mirror of
https://github.com/opendatalab/MinerU.git
synced 2026-10-25 17:42:08 +08:00
85 lines
3.9 KiB
Python
85 lines
3.9 KiB
Python
"""校验固定源码提交、release 标签和最终 wheel，输出可追溯的发布记录。"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import argparse
|
|
import hashlib
|
|
import json
|
|
import re
|
|
import subprocess
|
|
import tomllib
|
|
from email.parser import BytesParser
|
|
from pathlib import Path
|
|
from zipfile import ZipFile
|
|
|
|
from packaging.specifiers import SpecifierSet
|
|
from packaging.version import Version
|
|
|
|
|
|
def source_version(root: Path, tag: str) -> Version:
|
|
"""发版只验证已经提交的版本，拒绝用移动分支或临时改版本替代标签源码。"""
|
|
source = (root / "mineru/version.py").read_text(encoding="utf-8")
|
|
match = re.search(r'^__version__ = "([^"]+)"$', source, re.MULTILINE)
|
|
if match is None:
|
|
raise ValueError("Missing __version__ assignment")
|
|
version = Version(match.group(1))
|
|
if tag:
|
|
tag_match = re.fullmatch(r"mineru-(.+)-released", tag)
|
|
if tag_match is None or Version(tag_match.group(1)) != version:
|
|
raise ValueError(f"Release tag {tag!r} must match the committed version {version}; update it before tagging")
|
|
return version
|
|
|
|
|
|
def verify_wheel(root: Path, directory: Path, version: Version) -> dict[str, str]:
|
|
"""检查实际发布产物的版本、Python 范围和 classifiers，记录完整 wheel 摘要。"""
|
|
wheels = list(directory.glob("*.whl"))
|
|
if len(wheels) != 1:
|
|
raise ValueError(f"Expected exactly one wheel, found {len(wheels)}")
|
|
wheel = wheels[0]
|
|
project = tomllib.loads((root / "pyproject.toml").read_text(encoding="utf-8"))["project"]
|
|
with ZipFile(wheel) as archive:
|
|
metadata_paths = [name for name in archive.namelist() if name.endswith(".dist-info/METADATA")]
|
|
if len(metadata_paths) != 1:
|
|
raise ValueError("Wheel must have exactly one METADATA file")
|
|
metadata = BytesParser().parsebytes(archive.read(metadata_paths[0]))
|
|
for field in ("Name", "Version", "Requires-Python"):
|
|
if len(metadata.get_all(field, [])) != 1:
|
|
raise ValueError(f"Missing or duplicate wheel metadata: {field}")
|
|
if metadata["Name"] != project["name"] or Version(metadata["Version"]) != version:
|
|
raise ValueError("Wheel name/version does not match the release source")
|
|
if SpecifierSet(metadata["Requires-Python"]) != SpecifierSet(project["requires-python"]):
|
|
raise ValueError("Wheel Python requirement does not match pyproject.toml")
|
|
if set(metadata.get_all("Classifier", [])) != set(project["classifiers"]):
|
|
raise ValueError("Wheel classifiers do not match pyproject.toml")
|
|
with wheel.open("rb") as stream:
|
|
digest = hashlib.file_digest(stream, "sha256").hexdigest()
|
|
return {"wheel": wheel.name, "wheel_sha256": digest}
|
|
|
|
|
|
def main() -> None:
|
|
"""在构建前校验标签，在构建后补充 wheel 身份，两个阶段都锁定相同源码提交。"""
|
|
parser = argparse.ArgumentParser(description=__doc__)
|
|
parser.add_argument("--tag", default="")
|
|
parser.add_argument("--commit", required=True)
|
|
parser.add_argument("--wheel-dir", type=Path)
|
|
parser.add_argument("--output", type=Path)
|
|
args = parser.parse_args()
|
|
root = Path.cwd()
|
|
commit = subprocess.check_output(["git", "rev-parse", "HEAD"], text=True).strip()
|
|
if not re.fullmatch(r"[0-9a-f]{40}", args.commit) or commit != args.commit:
|
|
raise ValueError("Checkout must match the workflow commit SHA")
|
|
subprocess.run(["git", "diff", "--exit-code", "HEAD", "--", "mineru", "pyproject.toml"], check=True)
|
|
version = source_version(root, args.tag)
|
|
record = {"source_commit": commit, "tag": args.tag, "version": str(version)}
|
|
if args.wheel_dir is not None:
|
|
record.update(verify_wheel(root, args.wheel_dir, version))
|
|
result = json.dumps(record, indent=2)
|
|
if args.output is not None:
|
|
args.output.parent.mkdir(parents=True, exist_ok=True)
|
|
args.output.write_text(result + "\n", encoding="utf-8")
|
|
print(result)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|