterpret_cast<CMessageClass*>(result.get());
 
   // Insert the empty weakref into the base classes.
   if (InsertEmptyWeakref(
@@ -338,7 +309,7 @@ static PyObject* New(PyTypeObject* type,
 
   // Add the message to the DescriptorPool.
   if (cdescriptor_pool::RegisterMessageClass(newtype->py_descriptor_pool,
-                                             descriptor, result.get()) < 0) {
+                                             descriptor, newtype) < 0) {
     return NULL;
   }
 
@@ -349,7 +320,7 @@ static PyObject* New(PyTypeObject* type,
   return result.release();
 }
 
-static void Dealloc(PyMessageMeta *self) {
+static void Dealloc(CMessageClass *self) {
   Py_DECREF(self->py_message_descriptor);
   Py_DECREF(self->py_descriptor_pool);
   Py_TYPE(self)->tp_free(reinterpret_cast<PyObject*>(self));
@@ -378,10 +349,10 @@ static int InsertEmptyWeakref(PyTypeObject *base_type) {
 
 }  // namespace message_meta
 
-PyTypeObject PyMessageMeta_Type = {
+PyTypeObject CMessageClass_Type = {
   PyVarObject_HEAD_INIT(&PyType_Type, 0)
   FULL_MODULE_NAME ".MessageMeta",     // tp_name
-  sizeof(PyMessageMeta),               // tp_basicsize
+  sizeof(CMessageClass),               // tp_basicsize
   0,                                   // tp_itemsize
   (destructor)message_meta::Dealloc,   // tp_dealloc
   0,                                   // tp_print
@@ -419,16 +390,16 @@ PyTypeObject PyMessageMeta_Type = {
   message_meta::New,                   // tp_new
 };
 
-static PyMessageMeta* CheckMessageClass(PyTypeObject* cls) {
-  if (!PyObject_TypeCheck(cls, &PyMessageMeta_Type)) {
+static CMessageClass* CheckMessageClass(PyTypeObject* cls) {
+  if (!PyObject_TypeCheck(cls, &CMessageClass_Type)) {
     PyErr_Format(PyExc_TypeError, "Class %s is not a Message", cls->tp_name);
     return NULL;
   }
-  return reinterpret_cast<PyMessageMeta*>(cls);
+  return reinterpret_cast<CMessageClass*>(cls);
 }
 
 static const Descriptor* GetMessageDescriptor(PyTypeObject* cls) {
-  PyMessageMeta* type = CheckMessageClass(cls);
+  CMessageClass* type = CheckMessageClass(cls);
   if (type == NULL) {
     return NULL;
   }
@@ -783,9 +754,9 @@ namespace cmessage {
 
 PyDescriptorPool* GetDescriptorPoolForMessage(CMessage* message) {
   // No need to check the type: the type of instances of CMessage is always
-  // an instance of PyMessageMeta. Let's prove it with a debug-only check.
+  // an instance of CMessageClass. Let's prove it with a debug-only check.
   GOOGLE_DCHECK(PyObject_TypeCheck(message, &CMessage_Type));
-  return reinterpret_cast<PyMessageMeta*>(Py_TYPE(message))->py_descriptor_pool;
+  return reinterpret_cast<CMessageClass*>(Py_TYPE(message))->py_descriptor_pool;
 }
 
 MessageFactory* GetFactoryForMessage(CMessage* message) {
@@ -1090,6 +1061,10 @@ int InitAttributes(CMessage* self, PyObject* kwargs) {
                    PyString_AsString(name));
       return -1;
     }
+    if (value == Py_None) {
+      // field=None is the same as no field at all.
+      continue;
+    }
     if (descriptor->is_map()) {
       ScopedPyObjectPtr map(GetAttr(self, name));
       const FieldDescriptor* value_descriptor =
@@ -1220,9 +1195,9 @@ int InitAttributes(CMessage* self, PyObject* kwargs) {
 
 // Allocates an incomplete Python Message: the caller must fill self->message,
 // self->owner and eventually self->parent.
-CMessage* NewEmptyMessage(PyObject* type, const Descriptor *descriptor) {
+CMessage* NewEmptyMessage(CMessageClass* type) {
   CMessage* self = reinterpret_cast<CMessage*>(
-      PyType_GenericAlloc(reinterpret_cast<PyTypeObject*>(type), 0));
+      PyType_GenericAlloc(&type->super.ht_type, 0));
   if (self == NULL) {
     return NULL;
   }
@@ -1242,7 +1217,7 @@ CMessage* NewEmptyMessage(PyObject* type, const Descriptor *descriptor) {
 // Creates a new C++ message and takes ownership.
 static PyObject* New(PyTypeObject* cls,
                      PyObject* unused_args, PyObject* unused_kwargs) {
-  PyMessageMeta* type = CheckMessageClass(cls);
+  CMessageClass* type = CheckMessageClass(cls);
   if (type == NULL) {
     return NULL;
   }
@@ -1258,8 +1233,7 @@ static PyObject* New(PyTypeObject* cls,
     return NULL;
   }
 
-  CMessage* self = NewEmptyMessage(reinterpret_cast<PyObject*>(type),
-                                   message_descriptor);
+  CMessage* self = NewEmptyMessage(type);
   if (self == NULL) {
     return NULL;
   }
@@ -2023,10 +1997,34 @@ static PyObject* RegisterExtension(PyObject* cls,
     PyErr_SetString(PyExc_TypeError, "no extensions_by_number on class");
     return NULL;
   }
+
   ScopedPyObjectPtr number(PyObject_GetAttrString(extension_handle, "number"));
   if (number == NULL) {
     return NULL;
   }
+
+  // If the extension was already registered by number, check that it is the
+  // same.
+  existing_extension = PyDict_GetItem(extensions_by_number.get(), number.get());
+  if (existing_extension != NULL) {
+    const FieldDescriptor* existing_extension_descriptor =
+        GetExtensionDescriptor(existing_extension);
+    if (existing_extension_descriptor != descriptor) {
+      const Descriptor* msg_desc = GetMessageDescriptor(
+          reinterpret_cast<PyTypeObject*>(cls));
+      PyErr_Format(
+          PyExc_ValueError,
+          "Extensions \"%s\" and \"%s\" both try to extend message type "
+          "\"%s\" with field number %ld.",
+          existing_extension_descriptor->full_name().c_str(),
+          descriptor->full_name().c_str(),
+          msg_desc->full_name().c_str(),
+          PyInt_AsLong(number.get()));
+      return NULL;
+    }
+    // Nothing else to do.
+    Py_RETURN_NONE;
+  }
   if (PyDict_SetItem(extensions_by_number.get(), number.get(),
                      extension_handle) < 0) {
     return NULL;
@@ -2166,6 +2164,12 @@ static PyObject* ListFields(CMessage* self) {
   return all_fields.release();
 }
 
