maint: Add prettier pre-commit hook #3663

mathbunnyru · 2024-12-07T02:23:05Z

prettier is a much better tool than using hooks from https://github.com/pre-commit/pre-commit-hooks - the style is much better, always consistent and it actually works

mathbunnyru · 2024-12-07T02:24:21Z

.pre-commit-config.yaml

    hooks:
-    - id: rst-backticks
-    - id: rst-directive-colons
-    - id: rst-inline-touching-normal
+      - id: rst-backticks


This an example when existing hooks allow different styles (even in the same file)

mathbunnyru · 2024-12-07T02:24:32Z

.pre-commit-config.yaml

@@ -5,28 +5,25 @@ repos:
    hooks:
      - id: trailing-whitespace
      - id: end-of-file-fixer
-      - id: fix-encoding-pragma


This is deprecated

mathbunnyru · 2024-12-07T02:25:00Z

.pre-commit-config.yaml

-      - id: check-yaml
-        exclude: ^.+(/tests/|/recipe/).+$
-      - id: check-toml
-      - id: check-json
      - id: check-merge-conflict
-      - id: pretty-format-json
-        args: [--autofix]


These checks (except check-merge-conflict) no longer seem to be needed

mathbunnyru · 2024-12-07T02:25:26Z

.pre-commit-config.yaml

      - id: debug-statements
        language_version: python3
+  # Autoformat: YAML, JSON, Markdown, etc.
+  - repo: https://github.com/rbubley/mirrors-prettier


Using a fork, since official prettier pre-commit hook has been archived:
http://github.com/prettier/prettier/issues/15742

Hum, I'm kind of reluctant here. As I'm in favor of general improvements, using something that just has been archived (even if a mirror is available) doesn't seem a good idea to me.
https://github.com/rbubley/mirrors-prettier seems to rather be a workaround to let people keep using the hook and there are no guarantees that it will be maintained in the long term.
I would highly suggest to wait until a stable solution is available (could be something used by a significant number of people, regularly maintained, or officially superseding the archived one in pre-commit).

There is an important gotcha: the https://github.com/rbubley/mirrors-prettier is not a mirror/fork of prettier ore pre-commit.
It's a small repo to make vanilla prettier work as pre-commit hook.
If you take a look at the source code, you'll see that the part which matters is less than 100 lines.
So, there is not much to maintain and the code to automatically update the version of the prettier is already there.

Even in the worst case scenario, this fork will stop updating, we will have some prettier version in this repo.
It's already working great for lots of files like json/yaml/toml/markdown.
I don't think these files have that much new syntax features added to them (unlike python/c++ code, where you really want tools to work with newer code as well).
So, it's unlikely we'll encounter issues due to not using the latest version.

And the sooner we add prettier, the less diff this PR will generate.

mathbunnyru · 2024-12-07T02:38:48Z

releaser.py

@@ -39,8 +39,7 @@ def template_substitute(contents):
    today = datetime.date.today()
    fmt_today = today.strftime("%B %d, %Y")

-    header_line = f"{name} {version} ({fmt_today})"
-    res += f"{header_line}\n{'=' * len(header_line)}\n\n"
+    res += f"# {name} {version} ({fmt_today})\n\n"


releaser.py and changelog.py are changed manually by me, to make sure new changelogs are written correctly

mathbunnyru · 2024-12-07T02:40:26Z

CHANGELOG.md

@@ -40,15 +39,14 @@ Bug fixes:
 - [libmamba, micromamba] Create empty base prefix with `env update` by @Hind-M in https://github.com/mamba-org/mamba/pull/3519
 - [libmamba, micromamba] fix: Use POSIX-compliant scripts by @jjerphan in https://github.com/mamba-org/mamba/pull/3522
 - [libmamba, micromamba] maint: Clarify `env` subcommand documentation in help menu (cont'd) by @jjerphan in https://github.com/mamba-org/mamba/pull/3539
- [libmamba] fix: Handle space in `mamba` and `micromamba` executable absolute paths  by @NewUserHa in https://github.com/mamba-org/mamba/pull/3525
+- [libmamba] fix: Handle space in `mamba` and `micromamba` executable absolute paths by @NewUserHa in https://github.com/mamba-org/mamba/pull/3525


Sometimes prettier fixes double space and some other minor things in markdown text, don't think this is gonna cause trouble

mathbunnyru · 2024-12-07T02:45:07Z

Please, let me know if you don't like PRs with improvements to tooling around the code quality.
I think this can greatly improve readability and consistency and we only need to run the tool once, and then all the fixes will be automatic.

jjerphan

Thank you for this proposal, @mathbunnyru.

I am not against it, but I think we should ignore the formatting done by this PR then.

I am waiting for someone else from the team to comment.

mathbunnyru · 2024-12-25T13:59:19Z

@jjerphan @Hind-M should I close this PR then?

Hind-M · 2024-12-30T08:24:21Z

Any other opinion on this @Klaim @JohanMabille?

Klaim · 2024-12-30T10:27:24Z

Any other opinion on this @Klaim @JohanMabille?

I don't have a clear opinion at the moment but a few notes:

I share the same worry that relying on a potentially unmaintained script will backfire, I didn't check yet the details mentioned in review comments yet;
I'm not a fan of the array formatting in the JSONs as I find vertical lists easier to eyeball-scan scan when there are more than 3-4 elements - I suppose that's a formatting option to keep as before?;
It's not clear to me why the changes on the releasing scripts were necessary? (I don't mean the refactoring but the change of tokens searched).

mathbunnyru · 2025-01-09T11:17:56Z

Any other opinion on this @Klaim @JohanMabille?

I don't have a clear opinion at the moment but a few notes:

I share the same worry that relying on a potentially unmaintained script will backfire, I didn't check yet the details mentioned in review comments yet;

Please, take a look here: #3663 (comment)

I'm not a fan of the array formatting in the JSONs as I find vertical lists easier to eyeball-scan scan when there are more than 3-4 elements - I suppose that's a formatting option to keep as before?;

Unfortunately not: prettier/prettier-vscode#352
There is a workaround to insert // comment after first element of array, but that's not pretty at all.
prettier is not very configurable.

It's not clear to me why the changes on the releasing scripts were necessary? (I don't mean the refactoring but the change of tokens searched).

prettier changes markdown files from using alternative headers (like ==========) to using more flexible and common headings.
That's why we need to update the scripts to properly write changelogs and parse them.

mathbunnyru · 2025-01-09T11:27:47Z

I rebased and improved my commits - now Apply prettier pre-commit hook is where I actually run pre-commit run --all-files prettier and other commits are manually made, should be much easier to review.

I also ran python3 update_changelog.py followed by python3 releaser.py and the output was as expected.

mathbunnyru · 2025-01-09T11:34:22Z

@Klaim @jjerphan @Hind-M please, take a look, and if you have any questions, let me know.

I understand, that this is not straightforward PR, and there is a decision to be made, and I'm ok if you want to close the PR, rather than merge it.

At the same time, I think, all the files look significantly better than before.
And this change gives a consistency for the whole repo - C/C++ files are formatted by clang-format, (almost) all the other types - by prettier.
And if we decide to stop using prettier, it would be a few lines change in config, while we will keep quite beautiful config files.

jjerphan

LGTM iif there's a (optional) way to ignore the diff of the merge commit of this PR via blame.ignoreRevsFile. I am waiting for the rest of the team's opinion in this regard.

mathbunnyru · 2025-01-09T11:53:43Z

LGTM iif there's a (optional) way to ignore the diff of the merge commit of this PR via blame.ignoreRevsFile.

Thanks! I think you will squash the contents of this PR (if you decide to merge it), so I can only update .git-blame-ignore-revs after PR is merged.

I don't want to squash manually here, because it's quite difficult to rebase and review such PRs when everything is in one commit.

Klaim · 2025-01-09T14:55:56Z

I dont have any strong opinion about this, either way I'll follow what the rest of the team prefer. Maybe @JohanMabille has some opinon?

Hind-M

LGTM if it leads to improvements.
Just a side note on ignoring the diff of the merge commit, I think we should be consistent with the repo history here. And since previous changes on the pre-commit hooks were never ignored, we should continue that approach. Otherwise, we should consider ignoring all of them to avoid potential confusion for anyone reviewing the commit history without context.

jjerphan · 2025-01-10T15:51:03Z

Otherwise, we should consider ignoring all of them to avoid potential confusion for anyone reviewing the commit history without context.

