flake8 >4 support #132
Comments
|
Indeed. It does not look to be a quick patch though... 🤷
If someone wants to tackle this, I guess we could have a more thorough conversation about it. To answer your question: It would take a considerable refactor. |
|
Can you (or someone) define a more specific list of things that would need to be changed? I think supporting newer linter versions is essential and must come at some point, but I had such a good experience with Flakeheaven that I wouldn't like to stop using it in exchange for being able to update the linters. |
|
May be we can define list of microtasks (issues), which help make migrate to flake8 v5 more simple? For example |
|
I like this idea. We would need to refactor a bunch of stuff, and depending on the specific "patch", it would not be straightforward to support both versions as flakeheaven uses flake8 internal api which changes inadvertently. In those cases i guess we'd have to use a facade or similar to support both. This would also facilitate future migrations and contributions. So the general idea would be
So the initial plan I guess would look like
|
|
For steps 2 and 3, if it's going to be painful to support both versions, then perhaps there should be a clean break with flakeheaven 3.x supporting flake8 4 and flakeheaven 4.x supporting flake8 5? On the other hand, a facade might be useful if flake8 continues to change in breaking ways. |
|
I agree with @mcarans that supporting both versions could be hard, and it would also likely delay support for flake8 5 significantly. What about releasing a breaking version of flakeheaven which supports flake8 5 without necessarily retaining support for earlier versions? It would give a quicker option to those who want to use flake8 5 (the others can just not upgrade), and also provide a way to explore what compatibility issues are actually there. Then a further version can be developed that supports the newest and older versions, if possible. |
|
I also agree with @mcarans. I'd do as soon as possible a 4.0.0rc1 so that people can start using it and raise issues to be fixed. The idea of the facade makes a lot of sense too. Has someone started working in this direction? The pin of flake8 < 5 is killing me, so I'd be willing to spend some time on this |
|
@lyz-code sounds great! I had an initial approach feat/flake8v5 😳 , but havent had the time to complete it. Maybe we could setup a chat/talk (discord?) and list what would be required. |
|
I'll be here for the next 30 mins if you can join that would be great |
|
Pushed a commit to the feat/flake8v5 - see ddee34a (just commented out some imports to have a working start point) changes in the branch not strictly related to flake8v5:
Current MO so far:
I left it at So far, all commands except the |
|
Thanks @pwoolvett for the detailed description, it will make it much easier to contribute <3 |
feat(click#File System Isolation): File System Isolation
For basic command line tools with file system operations, the
`CliRunner.isolated_filesystem()` method is useful for setting the current
working directory to a new, empty folder.
```python
from click.testing import CliRunner
from cat import cat
def test_cat():
runner = CliRunner()
with runner.isolated_filesystem():
with open("hello.txt", "w") as f:
f.write("Hello World!")
result = runner.invoke(cat, ["hello.txt"])
assert result.exit_code == 0
assert result.output == "Hello World!\n"
```
Pass `temp_dir` to control where the temporary directory is created. The
directory will not be removed by Click in this case. This is useful to integrate
with a framework like Pytest that manages temporary files.
```python
def test_keep_dir(tmp_path):
runner = CliRunner()
with runner.isolated_filesystem(temp_dir=tmp_path) as td:
...
```
feat(python_snippets#Pathlib make parent directories if they don't exist): Pathlib make parent directories if they don't exist
```python
pathlib.Path("/tmp/sub1/sub2").mkdir(parents=True, exist_ok=True)
```
From the
[docs](https://docs.python.org/3/library/pathlib.html#pathlib.Path.mkdir):
- If `parents` is `true`, any missing parents of this path are created as
needed; they are created with the default permissions without taking mode into
account (mimicking the POSIX `mkdir -p` command).
- If `parents` is `false` (the default), a missing parent raises
`FileNotFoundError`.
- If `exist_ok` is `false` (the default), `FileExistsError` is raised if the
target directory already exists.