+static PyObject* DiscardUnknownFields(CMessage* self) {
+  AssureWritable(self);
+  self->message->DiscardUnknownFields();
+  Py_RETURN_NONE;
+}
+
 PyObject* FindInitializationErrors(CMessage* self) {
   Message* message = self->message;
   vector<string> errors;
@@ -2309,14 +2313,13 @@ PyObject* InternalGetSubMessage(
   const Message& sub_message = reflection->GetMessage(
       *self->message, field_descriptor, pool->message_factory);
 
-  PyObject *message_class = cdescriptor_pool::GetMessageClass(
+  CMessageClass* message_class = cdescriptor_pool::GetMessageClass(
       pool, field_descriptor->message_type());
   if (message_class == NULL) {
     return NULL;
   }
 
-  CMessage* cmsg = cmessage::NewEmptyMessage(message_class,
-                                             sub_message.GetDescriptor());
+  CMessage* cmsg = cmessage::NewEmptyMessage(message_class);
   if (cmsg == NULL) {
     return NULL;
   }
@@ -2585,6 +2588,8 @@ static PyMethodDef Methods[] = {
     "Clears a message field." },
   { "CopyFrom", (PyCFunction)CopyFrom, METH_O,
     "Copies a protocol message into the current message." },
+  { "DiscardUnknownFields", (PyCFunction)DiscardUnknownFields, METH_NOARGS,
+    "Discards the unknown fields." },
   { "FindInitializationErrors", (PyCFunction)FindInitializationErrors,
     METH_NOARGS,
     "Finds unset required fields." },
@@ -2654,7 +2659,7 @@ PyObject* GetAttr(CMessage* self, PyObject* name) {
     const Descriptor* entry_type = field_descriptor->message_type();
     const FieldDescriptor* value_type = entry_type->FindFieldByName("value");
     if (value_type->cpp_type() == FieldDescriptor::CPPTYPE_MESSAGE) {
-      PyObject* value_class = cdescriptor_pool::GetMessageClass(
+      CMessageClass* value_class = cdescriptor_pool::GetMessageClass(
           GetDescriptorPoolForMessage(self), value_type->message_type());
       if (value_class == NULL) {
         return NULL;
@@ -2677,7 +2682,7 @@ PyObject* GetAttr(CMessage* self, PyObject* name) {
   if (field_descriptor->label() == FieldDescriptor::LABEL_REPEATED) {
     PyObject* py_container = NULL;
     if (field_descriptor->cpp_type() == FieldDescriptor::CPPTYPE_MESSAGE) {
-      PyObject *message_class = cdescriptor_pool::GetMessageClass(
+      CMessageClass* message_class = cdescriptor_pool::GetMessageClass(
           GetDescriptorPoolForMessage(self), field_descriptor->message_type());
       if (message_class == NULL) {
         return NULL;
@@ -2749,7 +2754,7 @@ int SetAttr(CMessage* self, PyObject* name, PyObject* value) {
 }  // namespace cmessage
 
 PyTypeObject CMessage_Type = {
-  PyVarObject_HEAD_INIT(&PyMessageMeta_Type, 0)
+  PyVarObject_HEAD_INIT(&CMessageClass_Type, 0)
   FULL_MODULE_NAME ".CMessage",        // tp_name
   sizeof(CMessage),                    // tp_basicsize
   0,                                   //  tp_itemsize
@@ -2864,12 +2869,12 @@ bool InitProto2MessageModule(PyObject *m) {
   // Initialize constants defined in this file.
   InitGlobals();
 
-  PyMessageMeta_Type.tp_base = &PyType_Type;
-  if (PyType_Ready(&PyMessageMeta_Type) < 0) {
+  CMessageClass_Type.tp_base = &PyType_Type;
+  if (PyType_Ready(&CMessageClass_Type) < 0) {
     return false;
   }
   PyModule_AddObject(m, "MessageMeta",
-                     reinterpret_cast<PyObject*>(&PyMessageMeta_Type));
+                     reinterpret_cast<PyObject*>(&CMessageClass_Type));
 
   if (PyType_Ready(&CMessage_Type) < 0) {
     return false;
@@ -3077,9 +3082,10 @@ bool InitProto2MessageModule(PyObject *m) {
 }  // namespace protobuf
 
 static PyMethodDef ModuleMethods[] = {
-    {"SetAllowOversizeProtos",
-     (PyCFunction)google::protobuf::python::cmessage::SetAllowOversizeProtos,
-     METH_O, "Enable/disable oversize proto parsing."},
+  {"SetAllowOversizeProtos",
+    (PyCFunction)google::protobuf::python::cmessage::SetAllowOversizeProtos,
+    METH_O, "Enable/disable oversize proto parsing."},
+  { NULL, NULL}
 };
 
 #if PY_MAJOR_VERSION >= 3
diff --git a/python/google/protobuf/pyext/message.h b/python/google/protobuf/pyext/message.h
index c2b62649..9dce198f 100644
--- a/python/google/protobuf/pyext/message.h
+++ b/python/google/protobuf/pyext/message.h
@@ -116,12 +116,43 @@ typedef struct CMessage {
 
 extern PyTypeObject CMessage_Type;
 
+
+// The (meta) type of all Messages classes.
+// It allows us to cache some C++ pointers in the class object itself, they are
+// faster to extract than from the type's dictionary.
+
+struct CMessageClass {
+  // This is how CPython subclasses C structures: the base structure must be
+  // the first member of the object.
+  PyHeapTypeObject super;
+
+  // C++ descriptor of this message.
+  const Descriptor* message_descriptor;
+
+  // Owned reference, used to keep the pointer above alive.
+  PyObject* py_message_descriptor;
+
+  // The Python DescriptorPool used to create the class. It is needed to resolve
+  // fields descriptors, including extensions fields; its C++ MessageFactory is
+  // used to instantiate submessages.
+  // This can be different from DESCRIPTOR.file.pool, in the case of a custom
+  // DescriptorPool which defines new extensions.
+  // We own the reference, because it's important to keep the descriptors and
+  // factory alive.
+  PyDescriptorPool* py_descriptor_pool;
+
+  PyObject* AsPyObject() {
+    return reinterpret_cast<PyObject*>(this);
+  }
+};
+
+
 namespace cmessage {
 
 // Internal function to create a new empty Message Python object, but with empty
 // pointers to the C++ objects.
 // The caller must fill self->message, self->owner and eventually self->parent.
-CMessage* NewEmptyMessage(PyObject* type, const Descriptor* descriptor);
+CMessage* NewEmptyMessage(CMessageClass* type);
 
