feat(repro-audit): 新增 --repos-file 批量审计模式与确定性汇总排名，单测 10/10

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

examples/research/repro-audit/README.md

@@ -37,6 +37,10 @@ gitlink-cli auth login
 
 python3 scripts/repro_audit.py --owner <owner> --repo <repo> --output-dir outputs
 # 回写改进 tracking issue（请先确认报告内容）：加 --apply
+
+# 批量审计（实验室/课题组场景）：清单每行 owner/repo，# 为注释
+python3 scripts/repro_audit.py --repos-file repos.txt --output-dir outputs
+# 输出逐仓库报告 + repro-audit-summary.md 汇总排名（得分降序，退出码 2 表示存在 <70 分仓库）
 ```
 
 ## 已在真实科研仓库验证

examples/research/repro-audit/docs/application-report.md

@@ -16,7 +16,7 @@
 | 使用者 | 场景 | 价值 |
 |--------|------|------|
 | 课题组 | 论文投稿/开源发布前自查 | 逐项补齐复现要件，提升论文可信度 |
-| 实验室管理者 | 批量审计组内科研仓库 | 统一学术规范（许可证/引用/数据说明） |
+| 实验室管理者 | 批量审计组内科研仓库（`--repos-file` 清单模式，输出得分排名汇总表） | 统一学术规范（许可证/引用/数据说明） |
 | 期刊/会议 artifact 评审 | 快速初筛 | 评分卡作为客观初审依据 |
 | CI 门禁 | 科研仓库发布流程 | 退出码 2 阻断复现缺口明显的发布 |
 

examples/research/repro-audit/scripts/repro_audit.py

@@ -164,40 +164,89 @@ def render_report(owner, repo, ref, results):
     return "\n".join(lines), total
 
 
+def audit_repo(owner, repo, ref, out_dir, cli, apply_issue=False, echo=True):
+    """审计单个仓库，落盘报告，返回总分。"""
+    entries = list_entries(owner, repo, ref, cli=cli)
+    names = [n for n, t in entries if t != "dir"]
+    dirs = [n for n, t in entries if t == "dir"]
+    _, readme_text = fetch_readme(owner, repo, ref, names, cli=cli)
+    releases = count_releases(owner, repo, cli=cli)
+
+    results = audit(names, dirs, readme_text, releases)
+    report, total = render_report(owner, repo, ref, results)
+
+    path = out_dir / f"repro-audit-{owner}-{repo}.md"
+    path.write_text(report, encoding="utf-8")
+    if echo:
+        print(report)
+    print(f"\n报告已保存：{path}", file=sys.stderr)
+
+    if apply_issue:
+        run_cli([
+            "issue", "+create", "--owner", owner, "--repo", repo,
+            "--title", f"[repro-audit] 复现性审计报告（{total}/100）",
+            "--body", report, "--format", "json",
+        ], cli=cli)
+        print("已创建 tracking issue。", file=sys.stderr)
+    return total
+
+
+def read_repos_file(path):
+    """读取批量仓库清单：每行 owner/repo，# 开头为注释。"""
+    repos = []
+    for line in Path(path).read_text(encoding="utf-8").splitlines():
+        line = line.strip()
+        if not line or line.startswith("#"):
+            continue
+        owner, _, repo = line.partition("/")
+        if not owner or not repo:
+            raise ValueError(f"无效的仓库行（应为 owner/repo）：{line}")
+        repos.append((owner, repo))
+    return repos
+
+
+def render_summary(rows):
+    """批量审计汇总表（确定性：按得分降序、同分按名称）。"""
+    rows = sorted(rows, key=lambda r: (-r[1], r[0]))
+    lines = ["# 批量复现性审计汇总", "", "| 仓库 | 得分 | 等级 |", "|------|------|------|"]
+    for name, total in rows:
+        grade = next(g for t, g in GRADE if total >= t)
+        lines.append(f"| {name} | {total}/100 | {grade} |")
+    return "\n".join(lines) + "\n"
+
+
 def main():
     parser = argparse.ArgumentParser(description="科研项目复现性审计")
-    parser.add_argument("--owner", required=True)
-    parser.add_argument("--repo", required=True)
+    parser.add_argument("--owner")
+    parser.add_argument("--repo")
+    parser.add_argument("--repos-file", help="批量审计清单文件（每行 owner/repo，# 注释）")
     parser.add_argument("--ref", default="")
     parser.add_argument("--apply", action="store_true", help="把报告作为 tracking issue 回写（默认 dry-run）")
     parser.add_argument("--output-dir", default="outputs")
     parser.add_argument("--cli", default="gitlink-cli")
     args = parser.parse_args()
 
