I'd like to start with something like: "The __init__ function generated by @dataclass does not call base class __init__ methods. If base class __init__ methods need to be called, it's common to do so in a __post_init__ method."

Thanks for the feedback! I have adapted this text slightly in my last commit to reflect that the __init__ ought to be called if the base class has a non-default __init__ method, i.e. the one defined on object (this is also mentioned here in the docs). I was wondering if it should be mentioned that, when dealing with dataclasses, this is likely the case (I think that the generation of __init__ by @dataclass might obscure the presence of a non-default __init__).

That's a great example.

I don't know that anyone would know what "non-default __init__ method" means. I'd just leave it as "an __init__ method that needs to be called", or similar. Or maybe add words that say that in general, dataclass-generated __init__ methods to not need to be explicitly called, because the derived class __init__ takes care of initializing all fields.

This looks good. How about instead of "the derived class", use "the derived dataclass"? I'm trying to draw the distinction that if a non-dataclass is derived from a dataclass, then the base class's __init__ should in fact be called.

If I understand PEP 557 correctly, the __init__ method needs to be called if a dataclass inherits from a non-dataclass or if a non-dataclass inherits from a dataclass. I have tried to reflect this in the new wording.

If I understand PEP 557 correctly, the __init__ method needs to be called if a dataclass inherits from a non-dataclass or if a non-dataclass inherits from a dataclass. I have tried to reflect this in the new wording.

If a dataclass inherits from a non-dataclass, then it's probably that the base non-dataclass's __init__ does need to be called. That's what __post_init__ is for, among other uses.

If a non-dataclass inherits from a dataclass, then the non-dataclass most likely does need to call the dataclass's __init__.

It's only the case where a dataclass inherits from a dataclass that the base class's __init__ most likely doesn't need to be called. I say "most likely" because you could construct examples where it does need to be called, if the base dataclass's __post_init__ does some processing. But it's probably too complex to mention that here, I'm just mentioning it here to help you understand the issue. Apologies if I'm not being helpful.

This is very helpful! Thank you!

"not need to" isn't correct. Apologies if that was my error. Maybe simpler language is needed, like "... methods don't need to be called, since ...`.

Sorry, English is not my native tongue

If I understand PEP 557 correctly, the __init__ method needs to be called if a dataclass inherits from a non-dataclass or if a non-dataclass inherits from a dataclass. I have tried to reflect this in the new wording.

If a dataclass inherits from a non-dataclass, then it's probably that the base non-dataclass's __init__ does need to be called. That's what __post_init__ is for, among other uses.

If a non-dataclass inherits from a dataclass, then the non-dataclass most likely does need to call the dataclass's __init__.

It's only the case where a dataclass inherits from a dataclass that the base class's __init__ most likely doesn't need to be called. I say "most likely" because you could construct examples where it does need to be called, if the base dataclass's __post_init__ does some processing. But it's probably too complex to mention that here, I'm just mentioning it here to help you understand the issue. Apologies if I'm not being helpful.