mirror of
https://github.com/opendatalab/MinerU.git
synced 2026-10-25 17:42:08 +08:00
61 lines
2.6 KiB
Python
61 lines
2.6 KiB
Python
"""根据 GitHub 事件中的固定提交判定代码和文档是否需要验证。"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
import os
|
|
import re
|
|
import subprocess
|
|
from pathlib import Path
|
|
from typing import Any
|
|
|
|
|
|
def classify_paths(paths: list[str]) -> dict[str, bool]:
|
|
"""文档变更只构建文档，CI 公共脚本变更同时验证两个工作流。"""
|
|
docs = any(
|
|
path.startswith("docs/")
|
|
or path.startswith("README")
|
|
or path in {"mkdocs.yml", ".github/workflows/mkdocs.yml", "scripts/ci/changes.py"}
|
|
for path in paths
|
|
)
|
|
code = any(not (path.startswith(("docs/", "cmds/")) or path.endswith(".md") or path == "mkdocs.yml") for path in paths)
|
|
return {"code": code, "docs": docs}
|
|
|
|
|
|
def changed_paths(event_name: str, event: dict[str, Any], head: str) -> list[str] | None:
|
|
"""PR 对比合并提交与基分支；首次推送和手动/发布调用执行完整检查。"""
|
|
if event_name == "pull_request":
|
|
base = event["pull_request"]["base"]["sha"]
|
|
elif event_name == "push" and not event.get("ref", "").startswith("refs/tags/"):
|
|
base = event.get("before", "")
|
|
if not base or base == "0" * 40:
|
|
return None
|
|
else:
|
|
return None
|
|
if not all(re.fullmatch(r"[0-9a-f]{40}", value) for value in (base, head)):
|
|
raise ValueError("Expected full commit SHAs in the GitHub event")
|
|
# 被强制推送移出的提交可能不在 checkout 中；无法取回时保守运行全部检查。
|
|
if subprocess.run(["git", "cat-file", "-e", f"{base}^{{commit}}"], capture_output=True).returncode:
|
|
fetched = subprocess.run(["git", "fetch", "--no-tags", "--depth=1", "origin", base], capture_output=True)
|
|
if fetched.returncode:
|
|
return None
|
|
result = subprocess.run(
|
|
["git", "diff", "--name-only", "--no-renames", "-z", base, head], check=True, capture_output=True, text=True
|
|
)
|
|
return [path for path in result.stdout.split("\0") if path]
|
|
|
|
|
|
def main() -> None:
|
|
"""把变化分类写入 Actions 输出，供始终存在的汇总检查决定跳过哪些任务。"""
|
|
event = json.loads(Path(os.environ["GITHUB_EVENT_PATH"]).read_text(encoding="utf-8"))
|
|
paths = changed_paths(os.environ["GITHUB_EVENT_NAME"], event, os.environ["GITHUB_SHA"])
|
|
changes = {"code": True, "docs": True} if paths is None else classify_paths(paths)
|
|
with Path(os.environ["GITHUB_OUTPUT"]).open("a", encoding="utf-8") as output:
|
|
for name, changed in changes.items():
|
|
output.write(f"{name}={str(changed).lower()}\n")
|
|
print(json.dumps(changes))
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|