Python dataclass setting default list with values

Question

Can anyone help me fix this error. I just started using dataclass I wanted to put a default value so I can easily call from other function

I have this class

@dataclass(frozen=True)
class MyClass:
    my_list: list = ["list1", "list2", "list3"]
    my_list2: list = ["list1", "list2", "list3"]

But when i print print(MyClass.my_list) I'm getting this error

 raise ValueError(f'mutable default {type(f.default)} for field '
ValueError: mutable default <class 'list'> for field my_list is not allowed: use default_factory

I know it's allowed, but it seems kind of anti-pattern to make a frozen dataclass with mutable objects. Are you sure this is what you want to do? — sj95126
– sj95126, Commented Oct 14, 2021 at 3:54
Ahm only i wanted is to put a default list values. Although I can use a dictionary for easy use but I'm currently learning dataclass and this helps me with typing hints — mark12345
– mark12345, Commented Oct 14, 2021 at 3:56
@sj95126 you mean I don't need to include frozen? since it's already mutable? How about if i wanted to put other data type? like str — mark12345
– mark12345, Commented Oct 14, 2021 at 4:03
Actually, removing frozen=True won't change the error, but all I meant is that as a general concept, a 'frozen' object with mutable contents is somewhat contradictory. There can be complications with hashing (see the documentation) so it may be better to avoid frozen in this case if you don't need it. — sj95126
– sj95126, Commented Oct 14, 2021 at 4:21

Evan Summers · Accepted Answer · 2021-10-16 04:17:14Z

27

What it means by mutable default is that the lists provided as defaults will be the same individual objects in each instance of the dataclass. This would be confusing because mutating the list in an instance by e.g. appending to it would also append to the list in every other instance.

Instead, it wants you to provide a default_factory function that will make a new list for each instance:

from dataclasses import dataclass, field

@dataclass
class MyClass:
    my_list: list = field(default_factory=lambda: ["list1", "list2", "list3"])
    my_list2: list = field(default_factory=lambda: ["list1", "list2", "list3"])

edited Oct 16, 2021 at 4:17

answered Oct 14, 2021 at 3:49

Evan Summers

1,2729 silver badges19 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

mark12345 Over a year ago

Thank you it fixed! At first I got error: AttributeError: type object 'MyClass' has no attribute 'my_list I forgot to include parenthesis. Now I get it. thank you so much

mark12345 Over a year ago

I need to wait for 5minutes to accept your answer

ashtonium · Accepted Answer · 2024-09-04 14:06:18Z

3

As the first comment notes, it's a bit odd to have a mutable item in a dataclass. If you don't need it to be mutable, a simpler solution is to initialize it as a tuple.

from dataclasses import dataclass

@dataclass
class MyClass:
    my_list: tuple[str] = ("list1", "list2", "list3")
    my_list2: tuple[str] = ("list1", "list2", "list3")

(For example, you might want a tuple of fieldnames associated with your dataclass for serialization with DictWriter.)

edited Sep 4, 2024 at 14:06

ashtonium

2,4961 gold badge21 silver badges25 bronze badges

answered Jun 2, 2022 at 0:37

Sboisen

893 bronze badges

Collectives™ on Stack Overflow

Python dataclass setting default list with values

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related