-    entries = list_entries(args.owner, args.repo, args.ref, cli=args.cli)
-    names = [n for n, t in entries if t != "dir"]
-    dirs = [n for n, t in entries if t == "dir"]
-    _, readme_text = fetch_readme(args.owner, args.repo, args.ref, names, cli=args.cli)
-    releases = count_releases(args.owner, args.repo, cli=args.cli)
-
-    results = audit(names, dirs, readme_text, releases)
-    report, total = render_report(args.owner, args.repo, args.ref, results)
+    if not args.repos_file and not (args.owner and args.repo):
+        parser.error("需要 --owner 与 --repo，或 --repos-file")
 
     out_dir = Path(args.output_dir)
     out_dir.mkdir(parents=True, exist_ok=True)
-    path = out_dir / f"repro-audit-{args.owner}-{args.repo}.md"
-    path.write_text(report, encoding="utf-8")
-    print(report)
-    print(f"\n报告已保存：{path}", file=sys.stderr)
 
-    if args.apply:
-        run_cli([
-            "issue", "+create", "--owner", args.owner, "--repo", args.repo,
-            "--title", f"[repro-audit] 复现性审计报告（{total}/100）",
-            "--body", report, "--format", "json",
-        ], cli=args.cli)
-        print("已创建 tracking issue。", file=sys.stderr)
+    if args.repos_file:
+        rows = []
+        for owner, repo in read_repos_file(args.repos_file):
+            total = audit_repo(owner, repo, args.ref, out_dir, args.cli,
+                               apply_issue=args.apply, echo=False)
+            rows.append((f"{owner}/{repo}", total))
+        summary = render_summary(rows)
+        summary_path = out_dir / "repro-audit-summary.md"
+        summary_path.write_text(summary, encoding="utf-8")
+        print(summary)
+        print(f"汇总已保存：{summary_path}", file=sys.stderr)
+        return 0 if all(t >= 70 for _, t in rows) else 2
 
+    total = audit_repo(args.owner, args.repo, args.ref, out_dir, args.cli,
+                       apply_issue=args.apply)
     return 0 if total >= 70 else 2
 
 

examples/research/repro-audit/tests/test_audit.py

@@ -1,12 +1,13 @@
 """确定性回归护栏：同输入 → 同分 → 同等级。"""
 
 import sys
+import tempfile
 import unittest
 from pathlib import Path
 
 sys.path.insert(0, str(Path(__file__).resolve().parent.parent / "scripts"))
 
-from repro_audit import audit, render_report  # noqa: E402
+from repro_audit import audit, read_repos_file, render_report, render_summary  # noqa: E402
 
 GOOD_README = """# Project
 
@@ -68,6 +69,26 @@ class AuditTest(unittest.TestCase):
         for r in results:
             self.assertTrue(r.advice, f"{r.name} 应给出修复建议")
 
+    def test_read_repos_file(self):
+        with tempfile.NamedTemporaryFile("w", suffix=".txt", delete=False, encoding="utf-8") as f:
+            f.write("# 注释\n\nowner1/repo1\n  owner2/repo2  \n")
+            path = f.name
+        self.assertEqual(read_repos_file(path), [("owner1", "repo1"), ("owner2", "repo2")])
+
+    def test_read_repos_file_invalid_line(self):
+        with tempfile.NamedTemporaryFile("w", suffix=".txt", delete=False, encoding="utf-8") as f:
+            f.write("not-a-repo-line\n")
+            path = f.name
+        with self.assertRaises(ValueError):
+            read_repos_file(path)
+
+    def test_render_summary_sorted(self):
+        summary = render_summary([("o/low", 8), ("o/high", 92), ("o/mid", 58)])
+        rows = [line for line in summary.splitlines() if line.startswith("| o/")]
+        self.assertEqual([r.split(" | ")[0] for r in rows], ["| o/high", "| o/mid", "| o/low"])
+        self.assertIn("A（可复现性良好）", rows[0])
+        self.assertIn("D（复现困难）", rows[2])
+
 
 if __name__ == "__main__":
     unittest.main()