test_cmd_snapshot_hardening.py
python
sha256:84df9126d09aeec0b8f1b908f0b06c10913feec28f3514b382efb1ba6d619385
refactor: rename StructuredMergePlugin to AddressedMergePlu…
Sonnet 4.6
minor
⚠ breaking
23 days ago
| 1 | """Hardening tests for ``muse snapshot`` — security, performance, agent UX. |
| 2 | |
| 3 | Covers: |
| 4 | Unit (helpers): |
| 5 | - _safe_arcname: absolute path rejected, .. segments rejected, |
| 6 | prefix .. rejected, valid paths accepted |
| 7 | - _validate_snapshot_id_prefix: strips non-hex chars, caps at 64 |
| 8 | - _list_all_snapshots: symlink skipped |
| 9 | - _resolve_snapshot: prefix scan skips symlinks, full ID hit |
| 10 | |
| 11 | Unit (SnapshotRecord): |
| 12 | - note field persists through to_dict / from_dict round-trip |
| 13 | - note defaults to "" for old records without the field |
| 14 | |
| 15 | Security: |
| 16 | - Symlink inside .muse/snapshots/ skipped during list |
| 17 | - Symlink inside .muse/snapshots/ skipped during show prefix scan |
| 18 | - Symlink inside .muse/snapshots/ skipped during export prefix scan |
| 19 | - ANSI in note sanitized in text output, raw in JSON |
| 20 | - ANSI in rel_path sanitized in show --text output |
| 21 | - Broken --json shorthand on export no longer accepted (was broken bug) |
| 22 | |
| 23 | Error routing: |
| 24 | - snapshot read not-found goes to stderr |
| 25 | - snapshot export not-found goes to stderr |
| 26 | |
| 27 | JSON schema (create): |
| 28 | - All _SnapshotCreateJson fields present: repo_id, snapshot_id, |
| 29 | file_count, note, created_at |
| 30 | - note persisted and returned in JSON |
| 31 | |
| 32 | JSON schema (list): |
| 33 | - _SnapshotListItemJson fields: snapshot_id, file_count, note, created_at |
| 34 | - note round-trips through create → list |
| 35 | |
| 36 | JSON schema (show): |
| 37 | - _SnapshotReadJson fields: snapshot_id, created_at, file_count, |
| 38 | note, manifest |
| 39 | - show default is JSON (no flag needed) |
| 40 | - --text flag emits human-readable text |
| 41 | |
| 42 | JSON schema (export): |
| 43 | - _SnapshotExportJson fields: snapshot_id, output, format, |
| 44 | file_count, size_bytes |
| 45 | - size_bytes > 0 for non-empty archive |
| 46 | - format field matches archive type |
| 47 | |
| 48 | New features: |
| 49 | - note persisted in SnapshotRecord (not ephemeral) |
| 50 | - note shown in snapshot list text output |
| 51 | - note shown in snapshot read text output |
| 52 | - Old --format json / -f json flags rejected (clean migration) |
| 53 | |
| 54 | Integration: |
| 55 | - create → list → show → export pipeline (tar.gz + zip) |
| 56 | - Prefix scan resolves short ID in show and export |
| 57 | - Multiple snapshots sorted newest-first in list |
| 58 | - export --json + tar.gz produces valid archive AND JSON summary |
| 59 | |
| 60 | E2E: |
| 61 | - --help shows --json for create, list, export |
| 62 | - --help shows --text for show |
| 63 | - snapshot read --help describes default-JSON behaviour |
| 64 | |
| 65 | Stress: |
| 66 | - 200 snapshots list correctly |
| 67 | - 500-file snapshot create + show manifest integrity |
| 68 | - Concurrent create (5 threads) |
| 69 | - Concurrent list (10 threads) |
| 70 | """ |
| 71 | |
| 72 | from __future__ import annotations |
| 73 | |
| 74 | import hashlib |
| 75 | import json |
| 76 | import pathlib |
| 77 | import tarfile |
| 78 | import threading |
| 79 | import zipfile |
| 80 | |
| 81 | import pytest |
| 82 | from tests.cli_test_helper import CliRunner, InvokeResult |
| 83 | |
| 84 | from muse.core.object_store import object_path, write_object |
| 85 | from muse.core.ids import hash_snapshot |
| 86 | from muse.core.types import MsgpackValue |
| 87 | from muse.core.snapshots import ( |
| 88 | SnapshotRecord, |
| 89 | write_snapshot, |
| 90 | ) |
| 91 | from muse.cli.commands.snapshot_cmd import ( |
| 92 | _list_all_snapshots, |
| 93 | _resolve_snapshot, |
| 94 | _safe_arcname, |
| 95 | _validate_snapshot_id_prefix, |
| 96 | ) |
| 97 | from muse.core.types import Manifest, MsgpackDict, blob_id, split_id, short_id |
| 98 | from muse.core.paths import muse_dir |
| 99 | |
| 100 | runner = CliRunner() |
| 101 | cli = None # argparse migration — CliRunner ignores this arg |
| 102 | |
| 103 | _REPO_ID = "snapshot-hardening-test" |
| 104 | |
| 105 | |
| 106 | # --------------------------------------------------------------------------- |
| 107 | # TypedDicts for parsing JSON |
| 108 | # --------------------------------------------------------------------------- |
| 109 | |
| 110 | from typing import TypedDict |
| 111 | |
| 112 | |
| 113 | class _CreateOut(TypedDict): |
| 114 | repo_id: str |
| 115 | snapshot_id: str |
| 116 | file_count: int |
| 117 | note: str |
| 118 | created_at: str |
| 119 | |
| 120 | |
| 121 | class _ListItemOut(TypedDict): |
| 122 | snapshot_id: str |
| 123 | file_count: int |
| 124 | note: str |
| 125 | created_at: str |
| 126 | |
| 127 | |
| 128 | class _ReadOut(TypedDict): |
| 129 | snapshot_id: str |
| 130 | created_at: str |
| 131 | file_count: int |
| 132 | note: str |
| 133 | manifest: Manifest |
| 134 | |
| 135 | |
| 136 | class _ExportOut(TypedDict): |
| 137 | snapshot_id: str |
| 138 | output: str |
| 139 | format: str |
| 140 | file_count: int |
| 141 | size_bytes: int |
| 142 | |
| 143 | |
| 144 | # --------------------------------------------------------------------------- |
| 145 | # Helpers |
| 146 | # --------------------------------------------------------------------------- |
| 147 | |
| 148 | _invoke_lock = threading.Lock() |
| 149 | |
| 150 | |
| 151 | |
| 152 | |
| 153 | def _init_repo(path: pathlib.Path) -> pathlib.Path: |
| 154 | muse = muse_dir(path) |
| 155 | for d in ("commits", "snapshots", "objects", "refs/heads"): |
| 156 | (muse / d).mkdir(parents=True, exist_ok=True) |
| 157 | (muse / "HEAD").write_text("ref: refs/heads/main", encoding="utf-8") |
| 158 | (muse / "repo.json").write_text( |
| 159 | json.dumps({"repo_id": _REPO_ID, "domain": "code"}), encoding="utf-8" |
| 160 | ) |
| 161 | return path |
| 162 | |
| 163 | |
| 164 | def _env(repo: pathlib.Path) -> Manifest: |
| 165 | return {"MUSE_REPO_ROOT": str(repo)} |
| 166 | |
| 167 | |
| 168 | def _create_files(root: pathlib.Path, count: int = 3) -> list[str]: |
| 169 | names: list[str] = [] |
| 170 | for i in range(count): |
| 171 | name = f"file_{i}.txt" |
| 172 | (root / name).write_text(f"content {i}", encoding="utf-8") |
| 173 | names.append(name) |
| 174 | return names |
| 175 | |
| 176 | |
| 177 | def _invoke(args: list[str], env: Manifest) -> InvokeResult: |
| 178 | with _invoke_lock: |
| 179 | return runner.invoke(cli, args, env=env) |
| 180 | |
| 181 | |
| 182 | def _write_snapshot(root: pathlib.Path, note: str = "", n_files: int = 1) -> str: |
| 183 | """Create and store a snapshot record directly; return the snapshot_id.""" |
| 184 | manifest: Manifest = {} |
| 185 | for i in range(n_files): |
| 186 | data = f"object-{i}-{note}".encode() |
| 187 | obj_id = blob_id(data) |
| 188 | write_object(root, obj_id, data) |
| 189 | manifest[f"file_{i}.txt"] = obj_id |
| 190 | snap_id = hash_snapshot(manifest) |
| 191 | write_snapshot(root, SnapshotRecord(snapshot_id=snap_id, manifest=manifest, note=note)) |
| 192 | return snap_id |
| 193 | |
| 194 | |
| 195 | # --------------------------------------------------------------------------- |
| 196 | # Unit: _safe_arcname |
| 197 | # --------------------------------------------------------------------------- |
| 198 | |
| 199 | |
| 200 | def test_safe_arcname_rejects_absolute() -> None: |
| 201 | assert _safe_arcname("prefix", "/etc/passwd") is None |
| 202 | |
| 203 | |
| 204 | def test_safe_arcname_rejects_dotdot_in_rel() -> None: |
| 205 | assert _safe_arcname("prefix", "../traversal.txt") is None |
| 206 | |
| 207 | |
| 208 | def test_safe_arcname_rejects_dotdot_in_prefix() -> None: |
| 209 | assert _safe_arcname("../traversal", "file.txt") is None |
| 210 | |
| 211 | |
| 212 | def test_safe_arcname_valid_no_prefix() -> None: |
| 213 | result = _safe_arcname("", "path/to/file.txt") |
| 214 | assert result == "path/to/file.txt" |
| 215 | |
| 216 | |
| 217 | def test_safe_arcname_valid_with_prefix() -> None: |
| 218 | result = _safe_arcname("myproject", "path/to/file.txt") |
| 219 | assert result == "myproject/path/to/file.txt" |
| 220 | |
| 221 | |
| 222 | def test_safe_arcname_strips_trailing_slash_from_prefix() -> None: |
| 223 | result = _safe_arcname("myproject/", "file.txt") |
| 224 | assert result == "myproject/file.txt" |
| 225 | |
| 226 | |
| 227 | # --------------------------------------------------------------------------- |
| 228 | # Unit: _validate_snapshot_id_prefix |
| 229 | # --------------------------------------------------------------------------- |
| 230 | |
| 231 | |
| 232 | def test_validate_snapshot_id_prefix_strips_non_hex() -> None: |
| 233 | result = _validate_snapshot_id_prefix("abc123xyz!@#$") |
| 234 | assert result == "abc123" |
| 235 | |
| 236 | |
| 237 | def test_validate_snapshot_id_prefix_caps_at_64() -> None: |
| 238 | long_hex = "a" * 100 |
| 239 | result = _validate_snapshot_id_prefix(long_hex) |
| 240 | assert len(result) == 64 |
| 241 | |
| 242 | |
| 243 | def test_validate_snapshot_id_prefix_empty_input() -> None: |
| 244 | result = _validate_snapshot_id_prefix("") |
| 245 | assert result == "" |
| 246 | |
| 247 | |
| 248 | # --------------------------------------------------------------------------- |
| 249 | # Unit: _list_all_snapshots symlink guard |
| 250 | # --------------------------------------------------------------------------- |
| 251 | |
| 252 | |
| 253 | def test_list_all_snapshots_skips_symlink(tmp_path: pathlib.Path) -> None: |
| 254 | from muse.core.paths import objects_dir as _objects_dir |
| 255 | _init_repo(tmp_path) |
| 256 | snap_id = _write_snapshot(tmp_path) |
| 257 | # Plant a symlink inside the object store — iter_stored_objects must skip it. |
| 258 | objs_dir = _objects_dir(tmp_path) |
| 259 | shard_dir = objs_dir / "sha256" / "de" |
| 260 | shard_dir.mkdir(parents=True, exist_ok=True) |
| 261 | target = tmp_path / "malicious.txt" |
| 262 | target.write_bytes(b"not a snapshot") |
| 263 | link = shard_dir / ("ad" + "0" * 60) |
| 264 | try: |
| 265 | link.symlink_to(target) |
| 266 | except (OSError, NotImplementedError): |
| 267 | pytest.skip("symlinks not supported on this platform") |
| 268 | results = _list_all_snapshots(tmp_path) |
| 269 | snap_ids = [r.snapshot_id for r in results] |
| 270 | # Only the legitimately written snapshot must appear. |
| 271 | assert snap_ids == [snap_id] |
| 272 | |
| 273 | |
| 274 | def test_list_all_snapshots_returns_real_records(tmp_path: pathlib.Path) -> None: |
| 275 | _init_repo(tmp_path) |
| 276 | _write_snapshot(tmp_path, note="a") |
| 277 | _write_snapshot(tmp_path, note="b", n_files=2) |
| 278 | results = _list_all_snapshots(tmp_path) |
| 279 | assert len(results) == 2 |
| 280 | |
| 281 | |
| 282 | # --------------------------------------------------------------------------- |
