dataclasses.astuple (and .asdict) do deepcopy on all fields

@ericvsmith

BPO	43905
Nosy	@ericvsmith, @serhiy-storchaka, @miss-islington, @akulakov
PRs	bpo-43905: Expand dataclasses.astuple() and asdict() docs #26154 [3.10] bpo-43905: Expand dataclasses.astuple() and asdict() docs (GH-26154) #29851 [3.9] bpo-43905: Expand dataclasses.astuple() and asdict() docs (GH-26154) #29852

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = 'https://github.com/ericvsmith'
closed_at = <Date 2021-11-29.18:34:32.678>
created_at = <Date 2021-04-21.12:19:17.095>
labels = ['3.11', 'type-bug', '3.9', '3.10', 'docs']
title = 'dataclasses.astuple (and .asdict) do deepcopy on all fields'
updated_at = <Date 2021-11-29.18:44:26.912>
user = 'https://bugs.python.org/mandolaerik'

bugs.python.org fields:

activity = <Date 2021-11-29.18:44:26.912>
actor = 'andrei.avk'
assignee = 'eric.smith'
closed = True
closed_date = <Date 2021-11-29.18:34:32.678>
closer = 'eric.smith'
components = ['Documentation']
creation = <Date 2021-04-21.12:19:17.095>
creator = 'mandolaerik'
dependencies = []
files = []
hgrepos = []
issue_num = 43905
keywords = ['patch']
message_count = 14.0
messages = ['391516', '391518', '391519', '391520', '394015', '394019', '396284', '407241', '407243', '407306', '407310', '407311', '407314', '407316']
nosy_count = 5.0
nosy_names = ['eric.smith', 'serhiy.storchaka', 'miss-islington', 'andrei.avk', 'mandolaerik']
pr_nums = ['26154', '29851', '29852']
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue43905'
versions = ['Python 3.9', 'Python 3.10', 'Python 3.11']

Linked PRs

gh-88071: Update docstrings of dataclass' astuple and asdict #101806

It seems that the 'dataclass.astuple' function does a deepcopy of all fields. This is not documented. Two problems:

Dictionary keys that rely on object identity are ruined:

import dataclasses
@dataclasses.dataclass
class Foo:
    key: object
key = object()
lut = {key: 5}
(y,) = dataclasses.astuple(Foo(x))
\# KeyError
lut[y]

dataclasses can only be converted to a tuple if all fields are serializable:

import dataclasses
@dataclasses.dataclass
class Foo:
    f: object
foo = Foo(open('test.py'))
dataclasses.astuple(foo)

->

TypeError: cannot pickle '_io.TextIOWrapper' object

In my use case, I just want a list of all fields. I can do the following as a workaround:

(getattr(foo, field.name) for field in dataclasses.fields(foo))

Tested on Python 3.8.7 and 3.7.9.

Unfortunately this can't be changed, although I suppose it should be documented.

In general I think this API was a mistake, and should not have been added. There are just too many cases where it doesn't do what you want, or where it fails.

I'd like to deprecate it and remove it (along with asdict), but I fear that would be too disruptive.

Your approach seems reasonable to me for your use case.

Why deepcopy is used at all? It is a very specific feature which should not be used by default. If you want to make a deep copy of fields, you can call copy.deepcopy() explicitly.

    copy.deepcopy(dataclasses.astuple(obj))

or

    dataclasses.astuple(copy.deepcopy(obj))

The reason for the deep copying was to support changing a hierarchy of dataclasses into something that could be JSON serialized. But it didn't really work out. It recurses into dataclasses, namedtuples, lists, tuples, and dicts, and deep copies everything else.

As I said, it's a design flaw.

Would it make sense to make dataclasses iterable, like so?

    def __iter__(self):
        return (getattr(self, field.name) for field in fields(self))

With that in place, deprecating astuple would maybe be less disruptive?

No, iteration is explicitly a non-goal of PEP-557. See the section on namedtuple for why: https://www.python.org/dev/peps/pep-0557/#why-not-just-use-namedtuple

I've added a PR here: #26154

New changeset c1f93f0 by andrei kulakov in branch 'main':
bpo-43905: Expand dataclasses.astuple() and asdict() docs (GH-26154)
c1f93f0

New changeset 376b24e by Miss Islington (bot) in branch '3.9':
bpo-43905: Expand dataclasses.astuple() and asdict() docs (GH-26154)
376b24e

New changeset 32f1491 by Miss Islington (bot) in branch '3.10':
bpo-43905: Expand dataclasses.astuple() and asdict() docs (GH-26154)
32f1491

Thanks, @andrei.avk!

Thank you for reviewing Eric!

gh-88071: Update docstrings of dataclass' astuple and asdict (#101806)

pythongh-88071: Update docstrings of dataclass' astuple and asdict (p…

Navigation Menu

Repository navigation

dataclasses.astuple (and .asdict) do deepcopy on all fields #88071

Linked PRs

15 remaining items

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

dataclasses.astuple (and .asdict) do deepcopy on all fields #88071

Description

Linked PRs

Activity

mandolaerik commented on Apr 21, 2021

ericvsmith commented on Apr 21, 2021

serhiy-storchaka commented on Apr 21, 2021

ericvsmith commented on Apr 21, 2021

mandolaerik commented on May 20, 2021

ericvsmith commented on May 20, 2021

akulakov commented on Jun 21, 2021

15 remaining items

miss-islington commented on Nov 29, 2021

miss-islington commented on Nov 29, 2021

miss-islington commented on Nov 29, 2021

ericvsmith commented on Nov 29, 2021

akulakov commented on Nov 29, 2021

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions