Files
Agent-System-Dev/scripts/check-worktree.py
T
weichen b84667a697 Close review gaps in cleanup and evidence handling
Why: independent review found shared submodule metadata risk, quoted header leakage, stale JSON receipt blindness and missing abandon coverage.

Evidence: SELF 92 governance regressions pass; the original review receipt and replay output are recorded in the task card.

Impact: preserves external component worktrees, handles both evidence formats and adds targeted failure-path coverage; no product behavior changes.

Refs: .task/feature-governance-reliability.md; .task/reviews/governance-reliability.json
2026-10-05 13:07:10 +00:00

149 lines
7.0 KiB
Python

#!/usr/bin/env python3
"""Fail closed before task synchronization, verification, or worktree disposal."""
import os
from pathlib import Path
import subprocess
import sys
def git(repo, *args):
result = subprocess.run(["git", "-C", str(repo), *args], stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
if result.returncode:
raise RuntimeError(f"Git check failed in {repo}: {' '.join(args)}")
return result.stdout
def text(repo, *args):
return os.fsdecode(git(repo, *args)).strip()
def check_operations(repo):
for marker in ("index.lock", "HEAD.lock", "MERGE_HEAD", "CHERRY_PICK_HEAD",
"REVERT_HEAD", "rebase-merge", "rebase-apply", "BISECT_LOG"):
path = Path(text(repo, "rev-parse", "--path-format=absolute", "--git-path", marker))
if path.exists():
raise RuntimeError(f"active Git operation/lock in {repo}: {marker}; owner must finish it")
def check_clean(repo, ignored=False, allow_gitlink_drift=False):
args = ["status", "--porcelain=v1", "--untracked-files=all",
"--ignore-submodules=all" if allow_gitlink_drift else "--ignore-submodules=none"]
if ignored:
args.append("--ignored=matching")
status = git(repo, *args)
if status:
raise RuntimeError(f"unsaved, untracked{', or ignored' if ignored else ''} files in {repo}; preserve them before continuing")
def submodules(repo):
for item in git(repo, "ls-files", "--stage", "-z").split(b"\0"):
if not item:
continue
meta, path = item.split(b"\t", 1)
mode, sha, stage = meta.split()
if stage != b"0":
raise RuntimeError(f"unmerged index in {repo}; resolve conflicts first")
if mode == b"160000":
yield repo / os.fsdecode(path), sha.decode("ascii")
def worktrees(repo):
records = git(repo, "worktree", "list", "--porcelain", "-z").split(b"\0\0")
parsed = [record.split(b"\0") for record in records if record]
if not parsed or any(not record[0].startswith(b"worktree ") for record in parsed):
raise RuntimeError(f"cannot identify registered worktrees in {repo}")
return parsed
def check_component_worktrees(repo, removal_roots):
common = Path(text(repo, "rev-parse", "--path-format=absolute", "--git-common-dir")).resolve()
if not any(common == root or root in common.parents for root in removal_roots):
return
current = Path(text(repo, "rev-parse", "--absolute-git-dir")).resolve()
for index, record in enumerate(worktrees(repo)):
other = Path(os.fsdecode(record[0][9:])).resolve()
# Git may report an absorbed submodule's primary checkout as its common
# directory. It is the current checkout only when no linked gitdir is used.
if other != repo.resolve() and not (index == 0 and current == common):
raise RuntimeError(f"registered component worktree {other} depends on metadata in {common}; preserve or relocate it before removal")
def check_children(repo, mode, removal_roots=()):
for child, expected in submodules(repo):
if child.is_symlink():
raise RuntimeError(f"submodule path is a symbolic link: {child}")
if not (child / ".git").exists():
if mode in ("verify", "after-sync", "finish"):
raise RuntimeError(f"uninitialized submodule: {child}; initialize and inspect it first")
continue
if mode == "finish":
check_component_worktrees(child, removal_roots)
actual = text(child, "rev-parse", "HEAD")
check_operations(child)
check_clean(child, ignored=(mode == "finish"), allow_gitlink_drift=(mode == "before-update"))
branch = text(child, "branch", "--show-current")
if mode == "before-update" and branch and actual != expected:
raise RuntimeError(f"submodule {child} owns branch {branch}; keep that work and explicitly reconcile pin {expected} before retrying task-sync")
if mode in ("verify", "after-sync", "finish") and actual != expected:
raise RuntimeError(f"submodule SHA drift: {child}: expected {expected}, actual {actual}")
if mode == "finish" and git(child, "rev-list", "--all", "--reflog", "--not", expected).strip():
raise RuntimeError(f"submodule has commits outside its pinned history: {child}; preserve branches, stash and reflogs before removal")
check_children(child, mode, removal_roots)
def materialize(repo):
"""Check out one level, then inspect the newly exposed nested pins before moving them."""
for child, expected in submodules(repo):
check_children(repo, "before-update")
result = subprocess.run(["git", "-C", str(repo), "submodule", "update", "--init",
"--checkout", "--", str(child.relative_to(repo))])
if result.returncode:
raise RuntimeError(f"component materialization failed: {child}; preserve and reconcile before retrying")
materialize(child)
def check_finish_root(repo):
if Path.cwd().resolve() != repo:
raise RuntimeError("task-finish must run at the registered worktree root, not a subdirectory")
parsed = worktrees(repo)
main = Path(os.fsdecode(parsed[0][0][9:])).resolve()
if main == repo:
raise RuntimeError("task-finish must run in a linked task worktree, never the main checkout")
matches = [r for r in parsed if r[0].startswith(b"worktree ")
and Path(os.fsdecode(r[0][9:])).resolve() == repo]
if len(matches) != 1:
raise RuntimeError("worktree is not registered exactly once; preserve it for owner inspection")
if any(field.startswith((b"locked", b"prunable", b"bare", b"detached")) for field in matches[0][1:]):
raise RuntimeError("locked, detached, bare, or prunable worktree must be preserved")
common = Path(text(repo, "rev-parse", "--path-format=absolute", "--git-common-dir"))
if any(common.rglob("*.lock")):
raise RuntimeError("Git lock exists in shared repository metadata; wait for its owner")
def main():
mode = sys.argv[1] if len(sys.argv) > 1 else "verify"
if mode not in ("verify", "before-sync", "before-update", "materialize", "after-sync", "finish"):
raise RuntimeError("usage: check-worktree.py [verify|before-sync|before-update|materialize|after-sync|finish]")
repo = Path(text(Path.cwd(), "rev-parse", "--show-toplevel")).resolve()
check_operations(repo)
removal_roots = ()
if mode == "finish":
check_finish_root(repo)
removal_roots = (repo, Path(text(repo, "rev-parse", "--absolute-git-dir")).resolve())
if mode in ("before-sync", "after-sync", "finish"):
check_clean(repo, ignored=(mode == "finish"))
if mode == "materialize":
materialize(repo)
else:
check_children(repo, mode, removal_roots)
print(f"[worktree] {mode}: recursive component checks passed")
if __name__ == "__main__":
try:
main()
except RuntimeError as exc:
print(f"ERROR: {exc}", file=sys.stderr)
sys.exit(1)