 // Release a submessage from its proto tree, making it a new top-level messgae.
 // A new message will be created if this is a read-only default instance.
diff --git a/python/google/protobuf/pyext/repeated_composite_container.cc b/python/google/protobuf/pyext/repeated_composite_container.cc
index b01123b4..4f339e77 100644
--- a/python/google/protobuf/pyext/repeated_composite_container.cc
+++ b/python/google/protobuf/pyext/repeated_composite_container.cc
@@ -107,8 +107,7 @@ static int UpdateChildMessages(RepeatedCompositeContainer* self) {
   for (Py_ssize_t i = child_length; i < message_length; ++i) {
     const Message& sub_message = reflection->GetRepeatedMessage(
         *(self->message), self->parent_field_descriptor, i);
-    CMessage* cmsg = cmessage::NewEmptyMessage(self->subclass_init,
-                                               sub_message.GetDescriptor());
+    CMessage* cmsg = cmessage::NewEmptyMessage(self->child_message_class);
     ScopedPyObjectPtr py_cmsg(reinterpret_cast<PyObject*>(cmsg));
     if (cmsg == NULL) {
       return -1;
@@ -140,8 +139,7 @@ static PyObject* AddToAttached(RepeatedCompositeContainer* self,
   Message* sub_message =
       message->GetReflection()->AddMessage(message,
                                            self->parent_field_descriptor);
-  CMessage* cmsg = cmessage::NewEmptyMessage(self->subclass_init,
-                                             sub_message->GetDescriptor());
+  CMessage* cmsg = cmessage::NewEmptyMessage(self->child_message_class);
   if (cmsg == NULL)
     return NULL;
 
@@ -168,7 +166,7 @@ static PyObject* AddToReleased(RepeatedCompositeContainer* self,
 
   // Create a new Message detached from the rest.
   PyObject* py_cmsg = PyEval_CallObjectWithKeywords(
-      self->subclass_init, NULL, kwargs);
+      self->child_message_class->AsPyObject(), NULL, kwargs);
   if (py_cmsg == NULL)
     return NULL;
 
@@ -506,7 +504,7 @@ int SetOwner(RepeatedCompositeContainer* self,
 PyObject *NewContainer(
     CMessage* parent,
     const FieldDescriptor* parent_field_descriptor,
-    PyObject *concrete_class) {
+    CMessageClass* concrete_class) {
   if (!CheckFieldBelongsToMessage(parent_field_descriptor, parent->message)) {
     return NULL;
   }
@@ -523,7 +521,7 @@ PyObject *NewContainer(
   self->parent_field_descriptor = parent_field_descriptor;
   self->owner = parent->owner;
   Py_INCREF(concrete_class);
-  self->subclass_init = concrete_class;
+  self->child_message_class = concrete_class;
   self->child_messages = PyList_New(0);
 
   return reinterpret_cast<PyObject*>(self);
@@ -531,7 +529,7 @@ PyObject *NewContainer(
 
 static void Dealloc(RepeatedCompositeContainer* self) {
   Py_CLEAR(self->child_messages);
-  Py_CLEAR(self->subclass_init);
+  Py_CLEAR(self->child_message_class);
   // TODO(tibell): Do we need to call delete on these objects to make
   // sure their destructors are called?
   self->owner.reset();
diff --git a/python/google/protobuf/pyext/repeated_composite_container.h b/python/google/protobuf/pyext/repeated_composite_container.h
index 442ce7e3..25463037 100644
--- a/python/google/protobuf/pyext/repeated_composite_container.h
+++ b/python/google/protobuf/pyext/repeated_composite_container.h
@@ -58,6 +58,7 @@ using internal::shared_ptr;
 namespace python {
 
 struct CMessage;
+struct CMessageClass;
 
 // A RepeatedCompositeContainer can be in one of two states: attached
 // or released.
@@ -94,8 +95,8 @@ typedef struct RepeatedCompositeContainer {
   // calling Clear() or ClearField() on the parent.
   Message* message;
 
-  // A callable that is used to create new child messages.
-  PyObject* subclass_init;
+  // The type used to create new child messages.
+  CMessageClass* child_message_class;
 
   // A list of child messages.
   PyObject* child_messages;
@@ -110,7 +111,7 @@ namespace repeated_composite_container {
 PyObject *NewContainer(
     CMessage* parent,
     const FieldDescriptor* parent_field_descriptor,
-    PyObject *concrete_class);
+    CMessageClass *child_message_class);
 
 // Appends a new CMessage to the container and returns it.  The
 // CMessage is initialized using the content of kwargs.
diff --git a/python/google/protobuf/text_format.py b/python/google/protobuf/text_format.py
index a6f41ca8..6f1e3c8b 100755
--- a/python/google/protobuf/text_format.py
+++ b/python/google/protobuf/text_format.py
@@ -99,7 +99,7 @@ class TextWriter(object):
 
 def MessageToString(message, as_utf8=False, as_one_line=False,
                     pointy_brackets=False, use_index_order=False,
-                    float_format=None):
+                    float_format=None, use_field_number=False):
   """Convert protobuf message to text format.
 
   Floating point values can be formatted compactly with 15 digits of
@@ -118,15 +118,16 @@ def MessageToString(message, as_utf8=False, as_one_line=False,
       field number order.
     float_format: If set, use this to specify floating point number formatting
       (per the "Format Specification Mini-Language"); otherwise, str() is used.
+    use_field_number: If True, print field numbers instead of names.
 
   Returns:
     A string of the text formatted protocol buffer message.
   """
   out = TextWriter(as_utf8)
-  PrintMessage(message, out, as_utf8=as_utf8, as_one_line=as_one_line,
-               pointy_brackets=pointy_brackets,
-               use_index_order=use_index_order,
-               float_format=float_format)
+  printer = _Printer(out, 0, as_utf8, as_one_line,
+                     pointy_brackets, use_index_order, float_format,
+                     use_field_number)
+  printer.PrintMessage(message)
   result = out.getvalue()
   out.close()
   if as_one_line:
@@ -142,133 +143,187 @@ def _IsMapEntry(field):
 
 def PrintMessage(message, out, indent=0, as_utf8=False, as_one_line=False,
                  pointy_brackets=False, use_index_order=False,
-                 float_format=None):
-  fields = message.ListFields()
-  if use_index_order:
-    fields.sort(key=lambda x: x[0].index)
-  for field, value in fields:
-    if _IsMapEntry(field):
-      for key in sorted(value):
-        # This is slow for maps with submessage entires because it copies the
-        # entire tree.  Unfortunately this would take significant refactoring
-        # of this file to work around.
-        #
-        # TODO(haberman): refactor and optimize if this becomes an issue.
-        entry_submsg = field.message_type._concrete_class(
-            key=key, value=value[key])
-        PrintField(field, entry_submsg, out, indent, as_utf8, as_one_line,
-                   pointy_brackets=pointy_brackets,
-                   use_index_order=use_index_order, float_format=float_format)
-    elif field.label == descriptor.FieldDescriptor.LABEL_REPEATED:
-      for element in value:
-        PrintField(field, element, out, indent, as_utf8, as_one_line,
-                   pointy_brackets=pointy_brackets,
-                   use_index_order=use_index_order,
-                   float_format=float_format)
-    else:
-      PrintField(field, value, out, indent, as_utf8, as_one_line,
-                 pointy_brackets=pointy_brackets,
-                 use_index_order=use_index_order,
-                 float_format=float_format)
+                 float_format=None, use_field_number=False):
+  printer = _Printer(out, indent, as_utf8, as_one_line,
+                     pointy_brackets, use_index_order, float_format,
+                     use_field_number)
+  printer.PrintMessage(message)
 
