bazel-contrib / bazel-contrib/rules_python

Have `python_binary_naming_convention` directive also apply to `if __name__ == "__main__":` files

オープン
#1,815 コメント 2 件 リアクション 4 件 担当者 0 名 GitHub で見る
Can Close? gazelle
主要言語
Starlark
スター
688
フォーク
721
平均マージ
15時間 7分
マージ済み PR(30日)
76

説明

# 🚀 feature request

### Relevant Rules

+ gazelle

### Description

The `python_binary_naming_convention` directive currently only applies to binaries that are generated from `__main__.py` files.

`py_binary` targets that are generated from files that contain `if __name__ == "__main__":` are named after their source file. From the [README](https://github.com/bazelbuild/rules_python/blob/e86252ffd6d1a1bf32ae99933acc5ab49b78ec1e/gazelle/README.md#binaries) (emphasis mine):

> If `python_generation_mode` is set to `file`, then instead of one `py_binary` target per module, Gazelle will create one `py_binary` target for each file with such a line, and **the name of the target will match the name of the script**.

This results in name collisions when a single file is both a binary and a library and gazelle will report:

> gazelle: failed to generate target "//foobar" of kind "py_binary": a target of kind "py_library" with the same name already exists

The readme also says:

> Note that it's possible for another script to depend on a py_binary target and import from the py_binary's scripts. This can have possible negative effects on Bazel analysis time and runfiles size compared to depending on a py_library target. The simplest way to avoid these negative effects is to extract library code into a separate script without a main line. Gazelle will then create a py_library target for that library code, and other scripts can depend on that py_library target.

However, for existing projects that might not be feasible.

### Describe the solution you'd like

Have `python_binary_naming_convention` directive also apply to files that contain `if __name__ ...`.

Or add another directive that accomplishes such.

### Describe alternatives you've considered

Right now I'm just manually making all ~450 or so `py_binary` targets that our monorepo has :sob:

```starlark
# libs are generated by gazelle
py_library(
name = "foo",
srcs = ["foo.py"],
deps = [
":bar",
"//bin/bom:boo",
"@pypi:hello",
],
)

# binaries are manually added, using the lib as a dependency so that I don't have to keep `deps` in sync.
py_binary(
name = "foo_bin",
srcs = ["foo.py"],
deps = [":foo"],
)
```

コントリビューションガイド

コントリビューションガイドを開く

調査の方向性

Start with Gazelle's `python_binary_naming_convention` directive and the Binaries section of the Gazelle README, comparing generation from `__main__.py` with files containing `if __name__ == "__main__":`. Done means the convention applies to both cases and prevents the reported `py_library`/`py_binary` name collision.

索引モデルが issue の本文から書いたものです。

評価

技術スタック
python
領域
build-system, tooling
issue の種類
機能追加
難易度
4/5
見積もり時間
3〜5日
活発さ
活発
明瞭さ
おおむね明確
初心者へのやさしさ
55/100

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。