python-attrs / python-attrs/cattrs
Custom unstructure function is not called when a union type is wrapped in generic types multiple times
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.1k
- Forks
- 159
- Avg merge
- 12h 21m
- Merged PRs (30d)
- 6
Description
- cattrs version: 22.2.0
- Python version: 3.9
- Operating System: Ubuntu 20.04
Description
When you wrap a union type multiple times, e.g., Optional[list[Union[A, B]]], the custom unstructure function on the union type is not called. I believe this is related to https://github.com/python-attrs/cattrs/issues/129.
What I Did
import attrs
import cattrs
from typing import Optional, Union
c = cattrs.Converter()
@attrs.define
class A:
a: int
@attrs.define
class B:
a: int
@attrs.define
class C:
f: Optional[list[Union[A, B]]]
c.register_unstructure_hook(
Union[A, B],
lambda o: {"_type": o.__class__.__name__, **c.unstructure(o)},
)
c.register_structure_hook(
Union[A, B], lambda o, t: c.structure(o, A if o["_type"] == "A" else B)
)
inst = C([A(1), B(2)])
unstructured = c.unstructure(inst)
assert unstructured["f"][0]["_type"] == "A"
assert unstructured["f"][1]["_type"] == "B"
assert c.structure(unstructured, C) == inst
You should get the following error:
❯ python reproduce.py
Traceback (most recent call last):
File "/home/peter/programming/seagull/reproduce.py", line 34, in <module>
assert unstructured["f"][0]["_type"] == "A"
KeyError: '_type'
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by running the provided Python reproduction with cattrs 22.2.0 and Python 3.9, then trace how the converter handles the custom hooks for Optional[list[Union[A, B]]]. Done means the assertions for _type tags and round-trip structuring pass without the KeyError.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- backend
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Clearly specified
- Newbie friendliness
- 45/100