def _ConsumeSingleByteString(self):
"""Consume one token of a string literal.
String literals (whether bytes or text) can come in multiple adjacent
tokens which are automatically concatenated, like in C or Python. This
method only consumes one token.
Returns:
The token parsed.
Raises:
ParseError: When the wrong format data is found.
"""
text = self.token
if len(text) < 1 or text[0] not in _QUOTES:
raise self.ParseError('Expected string but found: %r' % (text,))
if len(text) < 2 or text[-1] != text[0]:
raise self.ParseError('String missing ending quote: %r' % (text,))
try:
result = text_encoding.CUnescape(text[1:-1])
except ValueError as e:
raise self.ParseError(str(e))
self.NextToken()
return result
def _SetFieldType(self, field_proto, field_desc, package, scope):
"""Sets the field's type, cpp_type, message_type and enum_type.
Args:
field_proto: Data about the field in proto format.
field_desc: The descriptor to modiy.
package: The package the field's container is in.
scope: Enclosing scope of available types.
"""
if field_proto.type_name:
desc = self._GetTypeFromScope(package, field_proto.type_name,
scope)
else:
desc = None
if not field_proto.HasField('type'):
if isinstance(desc, descriptor.Descriptor):
field_proto.type = descriptor.FieldDescriptor.TYPE_MESSAGE
else:
field_proto.type = descriptor.FieldDescriptor.TYPE_ENUM
field_desc.cpp_type = descriptor.FieldDescriptor.ProtoTypeToCppProtoType(
field_proto.type)
if (field_proto.type == descriptor.FieldDescriptor.TYPE_MESSAGE
or field_proto.type == descriptor.FieldDescriptor.TYPE_GROUP):
field_desc.message_type = desc
if field_proto.type == descriptor.FieldDescriptor.TYPE_ENUM:
field_desc.enum_type = desc
if field_proto.label == descriptor.FieldDescriptor.LABEL_REPEATED:
field_desc.has_default_value = False
field_desc.default_value = []
elif field_proto.HasField('default_value'):
field_desc.has_default_value = True
if (field_proto.type == descriptor.FieldDescriptor.TYPE_DOUBLE or
field_proto.type == descriptor.FieldDescriptor.TYPE_FLOAT):
field_desc.default_value = float(field_proto.default_value)
elif field_proto.type == descriptor.FieldDescriptor.TYPE_STRING:
field_desc.default_value = field_proto.default_value
elif field_proto.type == descriptor.FieldDescriptor.TYPE_BOOL:
field_desc.default_value = field_proto.default_value.lower(
) == 'true'
elif field_proto.type == descriptor.FieldDescriptor.TYPE_ENUM:
field_desc.default_value = field_desc.enum_type.values_by_name[
field_proto.default_value].number
elif field_proto.type == descriptor.FieldDescriptor.TYPE_BYTES:
field_desc.default_value = text_encoding.CUnescape(
field_proto.default_value)
else:
field_desc.default_value = int(field_proto.default_value)
else:
field_desc.has_default_value = False
field_desc.default_value = None
field_desc.type = field_proto.type
def _ConsumeSingleByteString(self):
"""Consume one token of a string literal.
String literals (whether bytes or text) can come in multiple adjacent
tokens which are automatically concatenated, like in C or Python. This
method only consumes one token.
"""
text = self.token
if len(text) < 1 or text[0] not in ('\'', '"'):
raise self._ParseError('Expected string.')
if len(text) < 2 or text[-1] != text[0]:
raise self._ParseError('String missing ending quote.')
try:
result = text_encoding.CUnescape(text[1:-1])
except ValueError, e:
raise self._ParseError(str(e))
def testCUnescape(self):
for escaped, escaped_utf8, unescaped in TEST_VALUES:
self.assertEqual(unescaped, text_encoding.CUnescape(escaped))
self.assertEqual(unescaped, text_encoding.CUnescape(escaped_utf8))