How do you tell pylint what the members of a protobuf-generated object are?

Question

I would like to release a python package for a set of protobuf messages. The protobuf compiler (protoc) generates a python library that does not actually define types/classes the typical sense, but rather dynamically constructs them. Is there any way to hint to pylint what the members and fields of these classes are?

For example, consider the following simple protobuf message specification:

message Person {
  required string name = 1;
  required int32 id = 2;
  optional string email = 3;
}

The compiler generates the following long pile of code:

# Generated by the protocol buffer compiler.  DO NOT EDIT!
# source: test.proto

import sys
_b=sys.version_info[0]<3 and (lambda x:x) or (lambda x:x.encode('latin1'))
from google.protobuf import descriptor as _descriptor
from google.protobuf import message as _message
from google.protobuf import reflection as _reflection
from google.protobuf import symbol_database as _symbol_database
from google.protobuf import descriptor_pb2
# @@protoc_insertion_point(imports)

_sym_db = _symbol_database.Default()




DESCRIPTOR = _descriptor.FileDescriptor(
  name='test.proto',
  package='',
  serialized_pb=_b('\n\ntest.proto\"1\n\x06Person\x12\x0c\n\x04name\x18\x01 \x02(\t\x12\n\n\x02id\x18\x02 \x02(\x05\x12\r\n\x05\x65mail\x18\x03 \x01(\t')
)
_sym_db.RegisterFileDescriptor(DESCRIPTOR)




_PERSON = _descriptor.Descriptor(
  name='Person',
  full_name='Person',
  filename=None,
  file=DESCRIPTOR,
  containing_type=None,
  fields=[
    _descriptor.FieldDescriptor(
      name='name', full_name='Person.name', index=0,
      number=1, type=9, cpp_type=9, label=2,
      has_default_value=False, default_value=_b("").decode('utf-8'),
      message_type=None, enum_type=None, containing_type=None,
      is_extension=False, extension_scope=None,
      options=None),
    _descriptor.FieldDescriptor(
      name='id', full_name='Person.id', index=1,
      number=2, type=5, cpp_type=1, label=2,
      has_default_value=False, default_value=0,
      message_type=None, enum_type=None, containing_type=None,
      is_extension=False, extension_scope=None,
      options=None),
    _descriptor.FieldDescriptor(
      name='email', full_name='Person.email', index=2,
      number=3, type=9, cpp_type=9, label=1,
      has_default_value=False, default_value=_b("").decode('utf-8'),
      message_type=None, enum_type=None, containing_type=None,
      is_extension=False, extension_scope=None,
      options=None),
  ],
  extensions=[
  ],
  nested_types=[],
  enum_types=[
  ],
  options=None,
  is_extendable=False,
  extension_ranges=[],
  oneofs=[
  ],
  serialized_start=14,
  serialized_end=63,
)

DESCRIPTOR.message_types_by_name['Person'] = _PERSON

Person = _reflection.GeneratedProtocolMessageType('Person', (_message.Message,), dict(
  DESCRIPTOR = _PERSON,
  __module__ = 'test_pb2'
  # @@protoc_insertion_point(class_scope:Person)
  ))
_sym_db.RegisterMessage(Person)


# @@protoc_insertion_point(module_scope)

This is an excellent question. In moving an org from using raw dictionaries to protobuf objects I was hoping we'd get some static analysis wins by using pylint to pick up on typos in member names. Would be very interested if someone's managed to get this to work with the generated classes. — aeijdenberg, May 06 '17 at 03:46
Very interesting question! The correct names are already in the code, so it seems like one could solve this by writing an [extension](https://github.com/PyCQA/pylint/tree/master/pylint/extensions) for pylint. — iFreilicht, Nov 20 '17 at 17:50

score 9 · Accepted Answer · answered Jul 17 '18 at 12:40

There didn't seem to be any support for automatically hinting to Pylint the names of the fields on protobuf messages, so I cobbled together a small extension to Pylint to statically determine these.

Introducing pylint-protobuf:

from example_pb2 import Person
p = Person()
p.invalid_field = 123

Usage:

$ pip install pylint-protobuf
$ pylint --load-plugins=pylint_protobuf example.py
************* Module example
E:  3, 0: Field 'invalid_field' does not appear in the declared
fields of protobuf-generated class 'Person' and will raise
AttributeError on access (protobuf-undefined-attribute)

Fair warning: it's pretty alpha, features not supported (yet!) include nested imports and repeated fields.

How do you tell pylint what the members of a protobuf-generated object are?

1 Answers1