 
 def PrintField(field, value, out, indent=0, as_utf8=False, as_one_line=False,
                pointy_brackets=False, use_index_order=False, float_format=None):
-  """Print a single field name/value pair.  For repeated fields, the value
-  should be a single element.
-  """
-
-  out.write(' ' * indent)
-  if field.is_extension:
-    out.write('[')
-    if (field.containing_type.GetOptions().message_set_wire_format and
-        field.type == descriptor.FieldDescriptor.TYPE_MESSAGE and
-        field.label == descriptor.FieldDescriptor.LABEL_OPTIONAL):
-      out.write(field.message_type.full_name)
-    else:
-      out.write(field.full_name)
-    out.write(']')
-  elif field.type == descriptor.FieldDescriptor.TYPE_GROUP:
-    # For groups, use the capitalized name.
-    out.write(field.message_type.name)
-  else:
-    out.write(field.name)
-
-  if field.cpp_type != descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
-    # The colon is optional in this case, but our cross-language golden files
-    # don't include it.
-    out.write(': ')
-
-  PrintFieldValue(field, value, out, indent, as_utf8, as_one_line,
-                  pointy_brackets=pointy_brackets,
-                  use_index_order=use_index_order,
-                  float_format=float_format)
-  if as_one_line:
-    out.write(' ')
-  else:
-    out.write('\n')
+  """Print a single field name/value pair."""
+  printer = _Printer(out, indent, as_utf8, as_one_line,
+                     pointy_brackets, use_index_order, float_format)
+  printer.PrintField(field, value)
 
 
 def PrintFieldValue(field, value, out, indent=0, as_utf8=False,
                     as_one_line=False, pointy_brackets=False,
                     use_index_order=False,
                     float_format=None):
-  """Print a single field value (not including name).  For repeated fields,
-  the value should be a single element."""
+  """Print a single field value (not including name)."""
+  printer = _Printer(out, indent, as_utf8, as_one_line,
+                     pointy_brackets, use_index_order, float_format)
+  printer.PrintFieldValue(field, value)
 
-  if pointy_brackets:
-    openb = '<'
-    closeb = '>'
-  else:
-    openb = '{'
-    closeb = '}'
-
-  if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
-    if as_one_line:
-      out.write(' %s ' % openb)
-      PrintMessage(value, out, indent, as_utf8, as_one_line,
-                   pointy_brackets=pointy_brackets,
-                   use_index_order=use_index_order,
-                   float_format=float_format)
-      out.write(closeb)
-    else:
-      out.write(' %s\n' % openb)
-      PrintMessage(value, out, indent + 2, as_utf8, as_one_line,
-                   pointy_brackets=pointy_brackets,
-                   use_index_order=use_index_order,
-                   float_format=float_format)
-      out.write(' ' * indent + closeb)
-  elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_ENUM:
-    enum_value = field.enum_type.values_by_number.get(value, None)
-    if enum_value is not None:
-      out.write(enum_value.name)
+
+class _Printer(object):
+  """Text format printer for protocol message."""
+
+  def __init__(self, out, indent=0, as_utf8=False, as_one_line=False,
+               pointy_brackets=False, use_index_order=False, float_format=None,
+               use_field_number=False):
+    """Initialize the Printer.
+
+    Floating point values can be formatted compactly with 15 digits of
+    precision (which is the most that IEEE 754 "double" can guarantee)
+    using float_format='.15g'. To ensure that converting to text and back to a
+    proto will result in an identical value, float_format='.17g' should be used.
+
+    Args:
+      out: To record the text format result.
+      indent: The indent level for pretty print.
+      as_utf8: Produce text output in UTF8 format.
+      as_one_line: Don't introduce newlines between fields.
+      pointy_brackets: If True, use angle brackets instead of curly braces for
+        nesting.
+      use_index_order: If True, print fields of a proto message using the order
+        defined in source code instead of the field number. By default, use the
+        field number order.
+      float_format: If set, use this to specify floating point number formatting
+        (per the "Format Specification Mini-Language"); otherwise, str() is
+        used.
+      use_field_number: If True, print field numbers instead of names.
+    """
+    self.out = out
+    self.indent = indent
+    self.as_utf8 = as_utf8
+    self.as_one_line = as_one_line
+    self.pointy_brackets = pointy_brackets
+    self.use_index_order = use_index_order
+    self.float_format = float_format
+    self.use_field_number = use_field_number
+
+  def PrintMessage(self, message):
+    """Convert protobuf message to text format.
+
+    Args:
+      message: The protocol buffers message.
+    """
+    fields = message.ListFields()
+    if self.use_index_order:
+      fields.sort(key=lambda x: x[0].index)
+    for field, value in fields:
+      if _IsMapEntry(field):
+        for key in sorted(value):
+          # This is slow for maps with submessage entires because it copies the
+          # entire tree.  Unfortunately this would take significant refactoring
+          # of this file to work around.
+          #
+          # TODO(haberman): refactor and optimize if this becomes an issue.
+          entry_submsg = field.message_type._concrete_class(
+              key=key, value=value[key])
+          self.PrintField(field, entry_submsg)
+      elif field.label == descriptor.FieldDescriptor.LABEL_REPEATED:
+        for element in value:
+          self.PrintField(field, element)
+      else:
+        self.PrintField(field, value)
+
+  def PrintField(self, field, value):
+    """Print a single field name/value pair."""
+    out = self.out
+    out.write(' ' * self.indent)
+    if self.use_field_number:
+      out.write(str(field.number))
     else:
-      out.write(str(value))
-  elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_STRING:
-    out.write('\"')
-    if isinstance(value, six.text_type):
-      out_value = value.encode('utf-8')
+      if field.is_extension:
+        out.write('[')
+        if (field.containing_type.GetOptions().message_set_wire_format and
+            field.type == descriptor.FieldDescriptor.TYPE_MESSAGE and
+            field.label == descriptor.FieldDescriptor.LABEL_OPTIONAL):
+          out.write(field.message_type.full_name)
+        else:
+          out.write(field.full_name)
+        out.write(']')
+      elif field.type == descriptor.FieldDescriptor.TYPE_GROUP:
+        # For groups, use the capitalized name.
+        out.write(field.message_type.name)
+      else:
+        out.write(field.name)
+
+    if field.cpp_type != descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
+      # The colon is optional in this case, but our cross-language golden files
+      # don't include it.
+      out.write(': ')
+
+    self.PrintFieldValue(field, value)
+    if self.as_one_line:
+      out.write(' ')
     else:
-      out_value = value
-    if field.type == descriptor.FieldDescriptor.TYPE_BYTES:
-      # We need to escape non-UTF8 chars in TYPE_BYTES field.
-      out_as_utf8 = False
+      out.write('\n')
+
+  def PrintFieldValue(self, field, value):
+    """Print a single field value (not including name).
+
+    For repeated fields, the value should be a single element.
+
+    Args:
+      field: The descriptor of the field to be printed.
+      value: The value of the field.
+    """
+    out = self.out
+    if self.pointy_brackets:
+      openb = '<'
+      closeb = '>'
     else:
-      out_as_utf8 = as_utf8
-    out.write(text_encoding.CEscape(out_value, out_as_utf8))
-    out.write('\"')
-  elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_BOOL:
-    if value:
-      out.write('true')
+      openb = '{'
+      closeb = '}'
+
+    if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
+      if self.as_one_line:
+        out.write(' %s ' % openb)
+        self.PrintMessage(value)
+        out.write(closeb)
+      else:
+        out.write(' %s\n' % openb)
+        self.indent += 2
+        self.PrintMessage(value)
+        self.indent -= 2
+        out.write(' ' * self.indent + closeb)
+    elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_ENUM:
+      enum_value = field.enum_type.values_by_number.get(value, None)
+      if enum_value is not None:
+        out.write(enum_value.name)
+      else:
+        out.write(str(value))
+    elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_STRING:
+      out.write('\"')
+      if isinstance(value, six.text_type):
+        out_value = value.encode('utf-8')
+      else:
+        out_value = value
+      if field.type == descriptor.FieldDescriptor.TYPE_BYTES:
+        # We need to escape non-UTF8 chars in TYPE_BYTES field.
+        out_as_utf8 = False
+      else:
+        out_as_utf8 = self.as_utf8
+      out.write(text_encoding.CEscape(out_value, out_as_utf8))
+      out.write('\"')
+    elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_BOOL:
+      if value:
+        out.write('true')
+      else:
+        out.write('false')
+    elif field.cpp_type in _FLOAT_TYPES and self.float_format is not None:
+      out.write('{1:{0}}'.format(self.float_format, value))
     else:
-      out.write('false')
-  elif field.cpp_type in _FLOAT_TYPES and float_format is not None:
-    out.write('{1:{0}}'.format(float_format, value))
-  else:
-    out.write(str(value))
+      out.write(str(value))
 
