Submit recent changes from internal branch, including "lite mode" for

C++ and Java.  See CHANGES.txt for more details.

1 files changed, 278 insertions, 3 deletions

diff --git a/python/google/protobuf/internal/text_format_test.py b/python/google/protobuf/internal/text_format_test.py
index 871590e7..0cf27186 100755
--- a/python/google/protobuf/internal/text_format_test.py
+++ b/python/google/protobuf/internal/text_format_test.py
@@ -42,11 +42,16 @@ from google.protobuf.internal import test_util
 from google.protobuf import unittest_pb2
 from google.protobuf import unittest_mset_pb2
 
-class TextFormatTest(unittest.TestCase):
-  def CompareToGoldenFile(self, text, golden_filename):
+
+class TextFormatTest(test_util.GoldenMessageTestCase):
+  def ReadGolden(self, golden_filename):
     f = test_util.GoldenFile(golden_filename)
     golden_lines = f.readlines()
     f.close()
+    return golden_lines
+
+  def CompareToGoldenFile(self, text, golden_filename):
+    golden_lines = self.ReadGolden(golden_filename)
     self.CompareToGoldenLines(text, golden_lines)
 
   def CompareToGoldenText(self, text, golden_text):
@@ -117,6 +122,276 @@ class TextFormatTest(unittest.TestCase):
     return text.replace('e+0','e+').replace('e+0','e+') \
                .replace('e-0','e-').replace('e-0','e-')
 
