Nested dataclass definitions not working with pydantic

[and3rson]

Initial Checks

• I have searched GitHub for a duplicate issue and I'm sure this is something new
• I have searched Google & StackOverflow for a solution and couldn't find anything
• I have read and followed the docs and still think this is a bug
• I am confident that the issue is with pydantic (not my code, or another library in the ecosystem like FastAPI or mypy)

Description

When defining dataclasses inside dataclasses, and then using those top-level dataclasses in pydantic models, I'm sometimes getting a TypeError: non-default argument '...' follows default argument.

Removing definitions of child dataclasses makes the code work properly.

Example Code

from typing import Optional, List
from dataclasses import dataclass, field

from pydantic import BaseModel


@dataclass
class Foo:
    a: Optional[int] = field(default=42)
    b: List = field(default_factory=list)

    @dataclass
    class Bar:
        pass


@dataclass
class Model1:
    foo: Foo
print(Model1(foo=Foo()))  # Model(foo=Foo(a=42, b=[]))


class Model2(BaseModel):
    foo: Foo
print(Model2(foo=Foo()))  # TypeError: non-default argument 'b' follows default argument
# Removing the definition of `Foo.Bar` makes this work properly.

Python, Pydantic & OS Version

pydantic version: 1.10.2
            pydantic compiled: True
                 install path: /home/anderson/src/***/.venv/lib/python3.9/site-packages/pydantic
               python version: 3.9.13 (main, Oct  5 2022, 17:32:54)  [GCC 12.2.0]
                     platform: Linux-6.0.9-arch1-1-x86_64-with-glibc2.36
     optional deps. installed: ['typing-extensions']

Affected Components

• Compatibility between releases
• Data validation/parsing
• Data serialization - .dict() and .json()
• JSON Schema
• Dataclasses
• Model Config
• Field Types - adding or changing a particular data type
• Function validation decorator
• Generic Models
• Other Model behaviour - construct(), pickling, private attributes, ORM mode
• Plugins and integration with other tools - mypy, FastAPI, python-devtools, Hypothesis, VS Code, PyCharm, etc.

[and3rson]

Possibly related: #2424

[and3rson]

Update: this is solved by using pydantic.dataclasses.dataclass instead. Still, the error is very confusing.

[PrettyWood]

@and3rson This will be fixed in 1.10.3

[samuelcolvin]

Closed via #4878