- If `exist_ok` is `true`, `FileExistsError` exceptions will be ignored (same
behavior as the POSIX `mkdir -p` command), but only if the last path component
is not an existing non-directory file.
feat(python_snippets#Pathlib touch a file): Pathlib touch a file
Create a file at this given path.
```python
pathlib.Path("/tmp/file.txt").touch(exist_ok=True)
```
If the file already exists, the function succeeds if `exist_ok` is `true` (and
its modification time is updated to the current time), otherwise
`FileExistsError` is raised.
If the parent directory doesn't exist you need to create it first.
```python
global_conf_path = xdg_home / "autoimport" / "config.toml"
global_conf_path.parent.mkdir(parents=True)
global_conf_path.touch(exist_ok=True)
```
fix(pdm#Solve circular dependencies): Solve circular dependencies by manual constraining
It also helps to run `pdm update` with the `-v` flag, that way you see which are
the candidates that are rejected, and you can put the constrain you want. For
example, I was seeing the next traceback:
```
pdm.termui: Conflicts detected:
pyflakes>=3.0.0 (from <Candidate autoflake 2.0.0 from https://pypi.org/simple/autoflake/>)
pyflakes<2.5.0,>=2.4.0 (from <Candidate flake8 4.0.1 from unknown>)
```
So I added a new dependency to pin it:
```
[tool.pdm.dev-dependencies]
dependencies = [
# Until flakeheaven supports flake8 5.x
# flakeheaven/flakeheaven#132
"flake8>=4.0.1,<5.0.0",
"pyflakes<2.5.0",
]
```
If none of the above works, you can override them:
```
[tool.pdm.overrides]
"importlib-metadata" = ">=3.10"
```
|
Maybe this issue should be renamed flake8 > 4 support since there is already flake8 v6. |
|
This would be a great upgrade indeed. Autopep8 cannot be installed to its latest version with poetry since the dependencies cannot be resolved on pycodestyle |
@pwoolvett I think supporting both flake8 versions is a bit too much work for only a temporary gain. I believe a breaking change is acceptable in this case, and people stuck on flake8 v4 can use older versions of flakeheaven. As it is right now, flakeheaven is becoming unusable as other tools have started moving to flake8 v5 or v6. Just my thoughts, but I do not understand the implementation details enough. |
|
Just an FYI, we started moving into heaven but it appears that we have to go back to hell until we can bring flake8 with us. We appreciate the work yall. |
Replace bare `except:` with `except Exception`, there are some cases where it can be narrowed further, but this is already an improvement over the current situation. This is somewhat pursuant to eliminating [flakeheaven](https://github.com/flakeheaven/flakeheaven), as it no longer supports the latest version of flake8 [[ref](flakeheaven/flakeheaven#132)]. But it also is just the right thing to do as bare exceptions can cause problems.
Replace bare `except:` with `except Exception`, there are some cases where it can be narrowed further, but this is already an improvement over the current situation. This is somewhat pursuant to eliminating [flakeheaven](https://github.com/flakeheaven/flakeheaven), as it no longer supports the latest version of flake8 [[ref](flakeheaven/flakeheaven#132)]. But it also is just the right thing to do as bare exceptions can cause problems.
Eliminate some occurrences of the following flake8 errors in tests: * E265 block comment should start with '# ' * E266 too many leading '#' for block comment * E402 module level import not at top of file * E712 comparison to False should be 'if cond is False:' or 'if not cond:' * E712 comparison to True should be 'if cond is True:' or 'if cond:' * E722 do not use bare 'except' * F401 ... imported but unused * F403 ... used; unable to detect undefined names * F405 ... may be undefined, or defined from star imports: ... * F541 f-string is missing placeholders * F841 local variable 'result' is assigned to but never used * N806 variable 'TEST_DIR' in function should be lowercase This is pursuant to eliminating [flakeheaven](https://github.com/flakeheaven/flakeheaven), as it no longer supports the latest version of flake8 [[ref](flakeheaven/flakeheaven#132)].