I would be in favor of doing this after the merge.

mathbunnyru · 2025-01-11T11:34:38Z

Great, as soon as this gets merged, I will add all commits that changed .pre-commit-config.yaml to the .git-blame-ignore-revs file.

mathbunnyru · 2025-01-12T11:53:01Z

Great, as soon as this gets merged, I will add all commits that changed .pre-commit-config.yaml to the .git-blame-ignore-revs file.

#3738

Created a PR to unify the process of making such changes.
Will update the file, when both PRs get merged.

jjerphan · 2025-01-13T08:09:20Z

Thank you for your patience, @mathbunnyru.

mathbunnyru commented Dec 7, 2024

View reviewed changes

mathbunnyru changed the title ~~Add pretter pre-commit hook~~ maint: Add pretter pre-commit hook Dec 7, 2024

jjerphan changed the title ~~maint: Add pretter pre-commit hook~~ maint: Add prettier pre-commit hook Dec 9, 2024

jjerphan added the release::ci_docs For PRs related to CI or documentation label Dec 10, 2024

jjerphan reviewed Dec 10, 2024

View reviewed changes

mathbunnyru force-pushed the add_prettier branch 2 times, most recently from 96320fa to 2fc282a Compare January 9, 2025 11:10

mathbunnyru added 4 commits January 9, 2025 11:23

Add prettier pre-commit hook and update releaser.py/update_changelog.py

42a746c

Update version.py.tmpl

524c36e

Apply prettier pre-commit hook

1427274

Fix markdown H1 indentation

b262d49

mathbunnyru force-pushed the add_prettier branch from 2fc282a to b262d49 Compare January 9, 2025 11:24

jjerphan reviewed Jan 9, 2025

View reviewed changes

jjerphan approved these changes Jan 10, 2025

View reviewed changes

Hind-M approved these changes Jan 10, 2025

View reviewed changes

jjerphan merged commit d0c7458 into mamba-org:main Jan 13, 2025
34 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

maint: Add prettier pre-commit hook #3663

maint: Add prettier pre-commit hook #3663

mathbunnyru commented Dec 7, 2024

mathbunnyru Dec 7, 2024

mathbunnyru Dec 7, 2024

mathbunnyru Dec 7, 2024

mathbunnyru Dec 7, 2024

Hind-M Dec 16, 2024

mathbunnyru Dec 16, 2024

mathbunnyru Dec 7, 2024

mathbunnyru Dec 7, 2024

mathbunnyru commented Dec 7, 2024

jjerphan left a comment •

edited

Loading

mathbunnyru commented Dec 25, 2024

Hind-M commented Dec 30, 2024

Klaim commented Dec 30, 2024

mathbunnyru commented Jan 9, 2025

mathbunnyru commented Jan 9, 2025

mathbunnyru commented Jan 9, 2025

jjerphan left a comment •

edited

Loading

mathbunnyru commented Jan 9, 2025

Klaim commented Jan 9, 2025

Hind-M left a comment

jjerphan commented Jan 10, 2025

mathbunnyru commented Jan 11, 2025

mathbunnyru commented Jan 12, 2025

jjerphan commented Jan 13, 2025

maint: Add prettier pre-commit hook #3663

maint: Add prettier pre-commit hook #3663

Conversation

mathbunnyru commented Dec 7, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mathbunnyru commented Dec 7, 2024

jjerphan left a comment • edited Loading

Choose a reason for hiding this comment

mathbunnyru commented Dec 25, 2024

Hind-M commented Dec 30, 2024

Klaim commented Dec 30, 2024

mathbunnyru commented Jan 9, 2025

mathbunnyru commented Jan 9, 2025

mathbunnyru commented Jan 9, 2025

jjerphan left a comment • edited Loading

Choose a reason for hiding this comment

mathbunnyru commented Jan 9, 2025

Klaim commented Jan 9, 2025

Hind-M left a comment

Choose a reason for hiding this comment

jjerphan commented Jan 10, 2025

mathbunnyru commented Jan 11, 2025

mathbunnyru commented Jan 12, 2025

jjerphan commented Jan 13, 2025

jjerphan left a comment •

edited

Loading

jjerphan left a comment •

edited

Loading