+  def testMergeGolden(self):
+    golden_text = '\n'.join(self.ReadGolden('text_format_unittest_data.txt'))
+    parsed_message = unittest_pb2.TestAllTypes()
+    text_format.Merge(golden_text, parsed_message)
+
+    message = unittest_pb2.TestAllTypes()
+    test_util.SetAllFields(message)
+    self.assertEquals(message, parsed_message)
+
+  def testMergeGoldenExtensions(self):
+    golden_text = '\n'.join(self.ReadGolden(
+        'text_format_unittest_extensions_data.txt'))
+    parsed_message = unittest_pb2.TestAllExtensions()
+    text_format.Merge(golden_text, parsed_message)
+
+    message = unittest_pb2.TestAllExtensions()
+    test_util.SetAllExtensions(message)
+    self.assertEquals(message, parsed_message)
+
+  def testMergeAllFields(self):
+    message = unittest_pb2.TestAllTypes()
+    test_util.SetAllFields(message)
+    ascii_text = text_format.MessageToString(message)
+
+    parsed_message = unittest_pb2.TestAllTypes()
+    text_format.Merge(ascii_text, parsed_message)
+    self.assertEqual(message, parsed_message)
+    self.ExpectAllFieldsSet(message)
+
+  def testMergeAllExtensions(self):
+    message = unittest_pb2.TestAllExtensions()
+    test_util.SetAllExtensions(message)
+    ascii_text = text_format.MessageToString(message)
+
+    parsed_message = unittest_pb2.TestAllExtensions()
+    text_format.Merge(ascii_text, parsed_message)
+    self.assertEqual(message, parsed_message)
+
+  def testMergeMessageSet(self):
+    message = unittest_pb2.TestAllTypes()
+    text = ('repeated_uint64: 1\n'
+            'repeated_uint64: 2\n')
+    text_format.Merge(text, message)
+    self.assertEqual(1, message.repeated_uint64[0])
+    self.assertEqual(2, message.repeated_uint64[1])
+
+    message = unittest_mset_pb2.TestMessageSetContainer()
+    text = ('message_set {\n'
+            '  [protobuf_unittest.TestMessageSetExtension1] {\n'
+            '    i: 23\n'
+            '  }\n'
+            '  [protobuf_unittest.TestMessageSetExtension2] {\n'
+            '    str: \"foo\"\n'
+            '  }\n'
+            '}\n')
+    text_format.Merge(text, message)
+    ext1 = unittest_mset_pb2.TestMessageSetExtension1.message_set_extension
+    ext2 = unittest_mset_pb2.TestMessageSetExtension2.message_set_extension
+    self.assertEquals(23, message.message_set.Extensions[ext1].i)
+    self.assertEquals('foo', message.message_set.Extensions[ext2].str)
+
+  def testMergeExotic(self):
+    message = unittest_pb2.TestAllTypes()
+    text = ('repeated_int64: -9223372036854775808\n'
+            'repeated_uint64: 18446744073709551615\n'
+            'repeated_double: 123.456\n'
+            'repeated_double: 1.23e+22\n'
+            'repeated_double: 1.23e-18\n'
+            'repeated_string: \n'
+            '\"\\000\\001\\007\\010\\014\\n\\r\\t\\013\\\\\\\'\\\"\"\n')
+    text_format.Merge(text, message)
+
+    self.assertEqual(-9223372036854775808, message.repeated_int64[0])
+    self.assertEqual(18446744073709551615, message.repeated_uint64[0])
+    self.assertEqual(123.456, message.repeated_double[0])
+    self.assertEqual(1.23e22, message.repeated_double[1])
+    self.assertEqual(1.23e-18, message.repeated_double[2])
+    self.assertEqual(
+        '\000\001\a\b\f\n\r\t\v\\\'\"', message.repeated_string[0])
+
+  def testMergeUnknownField(self):
+    message = unittest_pb2.TestAllTypes()
+    text = 'unknown_field: 8\n'
+    self.assertRaisesWithMessage(
+        text_format.ParseError,
+        ('1:1 : Message type "protobuf_unittest.TestAllTypes" has no field named '
+         '"unknown_field".'),
+        text_format.Merge, text, message)
+
+  def testMergeBadExtension(self):
+    message = unittest_pb2.TestAllTypes()
+    text = '[unknown_extension]: 8\n'
+    self.assertRaisesWithMessage(
+        text_format.ParseError,
+        '1:2 : Extension "unknown_extension" not registered.',
+        text_format.Merge, text, message)
+
+  def testMergeGroupNotClosed(self):
+    message = unittest_pb2.TestAllTypes()
+    text = 'RepeatedGroup: <'
+    self.assertRaisesWithMessage(
+        text_format.ParseError, '1:16 : Expected ">".',
+        text_format.Merge, text, message)
+
+    text = 'RepeatedGroup: {'
+    self.assertRaisesWithMessage(
+        text_format.ParseError, '1:16 : Expected "}".',
+        text_format.Merge, text, message)
+
+  def testMergeBadEnumValue(self):
+    message = unittest_pb2.TestAllTypes()
+    text = 'optional_nested_enum: BARR'
+    self.assertRaisesWithMessage(
+        text_format.ParseError,
+        ('1:23 : Enum type "protobuf_unittest.TestAllTypes.NestedEnum" '
+         'has no value named BARR.'),
+        text_format.Merge, text, message)
+
+    message = unittest_pb2.TestAllTypes()
+    text = 'optional_nested_enum: 100'
+    self.assertRaisesWithMessage(
+        text_format.ParseError,
+        ('1:23 : Enum type "protobuf_unittest.TestAllTypes.NestedEnum" '
+         'has no value with number 100.'),
+        text_format.Merge, text, message)
+
+  def assertRaisesWithMessage(self, e_class, e, func, *args, **kwargs):
+    """Same as assertRaises, but also compares the exception message."""
+    if hasattr(e_class, '__name__'):