 
-def Parse(text, message, allow_unknown_extension=False):
+def Parse(text, message,
+          allow_unknown_extension=False, allow_field_number=False):
   """Parses an text representation of a protocol message into a message.
 
   Args:
@@ -276,6 +331,7 @@ def Parse(text, message, allow_unknown_extension=False):
     message: A protocol buffer message to merge into.
     allow_unknown_extension: if True, skip over missing extensions and keep
       parsing
+    allow_field_number: if True, both field number and field name are allowed.
 
   Returns:
     The same message passed as argument.
@@ -285,10 +341,12 @@ def Parse(text, message, allow_unknown_extension=False):
   """
   if not isinstance(text, str):
     text = text.decode('utf-8')
-  return ParseLines(text.split('\n'), message, allow_unknown_extension)
+  return ParseLines(text.split('\n'), message, allow_unknown_extension,
+                    allow_field_number)
 
 
-def Merge(text, message, allow_unknown_extension=False):
+def Merge(text, message, allow_unknown_extension=False,
+          allow_field_number=False):
   """Parses an text representation of a protocol message into a message.
 
   Like Parse(), but allows repeated values for a non-repeated field, and uses
@@ -299,6 +357,7 @@ def Merge(text, message, allow_unknown_extension=False):
     message: A protocol buffer message to merge into.
     allow_unknown_extension: if True, skip over missing extensions and keep
       parsing
+    allow_field_number: if True, both field number and field name are allowed.
 
   Returns:
     The same message passed as argument.
@@ -306,10 +365,12 @@ def Merge(text, message, allow_unknown_extension=False):
   Raises:
     ParseError: On text parsing problems.
   """
-  return MergeLines(text.split('\n'), message, allow_unknown_extension)
+  return MergeLines(text.split('\n'), message, allow_unknown_extension,
+                    allow_field_number)
 
 
-def ParseLines(lines, message, allow_unknown_extension=False):
+def ParseLines(lines, message, allow_unknown_extension=False,
+               allow_field_number=False):
   """Parses an text representation of a protocol message into a message.
 
   Args:
@@ -317,6 +378,7 @@ def ParseLines(lines, message, allow_unknown_extension=False):
     message: A protocol buffer message to merge into.
     allow_unknown_extension: if True, skip over missing extensions and keep
       parsing
+    allow_field_number: if True, both field number and field name are allowed.
 
   Returns:
     The same message passed as argument.
@@ -324,11 +386,12 @@ def ParseLines(lines, message, allow_unknown_extension=False):
   Raises:
     ParseError: On text parsing problems.
   """
-  _ParseOrMerge(lines, message, False, allow_unknown_extension)
-  return message
+  parser = _Parser(allow_unknown_extension, allow_field_number)
+  return parser.ParseLines(lines, message)
 
 
-def MergeLines(lines, message, allow_unknown_extension=False):
+def MergeLines(lines, message, allow_unknown_extension=False,
+               allow_field_number=False):
   """Parses an text representation of a protocol message into a message.
 
   Args:
@@ -336,6 +399,7 @@ def MergeLines(lines, message, allow_unknown_extension=False):
     message: A protocol buffer message to merge into.
     allow_unknown_extension: if True, skip over missing extensions and keep
       parsing
+    allow_field_number: if True, both field number and field name are allowed.
 
   Returns:
     The same message passed as argument.
@@ -343,146 +407,272 @@ def MergeLines(lines, message, allow_unknown_extension=False):
   Raises:
     ParseError: On text parsing problems.
   """
-  _ParseOrMerge(lines, message, True, allow_unknown_extension)
-  return message
+  parser = _Parser(allow_unknown_extension, allow_field_number)
+  return parser.MergeLines(lines, message)
 
 
-def _ParseOrMerge(lines,
-                  message,
-                  allow_multiple_scalars,
-                  allow_unknown_extension=False):
-  """Converts an text representation of a protocol message into a message.
+class _Parser(object):
+  """Text format parser for protocol message."""
 
-  Args:
-    lines: Lines of a message's text representation.
-    message: A protocol buffer message to merge into.
-    allow_multiple_scalars: Determines if repeated values for a non-repeated
-      field are permitted, e.g., the string "foo: 1 foo: 2" for a
-      required/optional field named "foo".
-    allow_unknown_extension: if True, skip over missing extensions and keep
-      parsing
+  def __init__(self, allow_unknown_extension=False, allow_field_number=False):
+    self.allow_unknown_extension = allow_unknown_extension
+    self.allow_field_number = allow_field_number
 
-  Raises:
-    ParseError: On text parsing problems.
-  """
-  tokenizer = _Tokenizer(lines)
-  while not tokenizer.AtEnd():
-    _MergeField(tokenizer, message, allow_multiple_scalars,
-                allow_unknown_extension)
+  def ParseFromString(self, text, message):
+    """Parses an text representation of a protocol message into a message."""
+    if not isinstance(text, str):
+      text = text.decode('utf-8')
+    return self.ParseLines(text.split('\n'), message)
 
+  def ParseLines(self, lines, message):
+    """Parses an text representation of a protocol message into a message."""
+    self._allow_multiple_scalars = False
+    self._ParseOrMerge(lines, message)
+    return message
 
-def _MergeField(tokenizer,
-                message,
-                allow_multiple_scalars,
-                allow_unknown_extension=False):
-  """Merges a single protocol message field into a message.
+  def MergeFromString(self, text, message):
+    """Merges an text representation of a protocol message into a message."""
+    return self._MergeLines(text.split('\n'), message)
 
-  Args:
-    tokenizer: A tokenizer to parse the field name and values.
-    message: A protocol message to record the data.
-    allow_multiple_scalars: Determines if repeated values for a non-repeated
-      field are permitted, e.g., the string "foo: 1 foo: 2" for a
-      required/optional field named "foo".
-    allow_unknown_extension: if True, skip over missing extensions and keep
-      parsing.
+  def MergeLines(self, lines, message):
+    """Merges an text representation of a protocol message into a message."""
+    self._allow_multiple_scalars = True
+    self._ParseOrMerge(lines, message)
+    return message
 
-  Raises:
-    ParseError: In case of text parsing problems.
-  """
-  message_descriptor = message.DESCRIPTOR
-  if (hasattr(message_descriptor, 'syntax') and
-      message_descriptor.syntax == 'proto3'):
-    # Proto3 doesn't represent presence so we can't test if multiple
-    # scalars have occurred.  We have to allow them.
-    allow_multiple_scalars = True
-  if tokenizer.TryConsume('['):
-    name = [tokenizer.ConsumeIdentifier()]
-    while tokenizer.TryConsume('.'):
-      name.append(tokenizer.ConsumeIdentifier())
-    name = '.'.join(name)
-
-    if not message_descriptor.is_extendable:
-      raise tokenizer.ParseErrorPreviousToken(
-          'Message type "%s" does not have extensions.' %
-          message_descriptor.full_name)
-    # pylint: disable=protected-access
-    field = message.Extensions._FindExtensionByName(name)
-    # pylint: enable=protected-access
-    if not field:
-      if allow_unknown_extension:
-        field = None
-      else:
-        raise tokenizer.ParseErrorPreviousToken(
-            'Extension "%s" not registered.' % name)
-    elif message_descriptor != field.containing_type:
-      raise tokenizer.ParseErrorPreviousToken(
-          'Extension "%s" does not extend message type "%s".' % (
-              name, message_descriptor.full_name))
+  def _ParseOrMerge(self, lines, message):
+    """Converts an text representation of a protocol message into a message.
 
-    tokenizer.Consume(']')
+    Args:
+      lines: Lines of a message's text representation.
+      message: A protocol buffer message to merge into.
 
-  else:
-    name = tokenizer.ConsumeIdentifier()
-    field = message_descriptor.fields_by_name.get(name, None)
-
-    # Group names are expected to be capitalized as they appear in the
-    # .proto file, which actually matches their type names, not their field
-    # names.
-    if not field:
-      field = message_descriptor.fields_by_name.get(name.lower(), None)
-      if field and field.type != descriptor.FieldDescriptor.TYPE_GROUP:
-        field = None
-
-    if (field and field.type == descriptor.FieldDescriptor.TYPE_GROUP and
-        field.message_type.name != name):
-      field = None
-
-    if not field:
-      raise tokenizer.ParseErrorPreviousToken(
-          'Message type "%s" has no field named "%s".' % (
-              message_descriptor.full_name, name))
-
-  if field:
-    if not allow_multiple_scalars and field.containing_oneof:
-      # Check if there's a different field set in this oneof.
-      # Note that we ignore the case if the same field was set before, and we
-      # apply allow_multiple_scalars to non-scalar fields as well.
-      which_oneof = message.WhichOneof(field.containing_oneof.name)
-      if which_oneof is not None and which_oneof != field.name:
+    Raises:
+      ParseError: On text parsing problems.
+    """
+    tokenizer = _Tokenizer(lines)
+    while not tokenizer.AtEnd():
+      self._MergeField(tokenizer, message)
+
+  def _MergeField(self, tokenizer, message):
+    """Merges a single protocol message field into a message.
+
+    Args:
+      tokenizer: A tokenizer to parse the field name and values.
+      message: A protocol message to record the data.
+
+    Raises:
+      ParseError: In case of text parsing problems.
+    """
+    message_descriptor = message.DESCRIPTOR
+    if (hasattr(message_descriptor, 'syntax') and
+        message_descriptor.syntax == 'proto3'):
+      # Proto3 doesn't represent presence so we can't test if multiple
+      # scalars have occurred.  We have to allow them.
+      self._allow_multiple_scalars = True
+    if tokenizer.TryConsume('['):
+      name = [tokenizer.ConsumeIdentifier()]
+      while tokenizer.TryConsume('.'):
+        name.append(tokenizer.ConsumeIdentifier())
+      name = '.'.join(name)
+
+      if not message_descriptor.is_extendable:
+        raise tokenizer.ParseErrorPreviousToken(
+            'Message type "%s" does not have extensions.' %
+            message_descriptor.full_name)
+      # pylint: disable=protected-access
+      field = message.Extensions._FindExtensionByName(name)
+      # pylint: enable=protected-access
+      if not field:
+        if self.allow_unknown_extension:
+          field = None
+        else:
+          raise tokenizer.ParseErrorPreviousToken(
+              'Extension "%s" not registered.' % name)
+      elif message_descriptor != field.containing_type:
         raise tokenizer.ParseErrorPreviousToken(
-            'Field "%s" is specified along with field "%s", another member of '
-            'oneof "%s" for message type "%s".' % (
-                field.name, which_oneof, field.containing_oneof.name,
-                message_descriptor.full_name))
+            'Extension "%s" does not extend message type "%s".' % (
+                name, message_descriptor.full_name))
+
+      tokenizer.Consume(']')
 
-    if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
-      tokenizer.TryConsume(':')
-      merger = _MergeMessageField
     else:
-      tokenizer.Consume(':')
-      merger = _MergeScalarField
-
-    if (field.label == descriptor.FieldDescriptor.LABEL_REPEATED
-        and tokenizer.TryConsume('[')):
-      # Short repeated format, e.g. "foo: [1, 2, 3]"
-      while True:
-        merger(tokenizer, message, field, allow_multiple_scalars,
-               allow_unknown_extension)
-        if tokenizer.TryConsume(']'): break
-        tokenizer.Consume(',')
+      name = tokenizer.ConsumeIdentifier()
+      if self.allow_field_number and name.isdigit():
+        number = ParseInteger(name, True, True)
+        field = message_descriptor.fields_by_number.get(number, None)
+        if not field and message_descriptor.is_extendable:
+          field = message.Extensions._FindExtensionByNumber(number)
+      else:
+        field = message_descriptor.fields_by_name.get(name, None)
+
+        # Group names are expected to be capitalized as they appear in the
+        # .proto file, which actually matches their type names, not their field
+        # names.
+        if not field:
+          field = message_descriptor.fields_by_name.get(name.lower(), None)
+          if field and field.type != descriptor.FieldDescriptor.TYPE_GROUP:
+            field = None
+
+        if (field and field.type == descriptor.FieldDescriptor.TYPE_GROUP and
+            field.message_type.name != name):
+          field = None
+
+      if not field:
+        raise tokenizer.ParseErrorPreviousToken(
+            'Message type "%s" has no field named "%s".' % (
+                message_descriptor.full_name, name))
+
+    if field:
+      if not self._allow_multiple_scalars and field.containing_oneof:
+        # Check if there's a different field set in this oneof.
+        # Note that we ignore the case if the same field was set before, and we
+        # apply _allow_multiple_scalars to non-scalar fields as well.
+        which_oneof = message.WhichOneof(field.containing_oneof.name)
+        if which_oneof is not None and which_oneof != field.name:
+          raise tokenizer.ParseErrorPreviousToken(
+              'Field "%s" is specified along with field "%s", another member '
+              'of oneof "%s" for message type "%s".' % (
+                  field.name, which_oneof, field.containing_oneof.name,
+                  message_descriptor.full_name))
+
+      if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
+        tokenizer.TryConsume(':')
+        merger = self._MergeMessageField
+      else:
+        tokenizer.Consume(':')
+        merger = self._MergeScalarField
+
+      if (field.label == descriptor.FieldDescriptor.LABEL_REPEATED
+          and tokenizer.TryConsume('[')):
+        # Short repeated format, e.g. "foo: [1, 2, 3]"
+        while True:
+          merger(tokenizer, message, field)
+          if tokenizer.TryConsume(']'): break
+          tokenizer.Consume(',')
+
+      else:
+        merger(tokenizer, message, field)
+
+    else:  # Proto field is unknown.
+      assert self.allow_unknown_extension
+      _SkipFieldContents(tokenizer)
+
+    # For historical reasons, fields may optionally be separated by commas or
+    # semicolons.
+    if not tokenizer.TryConsume(','):
+      tokenizer.TryConsume(';')
+
+  def _MergeMessageField(self, tokenizer, message, field):
+    """Merges a single scalar field into a message.
+
+    Args:
+      tokenizer: A tokenizer to parse the field value.
+      message: The message of which field is a member.
+      field: The descriptor of the field to be merged.
+
+    Raises:
+      ParseError: In case of text parsing problems.
+    """
+    is_map_entry = _IsMapEntry(field)
 
+    if tokenizer.TryConsume('<'):
+      end_token = '>'
     else:
-      merger(tokenizer, message, field, allow_multiple_scalars,
-             allow_unknown_extension)
+      tokenizer.Consume('{')
+      end_token = '}'
+
+    if field.label == descriptor.FieldDescriptor.LABEL_REPEATED:
+      if field.is_extension:
+        sub_message = message.Extensions[field].add()
+      elif is_map_entry:
+        # pylint: disable=protected-access
+        sub_message = field.message_type._concrete_class()
+      else:
+        sub_message = getattr(message, field.name).add()
+    else:
+      if field.is_extension:
+        sub_message = message.Extensions[field]
+      else:
+        sub_message = getattr(message, field.name)
+      sub_message.SetInParent()
+
+    while not tokenizer.TryConsume(end_token):
+      if tokenizer.AtEnd():
+        raise tokenizer.ParseErrorPreviousToken('Expected "%s".' % (end_token,))
+      self._MergeField(tokenizer, sub_message)
+
+    if is_map_entry:
+      value_cpptype = field.message_type.fields_by_name['value'].cpp_type
+      if value_cpptype == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
+        value = getattr(message, field.name)[sub_message.key]
+        value.MergeFrom(sub_message.value)
+      else:
+        getattr(message, field.name)[sub_message.key] = sub_message.value
 
-  else:  # Proto field is unknown.
-    assert allow_unknown_extension
-    _SkipFieldContents(tokenizer)
+  def _MergeScalarField(self, tokenizer, message, field):
+    """Merges a single scalar field into a message.
 
-  # For historical reasons, fields may optionally be separated by commas or
-  # semicolons.
-  if not tokenizer.TryConsume(','):
-    tokenizer.TryConsume(';')
+    Args:
+      tokenizer: A tokenizer to parse the field value.
+      message: A protocol message to record the data.
+      field: The descriptor of the field to be merged.
+
+    Raises:
+      ParseError: In case of text parsing problems.
+      RuntimeError: On runtime errors.
+    """
+    _ = self.allow_unknown_extension
+    value = None
+
+    if field.type in (descriptor.FieldDescriptor.TYPE_INT32,
+                      descriptor.FieldDescriptor.TYPE_SINT32,
+                      descriptor.FieldDescriptor.TYPE_SFIXED32):
+      value = tokenizer.ConsumeInt32()
+    elif field.type in (descriptor.FieldDescriptor.TYPE_INT64,
+                        descriptor.FieldDescriptor.TYPE_SINT64,
+                        descriptor.FieldDescriptor.TYPE_SFIXED64):
+      value = tokenizer.ConsumeInt64()
+    elif field.type in (descriptor.FieldDescriptor.TYPE_UINT32,
+                        descriptor.FieldDescriptor.TYPE_FIXED32):
+      value = tokenizer.ConsumeUint32()
+    elif field.type in (descriptor.FieldDescriptor.TYPE_UINT64,
+                        descriptor.FieldDescriptor.TYPE_FIXED64):
+      value = tokenizer.ConsumeUint64()
+    elif field.type in (descriptor.FieldDescriptor.TYPE_FLOAT,
+                        descriptor.FieldDescriptor.TYPE_DOUBLE):
+      value = tokenizer.ConsumeFloat()
+    elif field.type == descriptor.FieldDescriptor.TYPE_BOOL:
+      value = tokenizer.ConsumeBool()
+    elif field.type == descriptor.FieldDescriptor.TYPE_STRING:
+      value = tokenizer.ConsumeString()
+    elif field.type == descriptor.FieldDescriptor.TYPE_BYTES:
+      value = tokenizer.ConsumeByteString()
+    elif field.type == descriptor.FieldDescriptor.TYPE_ENUM:
+      value = tokenizer.ConsumeEnum(field)
+    else:
+      raise RuntimeError('Unknown field type %d' % field.type)
+
+    if field.label == descriptor.FieldDescriptor.LABEL_REPEATED:
+      if field.is_extension:
+        message.Extensions[field].append(value)
+      else:
+        getattr(message, field.name).append(value)
+    else:
+      if field.is_extension:
+        if not self._allow_multiple_scalars and message.HasExtension(field):
+          raise tokenizer.ParseErrorPreviousToken(
+              'Message type "%s" should not have multiple "%s" extensions.' %
+              (message.DESCRIPTOR.full_name, field.full_name))
+        else:
+          message.Extensions[field] = value
+      else:
+        if not self._allow_multiple_scalars and message.HasField(field.name):
+          raise tokenizer.ParseErrorPreviousToken(
+              'Message type "%s" should not have multiple "%s" fields.' %
+              (message.DESCRIPTOR.full_name, field.name))
+        else:
+          setattr(message, field.name, value)
 
 
 def _SkipFieldContents(tokenizer):
@@ -555,10 +745,10 @@ def _SkipFieldValue(tokenizer):
   Raises:
     ParseError: In case an invalid field value is found.
   """
-  # String tokens can come in multiple adjacent string literals.
+  # String/bytes tokens can come in multiple adjacent string literals.
   # If we can consume one, consume as many as we can.
-  if tokenizer.TryConsumeString():
-    while tokenizer.TryConsumeString():
+  if tokenizer.TryConsumeByteString():
+    while tokenizer.TryConsumeByteString():
       pass
     return
 
@@ -569,132 +759,6 @@ def _SkipFieldValue(tokenizer):
     raise ParseError('Invalid field value: ' + tokenizer.token)
 
 
-def _MergeMessageField(tokenizer, message, field, allow_multiple_scalars,
-                       allow_unknown_extension):
-  """Merges a single scalar field into a message.
-
-  Args:
-    tokenizer: A tokenizer to parse the field value.
-    message: The message of which field is a member.
-    field: The descriptor of the field to be merged.
-    allow_multiple_scalars: Determines if repeated values for a non-repeated
-      field are permitted, e.g., the string "foo: 1 foo: 2" for a
-      required/optional field named "foo".
-    allow_unknown_extension: if True, skip over missing extensions and keep
-      parsing.
-
-  Raises:
-    ParseError: In case of text parsing problems.
-  """
-  is_map_entry = _IsMapEntry(field)
-
-  if tokenizer.TryConsume('<'):
-    end_token = '>'
-  else:
-    tokenizer.Consume('{')
-    end_token = '}'
-
-  if field.label == descriptor.FieldDescriptor.LABEL_REPEATED:
-    if field.is_extension:
-      sub_message = message.Extensions[field].add()
-    elif is_map_entry:
-      # pylint: disable=protected-access
-      sub_message = field.message_type._concrete_class()
-    else:
-      sub_message = getattr(message, field.name).add()
-  else:
-    if field.is_extension:
-      sub_message = message.Extensions[field]
-    else:
-      sub_message = getattr(message, field.name)
-    sub_message.SetInParent()
-
-  while not tokenizer.TryConsume(end_token):
-    if tokenizer.AtEnd():
-      raise tokenizer.ParseErrorPreviousToken('Expected "%s".' % (end_token,))
-    _MergeField(tokenizer, sub_message, allow_multiple_scalars,
-                allow_unknown_extension)
-
-  if is_map_entry:
-    value_cpptype = field.message_type.fields_by_name['value'].cpp_type
-    if value_cpptype == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
-      value = getattr(message, field.name)[sub_message.key]
-      value.MergeFrom(sub_message.value)
-    else:
-      getattr(message, field.name)[sub_message.key] = sub_message.value
-
-
-def _MergeScalarField(tokenizer, message, field, allow_multiple_scalars,
-                      allow_unknown_extension):
-  """Merges a single scalar field into a message.
-
-  Args:
-    tokenizer: A tokenizer to parse the field value.
-    message: A protocol message to record the data.
-    field: The descriptor of the field to be merged.
-    allow_multiple_scalars: Determines if repeated values for a non-repeated
-      field are permitted, e.g., the string "foo: 1 foo: 2" for a
-      required/optional field named "foo".
-    allow_unknown_extension: Unused, just here for consistency with
-      _MergeMessageField.
-
-  Raises:
-    ParseError: In case of text parsing problems.
-    RuntimeError: On runtime errors.
-  """
-  _ = allow_unknown_extension
-  value = None
-
-  if field.type in (descriptor.FieldDescriptor.TYPE_INT32,
-                    descriptor.FieldDescriptor.TYPE_SINT32,
-                    descriptor.FieldDescriptor.TYPE_SFIXED32):
-    value = tokenizer.ConsumeInt32()
-  elif field.type in (descriptor.FieldDescriptor.TYPE_INT64,
-                      descriptor.FieldDescriptor.TYPE_SINT64,
-                      descriptor.FieldDescriptor.TYPE_SFIXED64):
-    value = tokenizer.ConsumeInt64()
-  elif field.type in (descriptor.FieldDescriptor.TYPE_UINT32,
-                      descriptor.FieldDescriptor.TYPE_FIXED32):
-    value = tokenizer.ConsumeUint32()
-  elif field.type in (descriptor.FieldDescriptor.TYPE_UINT64,
-                      descriptor.FieldDescriptor.TYPE_FIXED64):
-    value = tokenizer.ConsumeUint64()
-  elif field.type in (descriptor.FieldDescriptor.TYPE_FLOAT,
-                      descriptor.FieldDescriptor.TYPE_DOUBLE):
-    value = tokenizer.ConsumeFloat()
-  elif field.type == descriptor.FieldDescriptor.TYPE_BOOL:
-    value = tokenizer.ConsumeBool()
-  elif field.type == descriptor.FieldDescriptor.TYPE_STRING:
-    value = tokenizer.ConsumeString()
-  elif field.type == descriptor.FieldDescriptor.TYPE_BYTES:
-    value = tokenizer.ConsumeByteString()
-  elif field.type == descriptor.FieldDescriptor.TYPE_ENUM:
-    value = tokenizer.ConsumeEnum(field)
-  else:
-    raise RuntimeError('Unknown field type %d' % field.type)
-
-  if field.label == descriptor.FieldDescriptor.LABEL_REPEATED:
-    if field.is_extension:
-      message.Extensions[field].append(value)
-    else:
-      getattr(message, field.name).append(value)
-  else:
-    if field.is_extension:
-      if not allow_multiple_scalars and message.HasExtension(field):
-        raise tokenizer.ParseErrorPreviousToken(
-            'Message type "%s" should not have multiple "%s" extensions.' %
-            (message.DESCRIPTOR.full_name, field.full_name))
-      else:
-        message.Extensions[field] = value
-    else:
-      if not allow_multiple_scalars and message.HasField(field.name):
-        raise tokenizer.ParseErrorPreviousToken(
-            'Message type "%s" should not have multiple "%s" fields.' %
-            (message.DESCRIPTOR.full_name, field.name))
-      else:
-        setattr(message, field.name, value)
-
-
 class _Tokenizer(object):
   """Protocol buffer text representation tokenizer.
 
@@ -925,9 +989,9 @@ class _Tokenizer(object):
     self.NextToken()
     return result
 
-  def TryConsumeString(self):
+  def TryConsumeByteString(self):
     try:
-      self.ConsumeString()
+      self.ConsumeByteString()
       return True
     except ParseError:
       return False