`var-annotated` behaves differently with daemon

[DetachHead]

Bug Report

class Foo:
    bar = None

when running with dmypy:

$ dmypy run -- foo.py
Restarting: configuration changed
Daemon stopped
Daemon started
foo.py: note: In class "Foo":
foo.py:2: error: Need type annotation for "bar"  [var-annotated]
        bar = None
        ^
Found 1 error in 1 file (checked 1 source file)

when running with mypy:

$ mypy foo.py
Success: no issues found in 1 source file

Your Environment

• Mypy version used: 0.910
• Mypy command-line flags: none
• Mypy configuration options from mypy.ini (and other config files):

  [mypy]
  exclude = venv
  mypy_path = stubs
  warn_redundant_casts = True
  show_error_context = True
  pretty = True
  follow_imports = skip
  allow_redefinition = True
  show_error_codes = True

• Python version used: 3.9.6
• Operating system and version: windows 10 1909

[KotlinIsland]

Isn't this just because the daemon implicitly enables local_partial_types?
https://mypy.readthedocs.io/en/stable/config_file.html#confval-local_partial_types

[DetachHead]

seems silly that the daemon secretly changes random flags, i would expect it to behave the same way

[hauntsaninja]

Yeah, this is because of local partial types. I think Jukka is planning on making that the default, so closing as duplicae of #10169

[KotlinIsland]

It's the default in basedmypy

[vergenzt]

Found this issue from googling, and sharing my further digging:

• it's turned on at

  mypy/mypy/dmypy_server.py

  Lines 197 to 199 in 55d4c17

  	# Fine-grained incremental doesn't support general partial types
  	# (details in https://github.com/python/mypy/issues/4492)
  	options.local_partial_types = True

• reasoning for the flag being turned on in daemon mode is because otherwise the daemon is buggy: see What to do about inferring types from multiple scopes? #4492
• in future, this error is going to be default for CLI invocations too: Enable local partial types by default #10169 (but it's not yet)

[KotlinIsland]

in future, this error is going to be default for CLI invocations too

it's the default in basedmypy