+      exc_name = e_class.__name__
+    else:
+      exc_name = str(e_class)
+
+    try:
+      func(*args, **kwargs)
+    except e_class, expr:
+      if str(expr) != e:
+        msg = '%s raised, but with wrong message: "%s" instead of "%s"'
+        raise self.failureException(msg % (exc_name,
+                                           str(expr).encode('string_escape'),
+                                           e.encode('string_escape')))
+      return
+    else:
+      raise self.failureException('%s not raised' % exc_name)
+
+
+class TokenizerTest(unittest.TestCase):
+
+  def testSimpleTokenCases(self):
+    text = ('identifier1:"string1"\n     \n\n'
+            'identifier2 : \n \n123  \n  identifier3 :\'string\'\n'
+            'identifiER_4 : 1.1e+2 ID5:-0.23 ID6:\'aaaa\\\'bbbb\'\n'
+            'ID7 : "aa\\"bb"\n\n\n\n ID8: {A:inf B:-inf C:true D:false}\n'
+            'ID9: 22 ID10: -111111111111111111 ID11: -22\n'
+            'ID12: 2222222222222222222')
+    tokenizer = text_format._Tokenizer(text)
+    methods = [(tokenizer.ConsumeIdentifier, 'identifier1'),
+               ':',
+               (tokenizer.ConsumeString, 'string1'),
+               (tokenizer.ConsumeIdentifier, 'identifier2'),
+               ':',
+               (tokenizer.ConsumeInt32, 123),
+               (tokenizer.ConsumeIdentifier, 'identifier3'),
+               ':',
+               (tokenizer.ConsumeString, 'string'),
+               (tokenizer.ConsumeIdentifier, 'identifiER_4'),
+               ':',
+               (tokenizer.ConsumeFloat, 1.1e+2),
+               (tokenizer.ConsumeIdentifier, 'ID5'),
+               ':',
+               (tokenizer.ConsumeFloat, -0.23),
+               (tokenizer.ConsumeIdentifier, 'ID6'),
+               ':',
+               (tokenizer.ConsumeString, 'aaaa\'bbbb'),
+               (tokenizer.ConsumeIdentifier, 'ID7'),
+               ':',
+               (tokenizer.ConsumeString, 'aa\"bb'),
+               (tokenizer.ConsumeIdentifier, 'ID8'),
+               ':',
+               '{',
+               (tokenizer.ConsumeIdentifier, 'A'),
+               ':',
+               (tokenizer.ConsumeFloat, float('inf')),
+               (tokenizer.ConsumeIdentifier, 'B'),
+               ':',
+               (tokenizer.ConsumeFloat, float('-inf')),
+               (tokenizer.ConsumeIdentifier, 'C'),
+               ':',
+               (tokenizer.ConsumeBool, True),
+               (tokenizer.ConsumeIdentifier, 'D'),
+               ':',
+               (tokenizer.ConsumeBool, False),
+               '}',
+               (tokenizer.ConsumeIdentifier, 'ID9'),
+               ':',
+               (tokenizer.ConsumeUint32, 22),
+               (tokenizer.ConsumeIdentifier, 'ID10'),
+               ':',
+               (tokenizer.ConsumeInt64, -111111111111111111),
+               (tokenizer.ConsumeIdentifier, 'ID11'),
+               ':',
+               (tokenizer.ConsumeInt32, -22),
+               (tokenizer.ConsumeIdentifier, 'ID12'),
+               ':',
+               (tokenizer.ConsumeUint64, 2222222222222222222)]
+
+    i = 0
+    while not tokenizer.AtEnd():
+      m = methods[i]
+      if type(m) == str:
+        token = tokenizer.token
+        self.assertEqual(token, m)
+        tokenizer.NextToken()
+      else:
+        self.assertEqual(m[1], m[0]())
+      i += 1
+
+  def testConsumeIntegers(self):
+    # This test only tests the failures in the integer parsing methods as well
+    # as the '0' special cases.
+    int64_max = (1 << 63) - 1
+    uint32_max = (1 << 32) - 1
+    text = '-1 %d %d' % (uint32_max + 1, int64_max + 1)
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeUint32)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeUint64)
+    self.assertEqual(-1, tokenizer.ConsumeInt32())
+
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeUint32)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeInt32)
+    self.assertEqual(uint32_max + 1, tokenizer.ConsumeInt64())
+
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeInt64)
+    self.assertEqual(int64_max + 1, tokenizer.ConsumeUint64())
+    self.assertTrue(tokenizer.AtEnd())
+
+    text = '-0 -0 0 0'
+    tokenizer = text_format._Tokenizer(text)
+    self.assertEqual(0, tokenizer.ConsumeUint32())
+    self.assertEqual(0, tokenizer.ConsumeUint64())
+    self.assertEqual(0, tokenizer.ConsumeUint32())
+    self.assertEqual(0, tokenizer.ConsumeUint64())
+    self.assertTrue(tokenizer.AtEnd())
+
+  def testConsumeByteString(self):
+    text = '"string1\''
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeByteString)
+
+    text = 'string1"'
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeByteString)
+
+    text = '\n"\\xt"'
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeByteString)
+
+    text = '\n"\\"'
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeByteString)
+
+    text = '\n"\\x"'
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeByteString)
+
+  def testConsumeBool(self):
+    text = 'not-a-bool'
+    tokenizer = text_format._Tokenizer(text)
+    self.assertRaises(text_format.ParseError, tokenizer.ConsumeBool)
+
+
 if __name__ == '__main__':
   unittest.main()
-