def PrintFieldValue(field, value, out, indent=0,
as_utf8=False, as_one_line=False):
"""Print a single field value (not including name). For repeated fields,
the value should be a single element."""
if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
if as_one_line:
out.write(b(' { '))
PrintMessage(value, out, indent, as_utf8, as_one_line)
out.write(b('}'))
else:
out.write(b(' {\n'))
PrintMessage(value, out, indent + 2, as_utf8, as_one_line)
out.write(b(' ') * indent + b('}'))
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(string_to_bytes(str(enum_value.name)))
else:
out.write(string_to_bytes(str(value)))
elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_STRING:
out.write(b('\"'))
if type(value) is unicode:
out.write(_CEscape(value.encode('utf-8'), as_utf8))
else:
out.write(_CEscape(value, as_utf8))
out.write(b('\"'))
elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_BOOL:
if value:
out.write(b("true"))
else:
out.write(b("false"))
else:
out.write(string_to_bytes(str(value)))
def PrintFieldValue(field, value, out, indent=0,
as_utf8=False, as_one_line=False):
"""Print a single field value (not including name). For repeated fields,
the value should be a single element."""
if field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_MESSAGE:
if as_one_line:
out.write(b' { ')
PrintMessage(value, out, indent, as_utf8, as_one_line)
out.write(b'}')
else:
out.write(b' {\n')
PrintMessage(value, out, indent + 2, as_utf8, as_one_line)
out.write(b' ' * indent + b'}')
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(string_to_bytes(str(enum_value.name)))
else:
out.write(string_to_bytes(str(value)))
elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_STRING:
out.write(b'\"')
if type(value) is unicode:
out.write(_CEscape(value.encode('utf-8'), as_utf8))
else:
out.write(_CEscape(value, as_utf8))
out.write(b'\"')
elif field.cpp_type == descriptor.FieldDescriptor.CPPTYPE_BOOL:
if value:
out.write(b"true")
else:
out.write(b"false")
else:
out.write(string_to_bytes(str(value)))
def _CUnescape(text):
def ReplaceHex(m):
return m.group(0)
# This is required because the 'string_escape' encoding doesn't
# allow single-digit hex escapes (like '\xf').
result = _CUNESCAPE_HEX.sub(ReplaceHex, text)
return string_to_bytes(loc_string_escape(result))
def _CUnescape(text):
def ReplaceHex(m):
return m.group(0)
# This is required because the 'string_escape' encoding doesn't
# allow single-digit hex escapes (like '\xf').
result = _CUNESCAPE_HEX.sub(ReplaceHex, text)
return string_to_bytes(loc_string_escape(result))
def _CUnescape(text):
def ReplaceHex(m):
# Only replace the match if the number of leading back slashes is odd. i.e.
# the slash itself is not escaped.
if len(m.group(1)) & 1:
return m.group(1) + b('x0') + m.group(2)
return m.group(0)
# This is required because the 'string_escape' encoding doesn't
# allow single-digit hex escapes (like '\xf').
result = _CUNESCAPE_HEX.sub(ReplaceHex, text)
return string_to_bytes(loc_string_escape(result))
def _CUnescape(text):
def ReplaceHex(m):
# Only replace the match if the number of leading back slashes is odd. i.e.
# the slash itself is not escaped.
if len(m.group(1)) & 1:
return m.group(1) + b('x0') + m.group(2)
return m.group(0)
# This is required because the 'string_escape' encoding doesn't
# allow single-digit hex escapes (like '\xf').
result = _CUNESCAPE_HEX.sub(ReplaceHex, text)
return string_to_bytes(loc_string_escape(result))
def escape(c):
b = ord(c)
if b == 10: return b"\\n" # optional escape
if b == 13: return b"\\r" # optional escape
if b == 9: return b"\\t" # optional escape
if b == 39: return b"\\'" # optional escape
if b == 34: return b'\\"' # necessary escape
if b == 92: return b"\\\\" # necessary escape
# necessary escapes
if not as_utf8 and (b >= 127 or b < 32): return string_to_bytes("\\%03o" % b)
return c
def escape(c):
o = bytes_as_num(c)
if o == 10: return b(r"\n") # optional escape
if o == 13: return b(r"\r") # optional escape
if o == 9: return b(r"\t") # optional escape
if o == 39: return b(r"\'") # optional escape
if o == 34: return b(r'\"') # necessary escape
if o == 92: return b(r"\\") # necessary escape
# necessary escapes
if not as_utf8 and (o >= 127 or o < 32): return string_to_bytes("\\%03o" % o)
return num_as_byte(c)
def escape(c):
o = bytes_as_num(c)
if o == 10: return rb"\n" # optional escape
if o == 13: return rb"\r" # optional escape
if o == 9: return rb"\t" # optional escape
if o == 39: return rb"\'" # optional escape
if o == 34: return rb'\"' # necessary escape
if o == 92: return rb"\\" # necessary escape
# necessary escapes
if not as_utf8 and (o >= 127 or o < 32): return string_to_bytes("\\%03o" % o)
return num_as_byte(c)
def escape(c):
b = ord(c)
if b == 10: return b"\\n" # optional escape
if b == 13: return b"\\r" # optional escape
if b == 9: return b"\\t" # optional escape
if b == 39: return b"\\'" # optional escape
if b == 34: return b'\\"' # necessary escape
if b == 92: return b"\\\\" # necessary escape
# necessary escapes
if not as_utf8 and (b >= 127 or b < 32):
return string_to_bytes("\\%03o" % b)
return c
def testMergeStringFieldUnescape(self):
message = unittest_pb2.TestAllTypes()
text = r'''repeated_string: "\xf\x62"
repeated_string: "\\xf\\x62"
repeated_string: "\\\xf\\\x62"
repeated_string: "\\\\xf\\\\x62"
repeated_string: "\\\\\xf\\\\\x62"
repeated_string: "\x5cx20"'''
text_format.Merge(string_to_bytes(text), message)
SLASH = '\\'
self.assertEqual('\x0fb', message.repeated_string[0])
self.assertEqual(SLASH + 'xf' + SLASH + 'x62',
message.repeated_string[1])
self.assertEqual(SLASH + '\x0f' + SLASH + 'b',
message.repeated_string[2])
self.assertEqual(SLASH + SLASH + 'xf' + SLASH + SLASH + 'x62',
message.repeated_string[3])
self.assertEqual(SLASH + SLASH + '\x0f' + SLASH + SLASH + 'b',
message.repeated_string[4])
self.assertEqual(SLASH + 'x20', message.repeated_string[5])
def StringByteSize(field_number, string): return BytesByteSize(field_number, string_to_bytes(string))
