beam: add manifest_check tool with tests

wizard-work/beam/manifest-check/manifest_check.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python3
+"""manifest_check: build and verify a sha256 file manifest for a directory."""
+from __future__ import annotations
+import argparse, hashlib, json, os, sys
+from pathlib import Path
+
+def sha256_file(p: Path) -> str:
+    h = hashlib.sha256()
+    with p.open("rb") as f:
+        for chunk in iter(lambda: f.read(65536), b""):
+            h.update(chunk)
+    return h.hexdigest()
+
+def build(root: Path) -> dict:
+    entries = []
+    for p in sorted(root.rglob("*")):
+        if p.is_file():
+            rel = p.relative_to(root).as_posix()
+            entries.append({"path": rel, "sha256": sha256_file(p), "size": p.stat().st_size})
+    return {"root": str(root), "count": len(entries), "entries": entries}
+
+def verify(manifest: dict, root: Path) -> tuple[bool, list[str]]:
+    errors = []
+    # expected files
+    expected = {e["path"]: e for e in manifest["entries"]}
+    actual_files = set()
+    for p in root.rglob("*"):
+        if p.is_file():
+            actual_files.add(p.relative_to(root).as_posix())
+    # missing
+    for path in expected:
+        if path not in actual_files:
+            errors.append(f"missing: {path}")
+    # extra
+    for path in actual_files:
+        if path not in expected:
+            errors.append(f"extra: {path}")
+    # content
+    for path, e in expected.items():
+        if path in actual_files:
+            p = root / path
+            actual_hash = sha256_file(p)
+            if actual_hash != e["sha256"]:
+                errors.append(f"changed: {path} (expected {e['sha256']}, got {actual_hash})")
+            if p.stat().st_size != e["size"]:
+                errors.append(f"size mismatch: {path}")
+    if manifest.get("count") != len(expected):
+        errors.append(f"count field {manifest['count']} != {len(expected)}")
+    return (len(errors) == 0, errors)
+
+def main():
+    ap = argparse.ArgumentParser()
+    sub = ap.add_subparsers(dest="cmd", required=True)
+    b = sub.add_parser("build")
+    b.add_argument("root")
+    b.add_argument("-o", "--output", required=True)
+    v = sub.add_parser("verify")
+    v.add_argument("manifest")
+    v.add_argument("root")
+    a = ap.parse_args()
+    if a.cmd == "build":
+        root = Path(a.root)
+        m = build(root)
+        Path(a.output).write_text(json.dumps(m, indent=2) + "\n")
+        print(f"built manifest: {m['count']} files -> {a.output}")
+        return 0
+    else:
+        manifest = json.loads(Path(a.manifest).read_text())
+        ok, errors = verify(manifest, Path(a.root))
+        if ok:
+            print(f"OK: {manifest['count']} files verified")
+            return 0
+        for e in errors:
+            print("FAIL:", e)
+        return 1
+
+if __name__ == "__main__":
+    sys.exit(main())

wizard-work/beam/manifest-check/test_manifest_check.py

@@ -0,0 +1,15 @@
+import subprocess, json, tempfile, os, pathlib, sys
+base = pathlib.Path("/root/troupe-standup/wizard-work/beam/manifest-check/manifest_check.py")
+d = pathlib.Path(tempfile.mkdtemp())
+(d/"a.txt").write_text("hello")
+(d/"sub").mkdir()
+(d/"sub/b.txt").write_text("world")
+m = pathlib.Path(tempfile.mkdtemp())/"manifest.json"
+subprocess.run([sys.executable, str(base), "build", str(d), "-o", str(m)], check=True)
+r = subprocess.run([sys.executable, str(base), "verify", str(m), str(d)])
+assert r.returncode == 0
+# tamper
+(d/"a.txt").write_text("tampered")
+r2 = subprocess.run([sys.executable, str(base), "verify", str(m), str(d)])
+assert r2.returncode == 1
+print("all tests passed")