Dataclass post_init_post_parse called multiple times.

[Danosaurus-rex]

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 using inheritance with pydantic.dataclasses.dataclass based objects with pydantic>=1.10 I seem to have encountered a bug that calls __post_init_post_parse__ methods multiple times if a __post_init__ method is called further down an inheritance chain.

Running the below code snippet outputs the print statement three times, whereas if __post_init__ is placed in AbstractClass the "post init post parse called" statement is called twice.

The exact same code snippet on pydantic<1.10 works as expected with the print statement only printed the once.

Including an overridden __post_init__ method in ConcreteClass only prints the statement once - but of course if there is more useful logic contained in BaseClass.__post_init__ (as would be the case for this pattern) then this is ignored.

Removing the @dataclass decorators from the base classes works fine however, except if we didn't have access to the code for those underlying base classes.

EDIT: It appears its due to these lines:
https://github.com/pydantic/pydantic/blob/v1.10.2/pydantic/dataclasses.py#L290-L304

Example Code

from pydantic.dataclasses import dataclass

@dataclass
class BaseClass:
    def __post_init__(self):
        pass

@dataclass
class AbstractClass(BaseClass):
    pass

@dataclass
class ConcreteClass(AbstractClass):
    def __post_init_post_parse__(self):
        print("post init post parse called")

ConcreteClass()

# prints:
# post init post parse called
# post init post parse called
# post init post parse called


# Example of nested dataclasses from docs: https://pydantic-docs.helpmanual.io/usage/dataclasses/#convert-stdlib-dataclasses-into-pydantic-dataclasses

from datetime import datetime
from typing import Optional

@dataclass
class Meta:
    modified_date: Optional[datetime]
    seen_count: int
    def __post_init__(self):
        print("POST INIT")

@dataclass
class File(Meta):
    filename: str
    def __post_init_post_parse__(self):
        print("POST INIT POST PARSE")

File(filename="file.txt", modified_date=datetime.now(), seen_count=1)

# prints:
# POST INIT
# POST INIT POST PARSE
# POST INIT POST PARSE

Python, Pydantic & OS Version

pydantic version: 1.10.2
            pydantic compiled: True
                 install path: *****
               python version: 3.8.13 (default, Mar 28 2022, 06:16:26)  [Clang 12.0.0 ]
                     platform: macOS-10.16-x86_64-i386-64bit
     optional deps. installed: ['dotenv', '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
• Settings Management
• Plugins and integration with other tools - mypy, FastAPI, python-devtools, Hypothesis, VS Code, PyCharm, etc.

[samuelcolvin]

See #4484.

@PrettyWood, any ideas?

[PrettyWood]

I'll check

[Kludex]

• This was solved on fix: avoid multiple calls of __post_init__ when dataclasses are inherited #4493. 🙏