Eliminate some occurrences of the following flake8 errors in tests: * E265 block comment should start with '# ' * E266 too many leading '#' for block comment * E402 module level import not at top of file * E712 comparison to False should be 'if cond is False:' or 'if not cond:' * E712 comparison to True should be 'if cond is True:' or 'if cond:' * E722 do not use bare 'except' * F401 ... imported but unused * F403 ... used; unable to detect undefined names * F405 ... may be undefined, or defined from star imports: ... * F541 f-string is missing placeholders * F841 local variable 'result' is assigned to but never used * N806 variable 'TEST_DIR' in function should be lowercase This is pursuant to eliminating [flakeheaven](https://github.com/flakeheaven/flakeheaven), as it no longer supports the latest version of flake8 [[ref](flakeheaven/flakeheaven#132)].
|
Hey guys! Thank you for the hard work, is there any eta for this? |
|
See this discussion. flakeheaven needs new maintainers although personally I am not using it any more and am using ruff now. In case it's of any use for those wanting to migrate, this is my team's setup with Black, ruff etc. being called from Hatch with configuration in a |
|
Thank you @mcarans, I wonder if you use with your current setup legacy code base support, like the baseline |
|
@Pixel-Minions Prior to Hatch and ruff with a Before I switched from flakeheaven to flake8-pyproject, it looked like this: |
|
botocove depends on flakeheaven and pep8-naming. I want botocove to use the latest version of its development dependencies. flakeheaven 3.3.0 needs flake8 4. pep8-naming 0.13.3 needs flake8 5. So I can't install these latest versions together because they conflict over flake8.
(I scored out the above because pep8-naming won't restore support for flake8 4. They formally dropped support for flake8 4 in PR 210 because it wasn't working anyway.) If flake8 5 is the future then flakeheaven needs to adapt to it to remain relevant. Use this [tool.poetry]
name = "test"
version = "0"
description = ""
authors = [""]
[tool.poetry.dependencies]
python = "^3.8"
[tool.poetry.dev-dependencies]
pep8-naming = ">=0.13.3"
flakeheaven = ">=3.3.0"
[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"$ poetry install
Updating dependencies
Resolving dependencies... (0.3s)
Because no versions of flakeheaven match >3.3.0
and flakeheaven (3.3.0) depends on flake8 (>=4.0.1,<5.0.0), flakeheaven (>=3.3.0) requires flake8 (>=4.0.1,<5.0.0).
And because pep8-naming (0.13.3) depends on flake8 (>=5.0.0)
and no versions of pep8-naming match >0.13.3, flakeheaven (>=3.3.0) is incompatible with pep8-naming (>=0.13.3).
So, because test depends on both pep8-naming (>=0.13.3) and flakeheaven (>=3.3.0), version solving failed. |
|
Not a real answer, but my solution was to simply switch to ruff completely. |
Sorry to hear that. I suggest ruff, although depending on the complexity of your lint it might take you some time. Alternatively, as you mentioned pinnning down pep8-naming should work. If you're willing to take a shot on the migration though, you're more than welcome! |
|
Thanks @berzi and @pwoolvett for the quick feedback to remind about ruff. Now if even the flakeheaven maintainer recommends ruff I have a compelling reason to change 🙂 As a quick fix I'll use an older version of pep8-naming and then I'll look into migrating to ruff. Thanks for all your lint support so far! |
|
In my experience the move to ruff is rather painless, it can even read the codes you have configured for flake8 & friends. The only issue is if you have custom linting rules, because ruff doesn't support those yet. Good luck with your project. :) |
Hi!
It looks like this supports flake8 v4, but not v5? What would it take to bring it up to flake8 v5?
Thank you!
The text was updated successfully, but these errors were encountered: