python-attrs / python-attrs/attrs

Eliminate evolve() boilerplate with evolvers/copy

Open
#501 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Feature
Dominant language
Python
Stars
5.8k
Forks
480
Avg merge
2h 15m
Merged PRs (30d)
2

Description

When I want to evolve() an object, I sometimes have to do a bit of boilerplate copy()ing:

from typing import List

import attr


@attr.s
class Car:
    model: str = attr.ib()
    occupants: List = attr.ib()


car1 = Car("Toyota", ["Alice", "Bob"])
car2 = attr.evolve(car1, model="Honda", occupants=car1.occupants.copy())

To eliminate this boilerplate, one option would be:

@attr.s
class Car:
    model: str = attr.ib()
    occupants: List = attr.ib(evolver=lambda x: x.copy())

Or similarly:


@attr.s
class Car:
    model: str = attr.ib()
    occupants: List = attr.ib()

    @occupants
    def evolver(self):
        return self.occupants.copy()

Or using copy.copy() via the __copy__ method:

@attr.s
class Car:
    model: str = attr.ib()
    occupants: List = attr.ib(evolve_copy=True)
    

The latter idea suggests the option of putting the copy functionality into @attr.s, which could add a __copy__ method to the class, and like the other methods it could be enabled/disabled for each attribute with attr.ib(copy=True/False), and evolve() could use copy.copy() by default.

@attr.s(copy=True)
class Car:
    model: str = attr.ib()
    occupants: List = attr.ib()
    components: pyrsistent.PSet = attr.ib(copy=False)

cc: @altendky

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the existing evolve(), attr.s, and attr.ib behavior described in the issue, then compare the proposed evolver, decorator, and copy approaches. The issue does not name implementation files or tests; the work is done when one agreed API is specified, implemented, and covered by tests for copied and non-copied attributes.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
tooling
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.