| 283 | # Unit: _resolve_snapshot prefix scan skips symlinks |
| 284 | # --------------------------------------------------------------------------- |
| 285 | |
| 286 | |
| 287 | def test_resolve_snapshot_prefix_skips_symlink(tmp_path: pathlib.Path) -> None: |
| 288 | from muse.core.paths import objects_dir as _objects_dir |
| 289 | _init_repo(tmp_path) |
| 290 | snap_id = _write_snapshot(tmp_path) |
| 291 | # Plant a symlink with prefix "aaaa" inside the object store. |
| 292 | objs_dir = _objects_dir(tmp_path) |
| 293 | shard_dir = objs_dir / "sha256" / "aa" |
| 294 | shard_dir.mkdir(parents=True, exist_ok=True) |
| 295 | target = tmp_path / "rogue.txt" |
| 296 | target.write_bytes(b"not a snapshot") |
| 297 | link = shard_dir / ("aa" + "0" * 60) |
| 298 | try: |
| 299 | link.symlink_to(target) |
| 300 | except (OSError, NotImplementedError): |
| 301 | pytest.skip("symlinks not supported on this platform") |
| 302 | # The symlink has a hex prefix "aaaa…" — resolving that prefix must skip it. |
| 303 | resolved = _resolve_snapshot(tmp_path, "aaaa") |
| 304 | # "aaaa" is not a hex prefix of the real snap_id — so should be None. |
| 305 | assert resolved is None or resolved.snapshot_id == snap_id |
| 306 | |
| 307 | |
| 308 | def test_resolve_snapshot_full_id_hit(tmp_path: pathlib.Path) -> None: |
| 309 | _init_repo(tmp_path) |
| 310 | snap_id = _write_snapshot(tmp_path, note="full hit") |
| 311 | resolved = _resolve_snapshot(tmp_path, snap_id) |
| 312 | assert resolved is not None |
| 313 | assert resolved.snapshot_id == snap_id |
| 314 | |
| 315 | |
| 316 | def test_resolve_snapshot_prefix_hit(tmp_path: pathlib.Path) -> None: |
| 317 | _init_repo(tmp_path) |
| 318 | snap_id = _write_snapshot(tmp_path, note="prefix hit") |
| 319 | resolved = _resolve_snapshot(tmp_path, short_id(snap_id)) |
| 320 | assert resolved is not None |
| 321 | assert resolved.snapshot_id == snap_id |
| 322 | |
| 323 | |
| 324 | def test_resolve_snapshot_miss(tmp_path: pathlib.Path) -> None: |
| 325 | _init_repo(tmp_path) |
| 326 | resolved = _resolve_snapshot(tmp_path, "0000000000000000000000000000000000000000000000000000000000000000") |
| 327 | assert resolved is None |
| 328 | |
| 329 | |
| 330 | # --------------------------------------------------------------------------- |
| 331 | # Unit: SnapshotRecord note round-trip |
| 332 | # --------------------------------------------------------------------------- |
| 333 | |
| 334 | |
| 335 | def test_snapshot_record_note_round_trips_to_dict() -> None: |
| 336 | snap = SnapshotRecord(snapshot_id="a" * 64, manifest={}, note="my note") |
| 337 | d = snap.to_dict() |
| 338 | assert d["note"] == "my note" |
| 339 | |
| 340 | |
| 341 | def test_snapshot_record_note_round_trips_from_dict() -> None: |
| 342 | snap = SnapshotRecord(snapshot_id="b" * 64, manifest={}, note="restored") |
| 343 | d: MsgpackDict = { |
| 344 | "snapshot_id": snap.snapshot_id, |
| 345 | "manifest": {}, |
| 346 | "created_at": snap.created_at.isoformat(), |
| 347 | "note": snap.note, |
| 348 | } |
| 349 | restored = SnapshotRecord.from_dict(d) |
| 350 | assert restored.note == "restored" |
| 351 | |
| 352 | |
| 353 | def test_snapshot_record_note_defaults_empty_for_old_records() -> None: |
| 354 | d: MsgpackDict = { |
| 355 | "snapshot_id": "c" * 64, |
| 356 | "manifest": {}, |
| 357 | "created_at": "2026-01-01T00:00:00+00:00", |
| 358 | # no "note" key — simulates an old record |
| 359 | } |
| 360 | restored = SnapshotRecord.from_dict(d) |
| 361 | assert restored.note == "" |
| 362 | |
| 363 | |
| 364 | # --------------------------------------------------------------------------- |
| 365 | # Security: symlink guard in show + export |
| 366 | # --------------------------------------------------------------------------- |
| 367 | |
| 368 | |
| 369 | def test_snapshot_read_symlink_not_resolved(tmp_path: pathlib.Path) -> None: |
| 370 | """Prefix scan in show must skip symlinks in the object store.""" |
| 371 | from muse.core.paths import objects_dir as _objects_dir |
| 372 | _init_repo(tmp_path) |
| 373 | _write_snapshot(tmp_path) |
| 374 | objs_dir = _objects_dir(tmp_path) |
| 375 | shard_dir = objs_dir / "sha256" / "00" |
| 376 | shard_dir.mkdir(parents=True, exist_ok=True) |
| 377 | target = tmp_path / "some_file.txt" |
| 378 | target.write_bytes(b"not a snapshot") |
| 379 | link = shard_dir / ("00" + "0" * 60) |
| 380 | try: |
| 381 | link.symlink_to(target) |
| 382 | except (OSError, NotImplementedError): |
| 383 | pytest.skip("symlinks not supported on this platform") |
| 384 | result = _invoke(["snapshot", "read", "0000000000000000"], env=_env(tmp_path)) |
| 385 | # Symlink skipped → prefix "0000" not found → exit_code != 0 |
| 386 | assert result.exit_code != 0 |
| 387 | |
| 388 | |
| 389 | def test_snapshot_export_symlink_not_resolved(tmp_path: pathlib.Path) -> None: |
| 390 | """Prefix scan in export must skip symlinks in the object store.""" |
| 391 | from muse.core.paths import objects_dir as _objects_dir |
| 392 | _init_repo(tmp_path) |
| 393 | _write_snapshot(tmp_path) |
| 394 | objs_dir = _objects_dir(tmp_path) |
| 395 | shard_dir = objs_dir / "sha256" / "bb" |
| 396 | shard_dir.mkdir(parents=True, exist_ok=True) |
| 397 | target = tmp_path / "some_file.txt" |
| 398 | target.write_bytes(b"not a snapshot") |
| 399 | link = shard_dir / ("bb" + "0" * 60) |
| 400 | try: |
| 401 | link.symlink_to(target) |
| 402 | except (OSError, NotImplementedError): |
| 403 | pytest.skip("symlinks not supported on this platform") |
| 404 | out_file = tmp_path / "out.tar.gz" |
| 405 | result = _invoke( |
| 406 | ["snapshot", "export", "bbbbbbbbbbbbbbbb", "--output", str(out_file)], |
| 407 | env=_env(tmp_path), |
| 408 | ) |
| 409 | # Symlink skipped → prefix "bbbb" not found → exit_code != 0 |
| 410 | assert result.exit_code != 0 |
| 411 | |
| 412 | |
| 413 | # --------------------------------------------------------------------------- |
| 414 | # Security: ANSI injection |
| 415 | # --------------------------------------------------------------------------- |
| 416 | |
| 417 | |
| 418 | def test_ansi_in_note_sanitized_in_text_output(tmp_path: pathlib.Path) -> None: |
| 419 | _init_repo(tmp_path) |
| 420 | _create_files(tmp_path, 1) |
| 421 | malicious_note = "\x1b[31mRED\x1b[0m" |
| 422 | result = _invoke(["snapshot", "create", "-m", malicious_note], env=_env(tmp_path)) |
| 423 | assert result.exit_code == 0 |
| 424 | assert "\x1b[" not in result.output |
| 425 | |
| 426 | |
| 427 | def test_ansi_in_note_raw_in_json_output(tmp_path: pathlib.Path) -> None: |
| 428 | _init_repo(tmp_path) |
| 429 | _create_files(tmp_path, 1) |
| 430 | malicious_note = "\x1b[31mRED\x1b[0m" |
| 431 | result = _invoke(["snapshot", "create", "--json", "-m", malicious_note], env=_env(tmp_path)) |
| 432 | assert result.exit_code == 0 |
| 433 | data: _CreateOut = json.loads(result.output) |
| 434 | assert "\x1b[" in data["note"] # JSON preserves raw bytes |
| 435 | |
| 436 | |
| 437 | def test_ansi_in_note_sanitized_in_list_text(tmp_path: pathlib.Path) -> None: |
| 438 | _init_repo(tmp_path) |
| 439 | _create_files(tmp_path, 1) |
| 440 | malicious_note = "\x1b[31mDanger\x1b[0m" |
| 441 | _invoke(["snapshot", "create", "-m", malicious_note], env=_env(tmp_path)) |
| 442 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 443 | assert result.exit_code == 0 |
| 444 | assert "\x1b[" not in result.output |
| 445 | |
| 446 | |
| 447 | # --------------------------------------------------------------------------- |
| 448 | # Error routing |
| 449 | # --------------------------------------------------------------------------- |
| 450 | |
| 451 | |
| 452 | def test_snapshot_read_not_found_stderr(tmp_path: pathlib.Path) -> None: |
| 453 | _init_repo(tmp_path) |
| 454 | result = _invoke(["snapshot", "read", "doesnotexist"], env=_env(tmp_path)) |
| 455 | assert result.exit_code != 0 |
| 456 | |
| 457 | |
| 458 | def test_snapshot_export_not_found_stderr(tmp_path: pathlib.Path) -> None: |
| 459 | _init_repo(tmp_path) |
| 460 | result = _invoke( |
| 461 | ["snapshot", "export", "doesnotexist", "--output", "/tmp/x.tar.gz"], |
| 462 | env=_env(tmp_path), |
| 463 | ) |
| 464 | assert result.exit_code != 0 |
| 465 | |
| 466 | |
| 467 | def test_old_format_flag_rejected(tmp_path: pathlib.Path) -> None: |
| 468 | _init_repo(tmp_path) |
| 469 | result = _invoke(["snapshot", "create", "-f", "json"], env=_env(tmp_path)) |
| 470 | # -f is no longer a valid flag for create → argparse rejects it |
| 471 | assert result.exit_code != 0 |
| 472 | |
| 473 | |
| 474 | # --------------------------------------------------------------------------- |
| 475 | # JSON schema: create |
| 476 | # --------------------------------------------------------------------------- |
| 477 | |
| 478 | |
| 479 | def test_create_json_all_fields(tmp_path: pathlib.Path) -> None: |
| 480 | _init_repo(tmp_path) |
| 481 | _create_files(tmp_path, 2) |
| 482 | result = _invoke(["snapshot", "create", "--json", "-m", "hello"], env=_env(tmp_path)) |
| 483 | assert result.exit_code == 0 |
| 484 | data: _CreateOut = json.loads(result.output) |
| 485 | assert data["repo_id"] == _REPO_ID |
| 486 | assert len(data["snapshot_id"]) == 71 |
| 487 | assert data["file_count"] >= 2 |
| 488 | assert data["note"] == "hello" |
| 489 | assert "T" in data["created_at"] |
| 490 | |
| 491 | |
| 492 | def test_create_json_note_persisted(tmp_path: pathlib.Path) -> None: |
| 493 | _init_repo(tmp_path) |
| 494 | _create_files(tmp_path, 1) |
| 495 | result = _invoke(["snapshot", "create", "--json", "-m", "saved"], env=_env(tmp_path)) |
| 496 | data: _CreateOut = json.loads(result.output) |
| 497 | assert data["note"] == "saved" |
| 498 | |
| 499 | |
| 500 | def test_create_json_no_note_empty_string(tmp_path: pathlib.Path) -> None: |
| 501 | _init_repo(tmp_path) |
| 502 | _create_files(tmp_path, 1) |
| 503 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 504 | data: _CreateOut = json.loads(result.output) |
| 505 | assert data["note"] == "" |
| 506 | |
| 507 | |
| 508 | def test_create_json_repo_id_present(tmp_path: pathlib.Path) -> None: |
| 509 | _init_repo(tmp_path) |
| 510 | _create_files(tmp_path, 1) |
| 511 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 512 | data: _CreateOut = json.loads(result.output) |
| 513 | assert data["repo_id"] == _REPO_ID |
| 514 | |
| 515 | |
| 516 | # --------------------------------------------------------------------------- |
| 517 | # JSON schema: list |
| 518 | # --------------------------------------------------------------------------- |
| 519 | |
| 520 | |
| 521 | def test_list_json_item_has_note(tmp_path: pathlib.Path) -> None: |
| 522 | _init_repo(tmp_path) |
| 523 | _create_files(tmp_path, 1) |
| 524 | _invoke(["snapshot", "create", "-m", "list-note"], env=_env(tmp_path)) |
| 525 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 526 | assert result.exit_code == 0 |
| 527 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 528 | assert len(items) == 1 |
| 529 | assert items[0]["note"] == "list-note" |
| 530 | |
| 531 | |
| 532 | def test_list_json_all_fields(tmp_path: pathlib.Path) -> None: |
| 533 | _init_repo(tmp_path) |
| 534 | _create_files(tmp_path, 1) |
| 535 | _invoke(["snapshot", "create"], env=_env(tmp_path)) |
| 536 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 537 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 538 | assert "snapshot_id" in items[0] |
| 539 | assert "file_count" in items[0] |
| 540 | assert "note" in items[0] |
| 541 | assert "created_at" in items[0] |
| 542 | |
| 543 | |
| 544 | def test_list_empty_json_is_array(tmp_path: pathlib.Path) -> None: |
| 545 | _init_repo(tmp_path) |
| 546 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 547 | assert result.exit_code == 0 |
| 548 | data = json.loads(result.output) |
| 549 | assert data["snapshots"] == [] |
| 550 | |
| 551 | |
| 552 | def test_list_note_in_text_output(tmp_path: pathlib.Path) -> None: |
| 553 | _init_repo(tmp_path) |
| 554 | _create_files(tmp_path, 1) |
| 555 | _invoke(["snapshot", "create", "-m", "a-note"], env=_env(tmp_path)) |
| 556 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 557 | assert result.exit_code == 0 |
| 558 | assert "a-note" in result.output |
| 559 | |
| 560 | |
| 561 | # --------------------------------------------------------------------------- |
| 562 | # JSON schema: show |
| 563 | # --------------------------------------------------------------------------- |
| 564 | |
| 565 | |
| 566 | def test_read_default_is_json(tmp_path: pathlib.Path) -> None: |
| 567 | _init_repo(tmp_path) |
| 568 | _create_files(tmp_path, 2) |
| 569 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 570 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 571 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 572 | assert result.exit_code == 0 |
| 573 | data: _ReadOut = json.loads(result.output) |
| 574 | assert data["snapshot_id"] == snap_id |
| 575 | |
| 576 | |
| 577 | def test_read_json_all_fields(tmp_path: pathlib.Path) -> None: |
| 578 | _init_repo(tmp_path) |
| 579 | _create_files(tmp_path, 2) |
| 580 | create_res = _invoke(["snapshot", "create", "--json", "-m", "show-note"], env=_env(tmp_path)) |
| 581 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 582 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 583 | data: _ReadOut = json.loads(result.output) |
| 584 | assert data["snapshot_id"] == snap_id |
| 585 | assert data["file_count"] >= 2 |
| 586 | assert data["note"] == "show-note" |
| 587 | assert isinstance(data["manifest"], dict) |
| 588 | assert "created_at" in data |
| 589 | |
| 590 | |
| 591 | def test_read_text_flag(tmp_path: pathlib.Path) -> None: |
| 592 | _init_repo(tmp_path) |
| 593 | _create_files(tmp_path, 1) |
| 594 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 595 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 596 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 597 | assert result.exit_code == 0 |
| 598 | assert "snapshot_id:" in result.output |
| 599 | |
| 600 | |
| 601 | def test_read_text_note_displayed(tmp_path: pathlib.Path) -> None: |
| 602 | _init_repo(tmp_path) |
| 603 | _create_files(tmp_path, 1) |
| 604 | create_res = _invoke( |
| 605 | ["snapshot", "create", "--json", "-m", "text-note"], env=_env(tmp_path) |
| 606 | ) |
| 607 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 608 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 609 | assert result.exit_code == 0 |
| 610 | assert "text-note" in result.output |
| 611 | |
| 612 | |
| 613 | # --------------------------------------------------------------------------- |
| 614 | # JSON schema: export |
| 615 | # --------------------------------------------------------------------------- |
| 616 | |
| 617 | |
| 618 | def test_export_json_all_fields_tar(tmp_path: pathlib.Path) -> None: |
| 619 | _init_repo(tmp_path) |
| 620 | _create_files(tmp_path, 2) |
| 621 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 622 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 623 | out_file = tmp_path / "out.tar.gz" |
| 624 | result = _invoke( |
| 625 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 626 | env=_env(tmp_path), |
| 627 | ) |
| 628 | assert result.exit_code == 0 |
| 629 | data: _ExportOut = json.loads(result.output) |
| 630 | assert data["snapshot_id"] == snap_id |
| 631 | assert data["output"] == str(out_file) |
| 632 | assert data["format"] == "tar.gz" |
| 633 | assert data["file_count"] >= 2 |
| 634 | assert data["size_bytes"] > 0 |
| 635 | |
| 636 | |
| 637 | def test_export_json_all_fields_zip(tmp_path: pathlib.Path) -> None: |
| 638 | _init_repo(tmp_path) |
| 639 | _create_files(tmp_path, 2) |
| 640 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 641 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 642 | out_file = tmp_path / "out.zip" |
| 643 | result = _invoke( |
| 644 | [ |
| 645 | "snapshot", "export", snap_id, |
| 646 | "--format", "zip", |
| 647 | "--output", str(out_file), |
| 648 | "--json", |
| 649 | ], |
| 650 | env=_env(tmp_path), |
| 651 | ) |
| 652 | assert result.exit_code == 0 |
| 653 | data: _ExportOut = json.loads(result.output) |
| 654 | assert data["format"] == "zip" |
| 655 | assert data["size_bytes"] > 0 |
| 656 | |
| 657 | |
| 658 | def test_export_json_and_archive_both_created(tmp_path: pathlib.Path) -> None: |
| 659 | _init_repo(tmp_path) |
| 660 | _create_files(tmp_path, 2) |
| 661 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 662 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 663 | out_file = tmp_path / "both.tar.gz" |
| 664 | result = _invoke( |
| 665 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 666 | env=_env(tmp_path), |
| 667 | ) |
| 668 | assert result.exit_code == 0 |
| 669 | assert out_file.exists() |
| 670 | assert tarfile.is_tarfile(str(out_file)) |
| 671 | data: _ExportOut = json.loads(result.output) |
| 672 | assert data["file_count"] >= 2 |
| 673 | |
| 674 | |
| 675 | # --------------------------------------------------------------------------- |
| 676 | # Integration: create → list → show → export pipeline |
| 677 | # --------------------------------------------------------------------------- |
| 678 | |
| 679 | |
| 680 | def test_pipeline_create_list_read_export(tmp_path: pathlib.Path) -> None: |
| 681 | _init_repo(tmp_path) |
| 682 | _create_files(tmp_path, 3) |
| 683 | |
| 684 | # 1. Create |
| 685 | create_res = _invoke( |
| 686 | ["snapshot", "create", "--json", "-m", "pipeline-note"], env=_env(tmp_path) |
| 687 | ) |
| 688 | assert create_res.exit_code == 0 |
| 689 | create_data: _CreateOut = json.loads(create_res.output) |
| 690 | snap_id = create_data["snapshot_id"] |
| 691 | assert create_data["note"] == "pipeline-note" |
| 692 | |
| 693 | # 2. List |
| 694 | list_res = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 695 | assert list_res.exit_code == 0 |
| 696 | list_items: list[_ListItemOut] = json.loads(list_res.output)["snapshots"] |
| 697 | assert any(item["snapshot_id"] == snap_id for item in list_items) |
| 698 | matching = next(i for i in list_items if i["snapshot_id"] == snap_id) |
| 699 | assert matching["note"] == "pipeline-note" |
| 700 | |
| 701 | # 3. Show |
| 702 | show_res = _invoke(["snapshot", "read", short_id(snap_id), "--json"], env=_env(tmp_path)) |
| 703 | assert show_res.exit_code == 0 |
| 704 | show_data: _ReadOut = json.loads(show_res.output) |
| 705 | assert show_data["snapshot_id"] == snap_id |
| 706 | assert show_data["note"] == "pipeline-note" |
| 707 | assert show_data["file_count"] == 3 |
| 708 | |
| 709 | # 4. Export tar.gz |
| 710 | out_tar = tmp_path / "pipe.tar.gz" |
| 711 | export_res = _invoke( |
| 712 | ["snapshot", "export", snap_id, "--output", str(out_tar), "--json"], |
| 713 | env=_env(tmp_path), |
| 714 | ) |
| 715 | assert export_res.exit_code == 0 |
| 716 | export_data: _ExportOut = json.loads(export_res.output) |
| 717 | assert export_data["file_count"] == 3 |
| 718 | assert out_tar.exists() |
| 719 | |
| 720 | # 5. Export zip |
| 721 | out_zip = tmp_path / "pipe.zip" |
| 722 | export_zip_res = _invoke( |
| 723 | [ |
| 724 | "snapshot", "export", snap_id, |
| 725 | "--format", "zip", |
| 726 | "--output", str(out_zip), |
| 727 | "--json", |
| 728 | ], |
| 729 | env=_env(tmp_path), |
| 730 | ) |
| 731 | assert export_zip_res.exit_code == 0 |
| 732 | assert zipfile.is_zipfile(str(out_zip)) |
| 733 | |
| 734 | |
| 735 | def test_multiple_snapshots_sorted_newest_first(tmp_path: pathlib.Path) -> None: |
| 736 | _init_repo(tmp_path) |
| 737 | for i in range(4): |
| 738 | _create_files(tmp_path, 1) |
| 739 | _invoke([f"snapshot", "create", "-m", f"snap-{i}"], env=_env(tmp_path)) |
| 740 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 741 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 742 | # Verify timestamps are non-increasing (newest first). |
| 743 | for j in range(len(items) - 1): |
| 744 | assert items[j]["created_at"] >= items[j + 1]["created_at"] |
| 745 | |
| 746 | |
| 747 | # --------------------------------------------------------------------------- |
| 748 | # E2E: help output |
| 749 | # --------------------------------------------------------------------------- |
| 750 | |
| 751 | |
| 752 | def test_create_help_shows_json_flag() -> None: |
| 753 | result = runner.invoke(cli, ["snapshot", "create", "--help"]) |
| 754 | assert result.exit_code == 0 |
| 755 | assert "--json" in result.output |
| 756 | |
| 757 | |
| 758 | def test_list_help_shows_json_flag() -> None: |
| 759 | result = runner.invoke(cli, ["snapshot", "list", "--help"]) |
| 760 | assert result.exit_code == 0 |
| 761 | assert "--json" in result.output |
| 762 | |
| 763 | |
| 764 | def test_read_help_mentions_json_flag() -> None: |
| 765 | result = runner.invoke(cli, ["snapshot", "read", "--help"]) |
| 766 | assert result.exit_code == 0 |
| 767 | assert "--json" in result.output |
| 768 | |
| 769 | |
| 770 | def test_export_help_shows_json_flag() -> None: |
| 771 | result = runner.invoke(cli, ["snapshot", "export", "--help"]) |
| 772 | assert result.exit_code == 0 |
| 773 | assert "--json" in result.output |
| 774 | |
| 775 | |
| 776 | def test_export_help_shows_format_choices() -> None: |
| 777 | result = runner.invoke(cli, ["snapshot", "export", "--help"]) |
| 778 | assert result.exit_code == 0 |
| 779 | assert "tar.gz" in result.output |
| 780 | assert "zip" in result.output |
| 781 | |
| 782 | |
| 783 | # --------------------------------------------------------------------------- |
| 784 | # Stress |
| 785 | # --------------------------------------------------------------------------- |
| 786 | |
| 787 | |
| 788 | def test_stress_200_snapshots_list(tmp_path: pathlib.Path) -> None: |
| 789 | _init_repo(tmp_path) |
| 790 | for i in range(200): |
| 791 | _write_snapshot(tmp_path, note=f"snap-{i}") |
| 792 | result = _invoke(["snapshot", "list", "--json", "--limit", "200"], env=_env(tmp_path)) |
| 793 | assert result.exit_code == 0 |
| 794 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 795 | assert len(items) == 200 |
| 796 | |
| 797 | |
| 798 | def test_stress_500_file_snapshot(tmp_path: pathlib.Path) -> None: |
| 799 | _init_repo(tmp_path) |
| 800 | for i in range(500): |
| 801 | (tmp_path / f"f{i}.txt").write_text(f"data-{i}", encoding="utf-8") |
| 802 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 803 | assert result.exit_code == 0 |
| 804 | data: _CreateOut = json.loads(result.output) |
| 805 | assert data["file_count"] >= 500 |
| 806 | |
| 807 | snap_id = data["snapshot_id"] |
| 808 | show_res = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 809 | assert show_res.exit_code == 0 |
| 810 | show_data: _ReadOut = json.loads(show_res.output) |
| 811 | assert show_data["file_count"] >= 500 |
| 812 | assert len(show_data["manifest"]) >= 500 |
| 813 | |
| 814 | |
| 815 | def test_stress_concurrent_create(tmp_path: pathlib.Path) -> None: |
| 816 | _init_repo(tmp_path) |
| 817 | for i in range(10): |
| 818 | (tmp_path / f"cf{i}.txt").write_text(f"c{i}", encoding="utf-8") |
| 819 | |
| 820 | errors: list[str] = [] |
| 821 | |
| 822 | def _create() -> None: |
| 823 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 824 | with _invoke_lock: |
| 825 | pass # lock already acquired inside _invoke |
| 826 | if result.exit_code != 0: |
| 827 | errors.append(f"exit_code={result.exit_code}") |
| 828 | |
| 829 | threads = [threading.Thread(target=_create) for _ in range(5)] |
| 830 | for t in threads: |
| 831 | t.start() |
| 832 | for t in threads: |
| 833 | t.join() |
| 834 | assert errors == [], f"Concurrent create errors: {errors}" |
| 835 | |
| 836 | |
| 837 | def test_stress_concurrent_list(tmp_path: pathlib.Path) -> None: |
| 838 | _init_repo(tmp_path) |
| 839 | for i in range(5): |
| 840 | _write_snapshot(tmp_path, note=f"concurrent-{i}") |
| 841 | |
| 842 | errors: list[str] = [] |
| 843 | |
| 844 | def _list() -> None: |
| 845 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 846 | if result.exit_code != 0: |
| 847 | errors.append(f"exit_code={result.exit_code}") |
| 848 | else: |
| 849 | try: |
| 850 | data = json.loads(result.output) |
| 851 | if len(data) < 5: |
| 852 | errors.append(f"expected 5 items, got {len(data)}") |
| 853 | except Exception as exc: |
| 854 | errors.append(str(exc)) |
| 855 | |
| 856 | threads = [threading.Thread(target=_list) for _ in range(10)] |
| 857 | for t in threads: |
| 858 | t.start() |
| 859 | for t in threads: |
| 860 | t.join() |
| 861 | assert errors == [], f"Concurrent list errors: {errors}" |
| 862 | |
| 863 | |
| 864 | # --------------------------------------------------------------------------- |
| 865 | # Extended / Security / Stress tests for ``muse snapshot create`` |
| 866 | # --------------------------------------------------------------------------- |
| 867 | |
| 868 | |
| 869 | class TestSnapshotCreateExtended: |
| 870 | """Unit, integration, and edge-case tests for ``muse snapshot create``.""" |
| 871 | |
| 872 | def test_create_help_contains_agent_quickstart(self) -> None: |
| 873 | result = runner.invoke(cli, ["snapshot", "create", "--help"]) |
| 874 | assert result.exit_code == 0 |
| 875 | assert "quickstart" in result.output.lower() or "muse snapshot create" in result.output |
| 876 | |
| 877 | def test_create_help_contains_json_schema(self) -> None: |
| 878 | result = runner.invoke(cli, ["snapshot", "create", "--help"]) |
| 879 | assert result.exit_code == 0 |
| 880 | assert "snapshot_id" in result.output |
| 881 | |
| 882 | def test_create_help_contains_exit_codes(self) -> None: |
| 883 | result = runner.invoke(cli, ["snapshot", "create", "--help"]) |
| 884 | assert result.exit_code == 0 |
| 885 | assert "exit code" in result.output.lower() or "0 —" in result.output |
| 886 | |
| 887 | def test_create_j_alias(self, tmp_path: pathlib.Path) -> None: |
| 888 | """-j is an alias for --json.""" |
| 889 | _init_repo(tmp_path) |
| 890 | _create_files(tmp_path, 2) |
| 891 | result = _invoke(["snapshot", "create", "-j"], env=_env(tmp_path)) |
| 892 | assert result.exit_code == 0 |
| 893 | data: _CreateOut = json.loads(result.output) |
| 894 | assert "snapshot_id" in data |
| 895 | |
| 896 | def test_create_snapshot_id_is_64_hex(self, tmp_path: pathlib.Path) -> None: |
| 897 | """snapshot_id in JSON output is exactly 64 hex characters.""" |
| 898 | _init_repo(tmp_path) |
| 899 | _create_files(tmp_path, 1) |
| 900 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 901 | assert result.exit_code == 0 |
| 902 | data: _CreateOut = json.loads(result.output) |
| 903 | assert len(data["snapshot_id"]) == 71 |
| 904 | assert all(c in "0123456789abcdef" for c in split_id(data["snapshot_id"])[1]) |
| 905 | |
| 906 | def test_create_file_count_matches_actual(self, tmp_path: pathlib.Path) -> None: |
| 907 | """file_count in JSON output matches the number of files created.""" |
| 908 | _init_repo(tmp_path) |
| 909 | _create_files(tmp_path, 7) |
| 910 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 911 | assert result.exit_code == 0 |
| 912 | data: _CreateOut = json.loads(result.output) |
| 913 | assert data["file_count"] >= 7 |
| 914 | |
| 915 | def test_create_created_at_is_iso8601(self, tmp_path: pathlib.Path) -> None: |
| 916 | """created_at field is ISO-8601 format (contains 'T' separator).""" |
| 917 | _init_repo(tmp_path) |
| 918 | _create_files(tmp_path, 1) |
| 919 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 920 | assert result.exit_code == 0 |
| 921 | data: _CreateOut = json.loads(result.output) |
| 922 | assert "T" in data["created_at"] |
| 923 | |
| 924 | def test_create_note_empty_when_not_supplied(self, tmp_path: pathlib.Path) -> None: |
| 925 | """note is empty string in JSON when -m not passed.""" |
| 926 | _init_repo(tmp_path) |
| 927 | _create_files(tmp_path, 1) |
| 928 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 929 | assert result.exit_code == 0 |
| 930 | data: _CreateOut = json.loads(result.output) |
| 931 | assert data["note"] == "" |
| 932 | |
| 933 | def test_create_note_persisted_in_json(self, tmp_path: pathlib.Path) -> None: |
| 934 | """note supplied via -m is reflected in JSON output.""" |
| 935 | _init_repo(tmp_path) |
| 936 | _create_files(tmp_path, 1) |
| 937 | result = _invoke(["snapshot", "create", "-m", "checkpoint", "--json"], env=_env(tmp_path)) |
| 938 | assert result.exit_code == 0 |
| 939 | data: _CreateOut = json.loads(result.output) |
| 940 | assert data["note"] == "checkpoint" |
| 941 | |
| 942 | def test_create_note_persists_to_list(self, tmp_path: pathlib.Path) -> None: |
| 943 | """note written via create is readable via list.""" |
| 944 | _init_repo(tmp_path) |
| 945 | _create_files(tmp_path, 1) |
| 946 | _invoke(["snapshot", "create", "-m", "roundtrip"], env=_env(tmp_path)) |
| 947 | list_result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 948 | assert list_result.exit_code == 0 |
| 949 | items: list[_ListItemOut] = json.loads(list_result.output)["snapshots"] |
| 950 | assert any(i["note"] == "roundtrip" for i in items) |
| 951 | |
| 952 | def test_create_note_persists_to_read(self, tmp_path: pathlib.Path) -> None: |
| 953 | """note written via create is readable via show.""" |
| 954 | _init_repo(tmp_path) |
| 955 | _create_files(tmp_path, 1) |
| 956 | create_result = _invoke( |
| 957 | ["snapshot", "create", "-m", "showcheck", "--json"], env=_env(tmp_path) |
| 958 | ) |
| 959 | snap_id = json.loads(create_result.output)["snapshot_id"] |
| 960 | show_result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 961 | assert show_result.exit_code == 0 |
| 962 | show_data: _ReadOut = json.loads(show_result.output) |
| 963 | assert show_data["note"] == "showcheck" |
| 964 | |
| 965 | def test_create_text_output_shows_short_id(self, tmp_path: pathlib.Path) -> None: |
| 966 | """Text output contains a 12-char prefix of the snapshot_id.""" |
| 967 | _init_repo(tmp_path) |
| 968 | _create_files(tmp_path, 1) |
| 969 | create_result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 970 | snap_id = json.loads(create_result.output)["snapshot_id"] |
| 971 | text_result = _invoke(["snapshot", "create"], env=_env(tmp_path)) |
| 972 | # Each create call produces a new snapshot; just verify format |
| 973 | assert text_result.exit_code == 0 |
| 974 | # Output should contain a hex-like prefix |
| 975 | assert any(c in "0123456789abcdef" for c in text_result.output) |
| 976 | |
| 977 | def test_create_text_output_shows_note(self, tmp_path: pathlib.Path) -> None: |
| 978 | """Text output shows note label when -m is supplied.""" |
| 979 | _init_repo(tmp_path) |
| 980 | _create_files(tmp_path, 1) |
| 981 | result = _invoke(["snapshot", "create", "-m", "my note"], env=_env(tmp_path)) |
| 982 | assert result.exit_code == 0 |
| 983 | assert "my note" in result.output |
| 984 | |
| 985 | def test_create_text_output_no_note_line_when_empty(self, tmp_path: pathlib.Path) -> None: |
| 986 | """Text output has no 'Note:' line when -m is not passed.""" |
| 987 | _init_repo(tmp_path) |
| 988 | _create_files(tmp_path, 1) |
| 989 | result = _invoke(["snapshot", "create"], env=_env(tmp_path)) |
| 990 | assert result.exit_code == 0 |
| 991 | assert "Note:" not in result.output |
| 992 | |
| 993 | def test_create_json_compact_no_indent(self, tmp_path: pathlib.Path) -> None: |
| 994 | """JSON output is compact (no indentation).""" |
| 995 | _init_repo(tmp_path) |
| 996 | _create_files(tmp_path, 1) |
| 997 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 998 | assert result.exit_code == 0 |
| 999 | # json.loads succeeds and the raw text has no indented lines |
| 1000 | assert "\n " not in result.output.strip() |
| 1001 | |
| 1002 | def test_create_repo_id_in_json(self, tmp_path: pathlib.Path) -> None: |
| 1003 | """repo_id field is present and non-empty in JSON output.""" |
| 1004 | _init_repo(tmp_path) |
| 1005 | _create_files(tmp_path, 1) |
| 1006 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1007 | assert result.exit_code == 0 |
| 1008 | data: _CreateOut = json.loads(result.output) |
| 1009 | assert data["repo_id"] == _REPO_ID |
| 1010 | |
| 1011 | def test_create_idempotent_same_files_same_id(self, tmp_path: pathlib.Path) -> None: |
| 1012 | """Two consecutive creates of the same working tree produce the same snapshot_id.""" |
| 1013 | _init_repo(tmp_path) |
| 1014 | _create_files(tmp_path, 3) |
| 1015 | r1 = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1016 | r2 = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1017 | assert r1.exit_code == 0 |
| 1018 | assert r2.exit_code == 0 |
| 1019 | assert json.loads(r1.output)["snapshot_id"] == json.loads(r2.output)["snapshot_id"] |
| 1020 | |
| 1021 | def test_create_different_files_different_id(self, tmp_path: pathlib.Path) -> None: |
| 1022 | """Adding a file between creates produces a different snapshot_id.""" |
| 1023 | _init_repo(tmp_path) |
| 1024 | _create_files(tmp_path, 2) |
| 1025 | r1 = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1026 | (tmp_path / "extra.txt").write_text("extra", encoding="utf-8") |
| 1027 | r2 = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1028 | assert r1.exit_code == 0 and r2.exit_code == 0 |
| 1029 | assert json.loads(r1.output)["snapshot_id"] != json.loads(r2.output)["snapshot_id"] |
| 1030 | |
| 1031 | |
| 1032 | class TestSnapshotCreateSecurity: |
| 1033 | """Security tests for ``muse snapshot create``.""" |
| 1034 | |
| 1035 | def test_create_ansi_note_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1036 | """ANSI escape codes in note are stripped from text output.""" |
| 1037 | _init_repo(tmp_path) |
| 1038 | _create_files(tmp_path, 1) |
| 1039 | malicious_note = "\x1b[31mDanger\x1b[0m" |
| 1040 | result = _invoke(["snapshot", "create", "-m", malicious_note], env=_env(tmp_path)) |
| 1041 | assert result.exit_code == 0 |
| 1042 | assert "\x1b[31m" not in result.output |
| 1043 | |
| 1044 | def test_create_ansi_note_raw_in_json(self, tmp_path: pathlib.Path) -> None: |
| 1045 | """ANSI escape codes in note are preserved raw in JSON output (agent data).""" |
| 1046 | _init_repo(tmp_path) |
| 1047 | _create_files(tmp_path, 1) |
| 1048 | malicious_note = "\x1b[31mDanger\x1b[0m" |
| 1049 | result = _invoke(["snapshot", "create", "-m", malicious_note, "--json"], env=_env(tmp_path)) |
| 1050 | assert result.exit_code == 0 |
| 1051 | data: _CreateOut = json.loads(result.output) |
| 1052 | assert data["note"] == malicious_note |
| 1053 | |
| 1054 | def test_create_control_char_note_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1055 | """CRLF and other control characters in note are stripped from text output.""" |
| 1056 | _init_repo(tmp_path) |
| 1057 | _create_files(tmp_path, 1) |
| 1058 | malicious_note = "good\r\ninjected line" |
| 1059 | result = _invoke(["snapshot", "create", "-m", malicious_note], env=_env(tmp_path)) |
| 1060 | assert result.exit_code == 0 |
| 1061 | assert "\r" not in result.output |
| 1062 | |
| 1063 | def test_create_very_long_note_no_crash(self, tmp_path: pathlib.Path) -> None: |
| 1064 | """A 10 000-character note does not crash the command.""" |
| 1065 | _init_repo(tmp_path) |
| 1066 | _create_files(tmp_path, 1) |
| 1067 | long_note = "x" * 10_000 |
| 1068 | result = _invoke(["snapshot", "create", "-m", long_note, "--json"], env=_env(tmp_path)) |
| 1069 | assert result.exit_code == 0 |
| 1070 | data: _CreateOut = json.loads(result.output) |
| 1071 | assert data["note"] == long_note |
| 1072 | |
| 1073 | def test_create_path_traversal_chars_in_note_no_crash(self, tmp_path: pathlib.Path) -> None: |
| 1074 | """Path-traversal-like characters in note do not crash or escape output.""" |
| 1075 | _init_repo(tmp_path) |
| 1076 | _create_files(tmp_path, 1) |
| 1077 | malicious_note = "../../etc/passwd" |
| 1078 | result = _invoke(["snapshot", "create", "-m", malicious_note, "--json"], env=_env(tmp_path)) |
| 1079 | assert result.exit_code == 0 |
| 1080 | data: _CreateOut = json.loads(result.output) |
| 1081 | assert data["note"] == malicious_note |
| 1082 | |
| 1083 | def test_create_snapshot_id_always_hex(self, tmp_path: pathlib.Path) -> None: |
| 1084 | """snapshot_id in text output is a safe hex substring with no control chars.""" |
| 1085 | _init_repo(tmp_path) |
| 1086 | _create_files(tmp_path, 1) |
| 1087 | # Get snapshot_id from JSON, verify text output contains its prefix |
| 1088 | json_result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1089 | snap_id = json.loads(json_result.output)["snapshot_id"] |
| 1090 | # Verify it is pure lowercase hex |
| 1091 | assert all(c in "0123456789abcdef" for c in split_id(snap_id)[1]) |
| 1092 | assert "\x1b" not in snap_id |
| 1093 | assert "\r" not in snap_id |
| 1094 | |
| 1095 | |
| 1096 | class TestSnapshotCreateStress: |
| 1097 | """Stress tests for ``muse snapshot create``.""" |
| 1098 | |
| 1099 | def test_create_1000_file_snapshot(self, tmp_path: pathlib.Path) -> None: |
| 1100 | """Snapshot of 1 000 files completes without error and reports correct count.""" |
| 1101 | _init_repo(tmp_path) |
| 1102 | for i in range(1000): |
| 1103 | (tmp_path / f"f{i}.dat").write_text(f"data-{i}", encoding="utf-8") |
| 1104 | result = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1105 | assert result.exit_code == 0 |
| 1106 | data: _CreateOut = json.loads(result.output) |
| 1107 | assert data["file_count"] >= 1000 |
| 1108 | |
| 1109 | def test_create_50_consecutive_snapshots(self, tmp_path: pathlib.Path) -> None: |
| 1110 | """50 consecutive creates all succeed and produce listable records.""" |
| 1111 | _init_repo(tmp_path) |
| 1112 | _create_files(tmp_path, 5) |
| 1113 | for i in range(50): |
| 1114 | (tmp_path / f"extra_{i}.txt").write_text(f"v{i}", encoding="utf-8") |
| 1115 | r = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1116 | assert r.exit_code == 0, f"Failed on iteration {i}: {r.output}" |
| 1117 | list_result = _invoke(["snapshot", "list", "--json", "--limit", "100"], env=_env(tmp_path)) |
| 1118 | assert list_result.exit_code == 0 |
| 1119 | items = json.loads(list_result.output)["snapshots"] |
| 1120 | assert len(items) >= 50 |
| 1121 | |
| 1122 | def test_create_concurrent_write_safety(self, tmp_path: pathlib.Path) -> None: |
| 1123 | """Concurrent snapshot creates on the same repo do not corrupt the store.""" |
| 1124 | _init_repo(tmp_path) |
| 1125 | for i in range(10): |
| 1126 | (tmp_path / f"cf{i}.txt").write_text(f"c{i}", encoding="utf-8") |
| 1127 | |
| 1128 | from muse.core.snapshots import write_snapshot |
| 1129 | from muse.core.ids import hash_snapshot |
| 1130 | |
| 1131 | errors: list[str] = [] |
| 1132 | |
| 1133 | def _do_create() -> None: |
| 1134 | try: |
| 1135 | # Use core directly — CliRunner serializes via _invoke_lock. |
| 1136 | manifest = {f"cf{i}.txt": blob_id(f"c{i}".encode()) for i in range(10)} |
| 1137 | snap_id = hash_snapshot(manifest) |
| 1138 | write_snapshot(tmp_path, SnapshotRecord( |
| 1139 | snapshot_id=snap_id, manifest=manifest, note="concurrent" |
| 1140 | )) |
| 1141 | except Exception as exc: # noqa: BLE001 |
| 1142 | errors.append(str(exc)) |
| 1143 | |
| 1144 | threads = [threading.Thread(target=_do_create) for _ in range(20)] |
| 1145 | for t in threads: |
| 1146 | t.start() |
| 1147 | for t in threads: |
| 1148 | t.join() |
| 1149 | assert not errors, f"Concurrent create failures: {errors}" |
| 1150 | |
| 1151 | |
| 1152 | # --------------------------------------------------------------------------- |
| 1153 | # Extended / Security / Stress tests for ``muse snapshot list`` |
| 1154 | # --------------------------------------------------------------------------- |
| 1155 | |
| 1156 | |
| 1157 | class TestSnapshotListExtended: |
| 1158 | """Unit, integration, and edge-case tests for ``muse snapshot list``.""" |
| 1159 | |
| 1160 | def test_list_help_contains_agent_quickstart(self) -> None: |
| 1161 | result = runner.invoke(cli, ["snapshot", "list", "--help"]) |
| 1162 | assert result.exit_code == 0 |
| 1163 | assert "quickstart" in result.output.lower() or "muse snapshot list" in result.output |
| 1164 | |
| 1165 | def test_list_help_contains_json_schema(self) -> None: |
| 1166 | result = runner.invoke(cli, ["snapshot", "list", "--help"]) |
| 1167 | assert result.exit_code == 0 |
| 1168 | assert "snapshot_id" in result.output |
| 1169 | |
| 1170 | def test_list_help_contains_exit_codes(self) -> None: |
| 1171 | result = runner.invoke(cli, ["snapshot", "list", "--help"]) |
| 1172 | assert result.exit_code == 0 |
| 1173 | assert "exit code" in result.output.lower() or "0 —" in result.output |
| 1174 | |
| 1175 | def test_list_j_alias(self, tmp_path: pathlib.Path) -> None: |
| 1176 | """-j is an alias for --json.""" |
| 1177 | _init_repo(tmp_path) |
| 1178 | _write_snapshot(tmp_path, note="alias-test") |
| 1179 | result = _invoke(["snapshot", "list", "-j"], env=_env(tmp_path)) |
| 1180 | assert result.exit_code == 0 |
| 1181 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 1182 | assert len(items) == 1 |
| 1183 | |
| 1184 | def test_list_json_compact_no_indent(self, tmp_path: pathlib.Path) -> None: |
| 1185 | """JSON output is compact (no indentation).""" |
| 1186 | _init_repo(tmp_path) |
| 1187 | _write_snapshot(tmp_path) |
| 1188 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1189 | assert result.exit_code == 0 |
| 1190 | assert "\n " not in result.output.strip() |
| 1191 | |
| 1192 | def test_list_empty_returns_empty_array_json(self, tmp_path: pathlib.Path) -> None: |
| 1193 | """Empty snapshot store emits '[]' with --json.""" |
| 1194 | _init_repo(tmp_path) |
| 1195 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1196 | assert result.exit_code == 0 |
| 1197 | assert json.loads(result.output)["snapshots"] == [] |
| 1198 | |
| 1199 | def test_list_empty_text_message(self, tmp_path: pathlib.Path) -> None: |
| 1200 | """Empty snapshot store prints a human message in text mode.""" |
| 1201 | _init_repo(tmp_path) |
| 1202 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 1203 | assert result.exit_code == 0 |
| 1204 | assert "no snapshots" in result.output.lower() |
| 1205 | |
| 1206 | def test_list_all_fields_present(self, tmp_path: pathlib.Path) -> None: |
| 1207 | """Every JSON item has snapshot_id, file_count, note, created_at.""" |
| 1208 | _init_repo(tmp_path) |
| 1209 | _write_snapshot(tmp_path, note="fields") |
| 1210 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1211 | assert result.exit_code == 0 |
| 1212 | item = json.loads(result.output)["snapshots"][0] |
| 1213 | assert "snapshot_id" in item |
| 1214 | assert "file_count" in item |
| 1215 | assert "note" in item |
| 1216 | assert "created_at" in item |
| 1217 | |
| 1218 | def test_list_snapshot_id_is_64_hex(self, tmp_path: pathlib.Path) -> None: |
| 1219 | """snapshot_id in each JSON item is 64 hex chars.""" |
| 1220 | _init_repo(tmp_path) |
| 1221 | _write_snapshot(tmp_path) |
| 1222 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1223 | assert result.exit_code == 0 |
| 1224 | sid = json.loads(result.output)["snapshots"][0]["snapshot_id"] |
| 1225 | assert len(sid) == 71 |
| 1226 | assert all(c in "0123456789abcdef" for c in split_id(sid)[1]) |
| 1227 | |
| 1228 | def test_list_created_at_iso8601(self, tmp_path: pathlib.Path) -> None: |
| 1229 | """created_at field contains 'T' (ISO-8601 separator).""" |
| 1230 | _init_repo(tmp_path) |
| 1231 | _write_snapshot(tmp_path) |
| 1232 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1233 | assert result.exit_code == 0 |
| 1234 | assert "T" in json.loads(result.output)["snapshots"][0]["created_at"] |
| 1235 | |
| 1236 | def test_list_newest_first_order(self, tmp_path: pathlib.Path) -> None: |
| 1237 | """Multiple snapshots appear newest-first in JSON output.""" |
| 1238 | import time as _time |
| 1239 | _init_repo(tmp_path) |
| 1240 | for i in range(5): |
| 1241 | _write_snapshot(tmp_path, note=f"snap-{i}", n_files=i + 1) |
| 1242 | _time.sleep(0.01) |
| 1243 | result = _invoke(["snapshot", "list", "--limit", "10", "--json"], env=_env(tmp_path)) |
| 1244 | assert result.exit_code == 0 |
| 1245 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 1246 | timestamps = [i["created_at"] for i in items] |
| 1247 | assert timestamps == sorted(timestamps, reverse=True) |
| 1248 | |
| 1249 | def test_list_limit_caps_results(self, tmp_path: pathlib.Path) -> None: |
| 1250 | """--limit N returns at most N snapshots.""" |
| 1251 | _init_repo(tmp_path) |
| 1252 | for i in range(10): |
| 1253 | _write_snapshot(tmp_path, note=f"s{i}") |
| 1254 | result = _invoke(["snapshot", "list", "--limit", "3", "--json"], env=_env(tmp_path)) |
| 1255 | assert result.exit_code == 0 |
| 1256 | assert len(json.loads(result.output)["snapshots"]) == 3 |
| 1257 | |
| 1258 | def test_list_limit_zero_exits_1(self, tmp_path: pathlib.Path) -> None: |
| 1259 | """--limit 0 is rejected with exit code 1.""" |
| 1260 | _init_repo(tmp_path) |
| 1261 | result = _invoke(["snapshot", "list", "--limit", "0"], env=_env(tmp_path)) |
| 1262 | assert result.exit_code == 1 |
| 1263 | |
| 1264 | def test_list_limit_negative_exits_1(self, tmp_path: pathlib.Path) -> None: |
| 1265 | """--limit -1 is rejected with exit code 1.""" |
| 1266 | _init_repo(tmp_path) |
| 1267 | result = _invoke(["snapshot", "list", "--limit", "-1"], env=_env(tmp_path)) |
| 1268 | assert result.exit_code == 1 |
| 1269 | |
| 1270 | def test_list_limit_error_mentions_limit(self, tmp_path: pathlib.Path) -> None: |
| 1271 | """Out-of-range --limit error output mentions 'limit'.""" |
| 1272 | _init_repo(tmp_path) |
| 1273 | result = _invoke(["snapshot", "list", "--limit", "0"], env=_env(tmp_path)) |
| 1274 | assert result.exit_code == 1 |
| 1275 | assert "limit" in result.stderr.lower() |
| 1276 | |
| 1277 | def test_list_note_in_text_output(self, tmp_path: pathlib.Path) -> None: |
| 1278 | """Note label appears in text output when present.""" |
| 1279 | _init_repo(tmp_path) |
| 1280 | _write_snapshot(tmp_path, note="my-label") |
| 1281 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 1282 | assert result.exit_code == 0 |
| 1283 | assert "my-label" in result.output |
| 1284 | |
| 1285 | def test_list_text_shows_short_id(self, tmp_path: pathlib.Path) -> None: |
| 1286 | """Text output shows the short_id prefix of snapshot_id.""" |
| 1287 | _init_repo(tmp_path) |
| 1288 | snap_id = _write_snapshot(tmp_path) |
| 1289 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 1290 | assert result.exit_code == 0 |
| 1291 | assert short_id(snap_id) in result.output |
| 1292 | |
| 1293 | def test_list_file_count_in_json(self, tmp_path: pathlib.Path) -> None: |
| 1294 | """file_count in JSON matches the number of files in the snapshot.""" |
| 1295 | _init_repo(tmp_path) |
| 1296 | _write_snapshot(tmp_path, n_files=7) |
| 1297 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1298 | assert result.exit_code == 0 |
| 1299 | assert json.loads(result.output)["snapshots"][0]["file_count"] == 7 |
| 1300 | |
| 1301 | |
| 1302 | class TestSnapshotListSecurity: |
| 1303 | """Security tests for ``muse snapshot list``.""" |
| 1304 | |
| 1305 | def test_list_ansi_note_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1306 | """ANSI escape codes in note are stripped from text output.""" |
| 1307 | _init_repo(tmp_path) |
| 1308 | _write_snapshot(tmp_path, note="\x1b[31mDanger\x1b[0m") |
| 1309 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 1310 | assert result.exit_code == 0 |
| 1311 | assert "\x1b[31m" not in result.output |
| 1312 | |
| 1313 | def test_list_ansi_note_raw_in_json(self, tmp_path: pathlib.Path) -> None: |
| 1314 | """ANSI escape codes in note are preserved raw in JSON output.""" |
| 1315 | _init_repo(tmp_path) |
| 1316 | malicious = "\x1b[31mDanger\x1b[0m" |
| 1317 | _write_snapshot(tmp_path, note=malicious) |
| 1318 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1319 | assert result.exit_code == 0 |
| 1320 | assert json.loads(result.output)["snapshots"][0]["note"] == malicious |
| 1321 | |
| 1322 | def test_list_control_char_note_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1323 | """CRLF in note is stripped from text output.""" |
| 1324 | _init_repo(tmp_path) |
| 1325 | _write_snapshot(tmp_path, note="good\r\ninjected") |
| 1326 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 1327 | assert result.exit_code == 0 |
| 1328 | assert "\r" not in result.output |
| 1329 | |
| 1330 | def test_list_symlink_in_objects_dir_skipped(self, tmp_path: pathlib.Path) -> None: |
| 1331 | """A symlink inside .muse/objects/ is skipped, not followed.""" |
| 1332 | from muse.core.paths import objects_dir as _objects_dir |
| 1333 | _init_repo(tmp_path) |
| 1334 | _write_snapshot(tmp_path, note="real") |
| 1335 | objs_dir = _objects_dir(tmp_path) |
| 1336 | shard_dir = objs_dir / "sha256" / "aa" |
| 1337 | shard_dir.mkdir(parents=True, exist_ok=True) |
| 1338 | fake = shard_dir / ("aa" * 31 + "0000") |
| 1339 | try: |
| 1340 | fake.symlink_to("/etc/passwd") |
| 1341 | except (OSError, NotImplementedError): |
| 1342 | pytest.skip("symlinks not supported on this platform") |
| 1343 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1344 | assert result.exit_code == 0 |
| 1345 | items: list[_ListItemOut] = json.loads(result.output)["snapshots"] |
| 1346 | assert len(items) == 1 |
| 1347 | assert items[0]["note"] == "real" |
| 1348 | |
| 1349 | def test_list_snapshot_id_prefix_in_text_is_safe_hex(self, tmp_path: pathlib.Path) -> None: |
| 1350 | """short_id(snapshot_id) in text output contains only hex chars.""" |
| 1351 | _init_repo(tmp_path) |
| 1352 | snap_id = _write_snapshot(tmp_path) |
| 1353 | result = _invoke(["snapshot", "list"], env=_env(tmp_path)) |
| 1354 | assert result.exit_code == 0 |
| 1355 | assert short_id(snap_id) in result.output |
| 1356 | assert "\x1b" not in result.output |
| 1357 | |
| 1358 | def test_list_very_long_note_no_crash(self, tmp_path: pathlib.Path) -> None: |
| 1359 | """A 10 000-character note does not crash list.""" |
| 1360 | _init_repo(tmp_path) |
| 1361 | _write_snapshot(tmp_path, note="x" * 10_000) |
| 1362 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1363 | assert result.exit_code == 0 |
| 1364 | assert json.loads(result.output)["snapshots"][0]["note"] == "x" * 10_000 |
| 1365 | |
| 1366 | |
| 1367 | class TestSnapshotListStress: |
| 1368 | """Stress tests for ``muse snapshot list``.""" |
| 1369 | |
| 1370 | def test_list_1000_snapshots(self, tmp_path: pathlib.Path) -> None: |
| 1371 | """Listing 1 000 snapshots with --limit 1000 returns all 1 000.""" |
| 1372 | _init_repo(tmp_path) |
| 1373 | for i in range(1000): |
| 1374 | _write_snapshot(tmp_path, note=f"s{i}") |
| 1375 | result = _invoke(["snapshot", "list", "--limit", "1000", "--json"], env=_env(tmp_path)) |
| 1376 | assert result.exit_code == 0 |
| 1377 | assert len(json.loads(result.output)["snapshots"]) == 1000 |
| 1378 | |
| 1379 | def test_list_default_limit_caps_at_20(self, tmp_path: pathlib.Path) -> None: |
| 1380 | """Default --limit of 20 caps a 50-snapshot store at 20 results.""" |
| 1381 | _init_repo(tmp_path) |
| 1382 | for i in range(50): |
| 1383 | _write_snapshot(tmp_path, note=f"s{i}") |
| 1384 | result = _invoke(["snapshot", "list", "--json"], env=_env(tmp_path)) |
| 1385 | assert result.exit_code == 0 |
| 1386 | assert len(json.loads(result.output)["snapshots"]) == 20 |
| 1387 | |
| 1388 | def test_list_concurrent_reads_safe(self, tmp_path: pathlib.Path) -> None: |
| 1389 | """Concurrent _list_all_snapshots core calls on the same repo do not crash.""" |
| 1390 | _init_repo(tmp_path) |
| 1391 | for i in range(10): |
| 1392 | _write_snapshot(tmp_path, note=f"c{i}") |
| 1393 | errors: list[str] = [] |
| 1394 | |
| 1395 | def _do_list() -> None: |
| 1396 | try: |
| 1397 | records = _list_all_snapshots(tmp_path) |
| 1398 | assert len(records) == 10 |
| 1399 | except Exception as exc: # noqa: BLE001 |
| 1400 | errors.append(str(exc)) |
| 1401 | |
| 1402 | threads = [threading.Thread(target=_do_list) for _ in range(15)] |
| 1403 | for t in threads: |
| 1404 | t.start() |
| 1405 | for t in threads: |
| 1406 | t.join() |
| 1407 | assert not errors, f"Concurrent failures: {errors}" |
| 1408 | |
| 1409 | |
| 1410 | # --------------------------------------------------------------------------- |
| 1411 | # Extended / Security / Stress tests for ``muse snapshot read`` |
| 1412 | # --------------------------------------------------------------------------- |
| 1413 | |
| 1414 | |
| 1415 | class TestSnapshotReadExtended: |
| 1416 | """Unit, integration, and edge-case tests for ``muse snapshot read``.""" |
| 1417 | |
| 1418 | def test_read_help_contains_agent_quickstart(self) -> None: |
| 1419 | result = runner.invoke(cli, ["snapshot", "read", "--help"]) |
| 1420 | assert result.exit_code == 0 |
| 1421 | assert "quickstart" in result.output.lower() or "muse snapshot read" in result.output |
| 1422 | |
| 1423 | def test_read_help_contains_json_schema(self) -> None: |
| 1424 | result = runner.invoke(cli, ["snapshot", "read", "--help"]) |
| 1425 | assert result.exit_code == 0 |
| 1426 | assert "snapshot_id" in result.output and "manifest" in result.output |
| 1427 | |
| 1428 | def test_read_help_contains_exit_codes(self) -> None: |
| 1429 | result = runner.invoke(cli, ["snapshot", "read", "--help"]) |
| 1430 | assert result.exit_code == 0 |
| 1431 | assert "exit code" in result.output.lower() or "0 —" in result.output |
| 1432 | |
| 1433 | def test_read_help_says_default_is_json(self) -> None: |
| 1434 | result = runner.invoke(cli, ["snapshot", "read", "--help"]) |
| 1435 | assert result.exit_code == 0 |
| 1436 | assert "json" in result.output.lower() |
| 1437 | |
| 1438 | def test_read_default_is_json_no_flag_needed(self, tmp_path: pathlib.Path) -> None: |
| 1439 | """show with no flags emits valid JSON.""" |
| 1440 | _init_repo(tmp_path) |
| 1441 | snap_id = _write_snapshot(tmp_path, note="default-json") |
| 1442 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1443 | assert result.exit_code == 0 |
| 1444 | data: _ReadOut = json.loads(result.output) |
| 1445 | assert data["snapshot_id"] == snap_id |
| 1446 | |
| 1447 | def test_read_json_compact_no_indent(self, tmp_path: pathlib.Path) -> None: |
| 1448 | """JSON output is compact (no indentation).""" |
| 1449 | _init_repo(tmp_path) |
| 1450 | snap_id = _write_snapshot(tmp_path) |
| 1451 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1452 | assert result.exit_code == 0 |
| 1453 | assert "\n " not in result.output.strip() |
| 1454 | |
| 1455 | def test_read_json_all_fields(self, tmp_path: pathlib.Path) -> None: |
| 1456 | """JSON output contains all five required fields.""" |
| 1457 | _init_repo(tmp_path) |
| 1458 | snap_id = _write_snapshot(tmp_path, note="fields-check", n_files=3) |
| 1459 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1460 | assert result.exit_code == 0 |
| 1461 | data: _ReadOut = json.loads(result.output) |
| 1462 | assert data["snapshot_id"] == snap_id |
| 1463 | assert "created_at" in data |
| 1464 | assert data["file_count"] == 3 |
| 1465 | assert data["note"] == "fields-check" |
| 1466 | assert isinstance(data["manifest"], dict) |
| 1467 | |
| 1468 | def test_read_manifest_sorted_alphabetically(self, tmp_path: pathlib.Path) -> None: |
| 1469 | """manifest keys in JSON output are sorted alphabetically.""" |
| 1470 | _init_repo(tmp_path) |
| 1471 | manifest = {f"z_file_{i}.txt": blob_id(f"z{i}".encode()) for i in range(5)} |
| 1472 | manifest.update({f"a_file_{i}.txt": blob_id(f"a{i}".encode()) for i in range(5)}) |
| 1473 | snap_id = hash_snapshot(manifest) |
| 1474 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 1475 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1476 | assert result.exit_code == 0 |
| 1477 | data: _ReadOut = json.loads(result.output) |
| 1478 | keys = list(data["manifest"].keys()) |
| 1479 | assert keys == sorted(keys) |
| 1480 | |
| 1481 | def test_read_prefix_resolves_short_id(self, tmp_path: pathlib.Path) -> None: |
| 1482 | """A 12-char prefix resolves to the full snapshot.""" |
| 1483 | _init_repo(tmp_path) |
| 1484 | snap_id = _write_snapshot(tmp_path, note="prefix-test") |
| 1485 | result = _invoke(["snapshot", "read", short_id(snap_id), "--json"], env=_env(tmp_path)) |
| 1486 | assert result.exit_code == 0 |
| 1487 | data: _ReadOut = json.loads(result.output) |
| 1488 | assert data["snapshot_id"] == snap_id |
| 1489 | |
| 1490 | def test_read_not_found_exits_1(self, tmp_path: pathlib.Path) -> None: |
| 1491 | """Unknown snapshot ID exits with code 1.""" |
| 1492 | _init_repo(tmp_path) |
| 1493 | result = _invoke(["snapshot", "read", "sha256:deadbeef"], env=_env(tmp_path)) |
| 1494 | assert result.exit_code == 1 |
| 1495 | |
| 1496 | def test_read_not_found_error_to_stderr(self, tmp_path: pathlib.Path) -> None: |
| 1497 | """Not-found message goes to stderr (captured in output by CliRunner).""" |
| 1498 | _init_repo(tmp_path) |
| 1499 | result = _invoke(["snapshot", "read", "sha256:deadbeef"], env=_env(tmp_path)) |
| 1500 | assert result.exit_code != 0 |
| 1501 | assert "not found" in result.stderr.lower() |
| 1502 | |
| 1503 | def test_read_text_flag_human_readable(self, tmp_path: pathlib.Path) -> None: |
| 1504 | """--text emits human-readable output (not JSON).""" |
| 1505 | _init_repo(tmp_path) |
| 1506 | snap_id = _write_snapshot(tmp_path, note="text-mode") |
| 1507 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1508 | assert result.exit_code == 0 |
| 1509 | # Text output starts with "snapshot_id:" label, not a JSON brace |
| 1510 | assert not result.output.strip().startswith("{") |
| 1511 | assert "snapshot_id:" in result.output |
| 1512 | |
| 1513 | def test_read_text_includes_note(self, tmp_path: pathlib.Path) -> None: |
| 1514 | """--text output includes the note label.""" |
| 1515 | _init_repo(tmp_path) |
| 1516 | snap_id = _write_snapshot(tmp_path, note="my-note") |
| 1517 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1518 | assert result.exit_code == 0 |
| 1519 | assert "my-note" in result.output |
| 1520 | |
| 1521 | def test_read_text_lists_files(self, tmp_path: pathlib.Path) -> None: |
| 1522 | """--text output lists file names from the manifest.""" |
| 1523 | _init_repo(tmp_path) |
| 1524 | snap_id = _write_snapshot(tmp_path, n_files=3) |
| 1525 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1526 | assert result.exit_code == 0 |
| 1527 | assert "file_0.txt" in result.output |
| 1528 | |
| 1529 | def test_read_file_count_matches_manifest(self, tmp_path: pathlib.Path) -> None: |
| 1530 | """file_count in JSON equals the number of keys in manifest.""" |
| 1531 | _init_repo(tmp_path) |
| 1532 | snap_id = _write_snapshot(tmp_path, n_files=9) |
| 1533 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1534 | assert result.exit_code == 0 |
| 1535 | data: _ReadOut = json.loads(result.output) |
| 1536 | assert data["file_count"] == len(data["manifest"]) |
| 1537 | |
| 1538 | def test_read_created_at_iso8601(self, tmp_path: pathlib.Path) -> None: |
| 1539 | """created_at field is ISO-8601 (contains 'T' separator).""" |
| 1540 | _init_repo(tmp_path) |
| 1541 | snap_id = _write_snapshot(tmp_path) |
| 1542 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1543 | assert result.exit_code == 0 |
| 1544 | assert "T" in json.loads(result.output)["created_at"] |
| 1545 | |
| 1546 | def test_read_note_empty_string_when_not_set(self, tmp_path: pathlib.Path) -> None: |
| 1547 | """note is empty string in JSON when not supplied at create time.""" |
| 1548 | _init_repo(tmp_path) |
| 1549 | snap_id = _write_snapshot(tmp_path, note="") |
| 1550 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1551 | assert result.exit_code == 0 |
| 1552 | assert json.loads(result.output)["note"] == "" |
| 1553 | |
| 1554 | def test_read_text_no_note_line_when_empty(self, tmp_path: pathlib.Path) -> None: |
| 1555 | """--text output has no 'note:' line when note is empty.""" |
| 1556 | _init_repo(tmp_path) |
| 1557 | snap_id = _write_snapshot(tmp_path, note="") |
| 1558 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1559 | assert result.exit_code == 0 |
| 1560 | assert "note:" not in result.output.lower() |
| 1561 | |
| 1562 | |
| 1563 | class TestSnapshotReadSecurity: |
| 1564 | """Security tests for ``muse snapshot read``.""" |
| 1565 | |
| 1566 | def test_read_ansi_note_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1567 | """ANSI in note is stripped from --text output.""" |
| 1568 | _init_repo(tmp_path) |
| 1569 | malicious = "\x1b[31mDanger\x1b[0m" |
| 1570 | snap_id = _write_snapshot(tmp_path, note=malicious) |
| 1571 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1572 | assert result.exit_code == 0 |
| 1573 | assert "\x1b[31m" not in result.output |
| 1574 | |
| 1575 | def test_read_ansi_note_raw_in_json(self, tmp_path: pathlib.Path) -> None: |
| 1576 | """ANSI in note is preserved raw in JSON output (agent data).""" |
| 1577 | _init_repo(tmp_path) |
| 1578 | malicious = "\x1b[31mDanger\x1b[0m" |
| 1579 | snap_id = _write_snapshot(tmp_path, note=malicious) |
| 1580 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1581 | assert result.exit_code == 0 |
| 1582 | assert json.loads(result.output)["note"] == malicious |
| 1583 | |
| 1584 | def test_read_ansi_rel_path_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1585 | """ANSI in a manifest path is stripped from --text output.""" |
| 1586 | _init_repo(tmp_path) |
| 1587 | malicious_path = "\x1b[32msrc/malicious.py\x1b[0m" |
| 1588 | manifest = {malicious_path: blob_id(b"malicious")} |
| 1589 | snap_id = hash_snapshot(manifest) |
| 1590 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 1591 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1592 | assert result.exit_code == 0 |
| 1593 | assert "\x1b[32m" not in result.output |
| 1594 | |
| 1595 | def test_read_control_char_note_stripped_in_text(self, tmp_path: pathlib.Path) -> None: |
| 1596 | """CRLF in note is stripped from --text output.""" |
| 1597 | _init_repo(tmp_path) |
| 1598 | snap_id = _write_snapshot(tmp_path, note="good\r\ninjected") |
| 1599 | result = _invoke(["snapshot", "read", snap_id], env=_env(tmp_path)) |
| 1600 | assert result.exit_code == 0 |
| 1601 | assert "\r" not in result.output |
| 1602 | |
| 1603 | def test_read_not_found_id_sanitized_in_error(self, tmp_path: pathlib.Path) -> None: |
| 1604 | """ANSI in a not-found snapshot ID is stripped from the error message.""" |
| 1605 | _init_repo(tmp_path) |
| 1606 | malicious_id = "\x1b[31mdeadbeef\x1b[0m" |
| 1607 | result = _invoke(["snapshot", "read", malicious_id], env=_env(tmp_path)) |
| 1608 | assert result.exit_code != 0 |
| 1609 | assert "\x1b[31m" not in result.output |
| 1610 | |
| 1611 | def test_read_symlink_in_prefix_scan_skipped(self, tmp_path: pathlib.Path) -> None: |
| 1612 | """A symlink in the object store is skipped during prefix scan.""" |
| 1613 | from muse.core.paths import objects_dir as _objects_dir |
| 1614 | _init_repo(tmp_path) |
| 1615 | real_id = _write_snapshot(tmp_path, note="real") |
| 1616 | objs_dir = _objects_dir(tmp_path) |
| 1617 | # Plant a symlink in the object store whose shard matches the real ID prefix. |
| 1618 | _, real_hex = split_id(real_id) |
| 1619 | shard_name = real_hex[:2] |
| 1620 | shard_dir = objs_dir / "sha256" / shard_name |
| 1621 | shard_dir.mkdir(parents=True, exist_ok=True) |
| 1622 | fake_name = real_hex[:4] + "f" * 58 |
| 1623 | fake = shard_dir / fake_name |
| 1624 | try: |
| 1625 | fake.symlink_to("/etc/passwd") |
| 1626 | except (OSError, NotImplementedError): |
| 1627 | pytest.skip("symlinks not supported on this platform") |
| 1628 | # Full ID lookup should still work — symlink at different ID is irrelevant |
| 1629 | result = _invoke(["snapshot", "read", real_id, "--json"], env=_env(tmp_path)) |
| 1630 | assert result.exit_code == 0 |
| 1631 | assert json.loads(result.output)["note"] == "real" |
| 1632 | |
| 1633 | |
| 1634 | class TestSnapshotReadStress: |
| 1635 | """Stress tests for ``muse snapshot read``.""" |
| 1636 | |
| 1637 | def test_read_500_file_manifest_json(self, tmp_path: pathlib.Path) -> None: |
| 1638 | """show returns all 500 manifest entries for a large snapshot.""" |
| 1639 | _init_repo(tmp_path) |
| 1640 | manifest = {f"f{i:04d}.dat": blob_id(f"data{i}".encode()) for i in range(500)} |
| 1641 | snap_id = hash_snapshot(manifest) |
| 1642 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 1643 | result = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1644 | assert result.exit_code == 0 |
| 1645 | data: _ReadOut = json.loads(result.output) |
| 1646 | assert data["file_count"] == 500 |
| 1647 | assert len(data["manifest"]) == 500 |
| 1648 | |
| 1649 | def test_read_50_consecutive_shows(self, tmp_path: pathlib.Path) -> None: |
| 1650 | """50 consecutive show calls on different snapshots all succeed.""" |
| 1651 | _init_repo(tmp_path) |
| 1652 | ids = [_write_snapshot(tmp_path, note=f"snap-{i}") for i in range(50)] |
| 1653 | for snap_id in ids: |
| 1654 | r = _invoke(["snapshot", "read", snap_id, "--json"], env=_env(tmp_path)) |
| 1655 | assert r.exit_code == 0, f"Failed for {short_id(snap_id)}: {r.output}" |
| 1656 | assert json.loads(r.output)["snapshot_id"] == snap_id |
| 1657 | |
| 1658 | def test_read_concurrent_reads_safe(self, tmp_path: pathlib.Path) -> None: |
| 1659 | """Concurrent _resolve_snapshot calls on the same snapshot do not crash.""" |
| 1660 | _init_repo(tmp_path) |
| 1661 | snap_id = _write_snapshot(tmp_path, note="concurrent", n_files=5) |
| 1662 | errors: list[str] = [] |
| 1663 | |
| 1664 | def _do_show() -> None: |
| 1665 | try: |
| 1666 | rec = _resolve_snapshot(tmp_path, snap_id) |
| 1667 | assert rec is not None |
| 1668 | assert rec.snapshot_id == snap_id |
| 1669 | assert rec.note == "concurrent" |
| 1670 | except Exception as exc: # noqa: BLE001 |
| 1671 | errors.append(str(exc)) |
| 1672 | |
| 1673 | threads = [threading.Thread(target=_do_show) for _ in range(20)] |
| 1674 | for t in threads: |
| 1675 | t.start() |
| 1676 | for t in threads: |
| 1677 | t.join() |
| 1678 | assert not errors, f"Concurrent failures: {errors}" |
| 1679 | |
| 1680 | |
| 1681 | # --------------------------------------------------------------------------- |
| 1682 | # Extended / Security / Stress tests for ``muse snapshot export`` |
| 1683 | # --------------------------------------------------------------------------- |
| 1684 | |
| 1685 | |
| 1686 | class TestSnapshotExportExtended: |
| 1687 | """Unit, integration, and edge-case tests for ``muse snapshot export``.""" |
| 1688 | |
| 1689 | def test_export_help_contains_agent_quickstart(self) -> None: |
| 1690 | result = runner.invoke(cli, ["snapshot", "export", "--help"]) |
| 1691 | assert result.exit_code == 0 |
| 1692 | assert "quickstart" in result.output.lower() or "muse snapshot export" in result.output |
| 1693 | |
| 1694 | def test_export_help_contains_json_schema(self) -> None: |
| 1695 | result = runner.invoke(cli, ["snapshot", "export", "--help"]) |
| 1696 | assert result.exit_code == 0 |
| 1697 | assert "size_bytes" in result.output |
| 1698 | |
| 1699 | def test_export_help_contains_exit_codes(self) -> None: |
| 1700 | result = runner.invoke(cli, ["snapshot", "export", "--help"]) |
| 1701 | assert result.exit_code == 0 |
| 1702 | assert "exit code" in result.output.lower() or "0 —" in result.output |
| 1703 | |
| 1704 | def test_export_j_alias(self, tmp_path: pathlib.Path) -> None: |
| 1705 | """-j is an alias for --json.""" |
| 1706 | _init_repo(tmp_path) |
| 1707 | _create_files(tmp_path, 2) |
| 1708 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1709 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1710 | out_file = tmp_path / "alias.tar.gz" |
| 1711 | result = _invoke( |
| 1712 | ["snapshot", "export", snap_id, "--output", str(out_file), "-j"], |
| 1713 | env=_env(tmp_path), |
| 1714 | ) |
| 1715 | assert result.exit_code == 0 |
| 1716 | data: _ExportOut = json.loads(result.output) |
| 1717 | assert data["snapshot_id"] == snap_id |
| 1718 | |
| 1719 | def test_export_tar_gz_default_format(self, tmp_path: pathlib.Path) -> None: |
| 1720 | """Default format is tar.gz; JSON reports format correctly.""" |
| 1721 | _init_repo(tmp_path) |
| 1722 | _create_files(tmp_path, 1) |
| 1723 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1724 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1725 | out_file = tmp_path / "out.tar.gz" |
| 1726 | result = _invoke( |
| 1727 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1728 | env=_env(tmp_path), |
| 1729 | ) |
| 1730 | assert result.exit_code == 0 |
| 1731 | assert json.loads(result.output)["format"] == "tar.gz" |
| 1732 | assert tarfile.is_tarfile(str(out_file)) |
| 1733 | |
| 1734 | def test_export_zip_format(self, tmp_path: pathlib.Path) -> None: |
| 1735 | """--format zip writes a valid zip archive.""" |
| 1736 | _init_repo(tmp_path) |
| 1737 | _create_files(tmp_path, 2) |
| 1738 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1739 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1740 | out_file = tmp_path / "out.zip" |
| 1741 | result = _invoke( |
| 1742 | ["snapshot", "export", snap_id, "--format", "zip", "--output", str(out_file), "--json"], |
| 1743 | env=_env(tmp_path), |
| 1744 | ) |
| 1745 | assert result.exit_code == 0 |
| 1746 | assert json.loads(result.output)["format"] == "zip" |
| 1747 | assert zipfile.is_zipfile(str(out_file)) |
| 1748 | |
| 1749 | def test_export_json_all_fields_present(self, tmp_path: pathlib.Path) -> None: |
| 1750 | """JSON output contains all five required fields.""" |
| 1751 | _init_repo(tmp_path) |
| 1752 | _create_files(tmp_path, 2) |
| 1753 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1754 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1755 | out_file = tmp_path / "fields.tar.gz" |
| 1756 | result = _invoke( |
| 1757 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1758 | env=_env(tmp_path), |
| 1759 | ) |
| 1760 | assert result.exit_code == 0 |
| 1761 | data: _ExportOut = json.loads(result.output) |
| 1762 | assert "snapshot_id" in data |
| 1763 | assert "output" in data |
| 1764 | assert "format" in data |
| 1765 | assert "file_count" in data |
| 1766 | assert "size_bytes" in data |
| 1767 | |
| 1768 | def test_export_json_compact_no_indent(self, tmp_path: pathlib.Path) -> None: |
| 1769 | """JSON output is compact (no indentation).""" |
| 1770 | _init_repo(tmp_path) |
| 1771 | _create_files(tmp_path, 1) |
| 1772 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1773 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1774 | out_file = tmp_path / "compact.tar.gz" |
| 1775 | result = _invoke( |
| 1776 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1777 | env=_env(tmp_path), |
| 1778 | ) |
| 1779 | assert result.exit_code == 0 |
| 1780 | assert "\n " not in result.output.strip() |
| 1781 | |
| 1782 | def test_export_size_bytes_positive(self, tmp_path: pathlib.Path) -> None: |
| 1783 | """size_bytes > 0 for a non-empty archive.""" |
| 1784 | _init_repo(tmp_path) |
| 1785 | _create_files(tmp_path, 3) |
| 1786 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1787 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1788 | out_file = tmp_path / "size.tar.gz" |
| 1789 | result = _invoke( |
| 1790 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1791 | env=_env(tmp_path), |
| 1792 | ) |
| 1793 | assert result.exit_code == 0 |
| 1794 | assert json.loads(result.output)["size_bytes"] > 0 |
| 1795 | |
| 1796 | def test_export_file_count_matches(self, tmp_path: pathlib.Path) -> None: |
| 1797 | """file_count in JSON matches number of files created.""" |
| 1798 | _init_repo(tmp_path) |
| 1799 | _create_files(tmp_path, 5) |
| 1800 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1801 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1802 | out_file = tmp_path / "count.tar.gz" |
| 1803 | result = _invoke( |
| 1804 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1805 | env=_env(tmp_path), |
| 1806 | ) |
| 1807 | assert result.exit_code == 0 |
| 1808 | assert json.loads(result.output)["file_count"] >= 5 |
| 1809 | |
| 1810 | def test_export_output_path_in_json(self, tmp_path: pathlib.Path) -> None: |
| 1811 | """output field in JSON matches the --output argument.""" |
| 1812 | _init_repo(tmp_path) |
| 1813 | _create_files(tmp_path, 1) |
| 1814 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1815 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1816 | out_file = tmp_path / "myarchive.tar.gz" |
| 1817 | result = _invoke( |
| 1818 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1819 | env=_env(tmp_path), |
| 1820 | ) |
| 1821 | assert result.exit_code == 0 |
| 1822 | assert json.loads(result.output)["output"] == str(out_file) |
| 1823 | |
| 1824 | def test_export_archive_actually_created(self, tmp_path: pathlib.Path) -> None: |
| 1825 | """The archive file is present on disk after export.""" |
| 1826 | _init_repo(tmp_path) |
| 1827 | _create_files(tmp_path, 2) |
| 1828 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1829 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1830 | out_file = tmp_path / "present.tar.gz" |
| 1831 | result = _invoke( |
| 1832 | ["snapshot", "export", snap_id, "--output", str(out_file)], |
| 1833 | env=_env(tmp_path), |
| 1834 | ) |
| 1835 | assert result.exit_code == 0 |
| 1836 | assert out_file.exists() |
| 1837 | |
| 1838 | def test_export_prefix_nests_files_in_tar(self, tmp_path: pathlib.Path) -> None: |
| 1839 | """--prefix nests all files under a directory inside the tar archive.""" |
| 1840 | _init_repo(tmp_path) |
| 1841 | _create_files(tmp_path, 2) |
| 1842 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1843 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1844 | out_file = tmp_path / "prefixed.tar.gz" |
| 1845 | result = _invoke( |
| 1846 | ["snapshot", "export", snap_id, "--output", str(out_file), "--prefix", "mydir"], |
| 1847 | env=_env(tmp_path), |
| 1848 | ) |
| 1849 | assert result.exit_code == 0 |
| 1850 | with tarfile.open(str(out_file)) as tf: |
| 1851 | names = tf.getnames() |
| 1852 | assert all(n.startswith("mydir/") for n in names) |
| 1853 | |
| 1854 | def test_export_not_found_exits_1(self, tmp_path: pathlib.Path) -> None: |
| 1855 | """Unknown snapshot ID exits with code 1.""" |
| 1856 | _init_repo(tmp_path) |
| 1857 | out_file = tmp_path / "nope.tar.gz" |
| 1858 | result = _invoke( |
| 1859 | ["snapshot", "export", "deadbeef", "--output", str(out_file)], |
| 1860 | env=_env(tmp_path), |
| 1861 | ) |
| 1862 | assert result.exit_code == 1 |
| 1863 | |
| 1864 | def test_export_prefix_scan_resolves_short_id(self, tmp_path: pathlib.Path) -> None: |
| 1865 | """A 12-char prefix resolves to the correct snapshot for export.""" |
| 1866 | _init_repo(tmp_path) |
| 1867 | _create_files(tmp_path, 1) |
| 1868 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1869 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1870 | out_file = tmp_path / "prefix_resolve.tar.gz" |
| 1871 | result = _invoke( |
| 1872 | ["snapshot", "export", short_id(snap_id), "--output", str(out_file), "--json"], |
| 1873 | env=_env(tmp_path), |
| 1874 | ) |
| 1875 | assert result.exit_code == 0 |
| 1876 | assert json.loads(result.output)["snapshot_id"] == snap_id |
| 1877 | |
| 1878 | def test_export_snapshot_id_in_json_is_full_hex(self, tmp_path: pathlib.Path) -> None: |
| 1879 | """snapshot_id in JSON is the full 64-char hex ID.""" |
| 1880 | _init_repo(tmp_path) |
| 1881 | _create_files(tmp_path, 1) |
| 1882 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1883 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1884 | out_file = tmp_path / "id_check.tar.gz" |
| 1885 | result = _invoke( |
| 1886 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1887 | env=_env(tmp_path), |
| 1888 | ) |
| 1889 | assert result.exit_code == 0 |
| 1890 | sid = json.loads(result.output)["snapshot_id"] |
| 1891 | assert len(sid) == 71 |
| 1892 | assert all(c in "0123456789abcdef" for c in split_id(sid)[1]) |
| 1893 | |
| 1894 | def test_export_text_output_mentions_path(self, tmp_path: pathlib.Path) -> None: |
| 1895 | """Text output mentions the archive filename.""" |
| 1896 | _init_repo(tmp_path) |
| 1897 | _create_files(tmp_path, 1) |
| 1898 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1899 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1900 | out_file = tmp_path / "mentioned.tar.gz" |
| 1901 | result = _invoke( |
| 1902 | ["snapshot", "export", snap_id, "--output", str(out_file)], |
| 1903 | env=_env(tmp_path), |
| 1904 | ) |
| 1905 | assert result.exit_code == 0 |
| 1906 | assert "mentioned.tar.gz" in result.output |
| 1907 | |
| 1908 | |
| 1909 | class TestSnapshotExportSecurity: |
| 1910 | """Security tests for ``muse snapshot export``.""" |
| 1911 | |
| 1912 | def test_export_not_found_id_sanitized(self, tmp_path: pathlib.Path) -> None: |
| 1913 | """ANSI in a not-found snapshot ID is stripped from the error message.""" |
| 1914 | _init_repo(tmp_path) |
| 1915 | malicious_id = "\x1b[31mdeadbeef\x1b[0m" |
| 1916 | out_file = tmp_path / "nope.tar.gz" |
| 1917 | result = _invoke( |
| 1918 | ["snapshot", "export", malicious_id, "--output", str(out_file)], |
| 1919 | env=_env(tmp_path), |
| 1920 | ) |
| 1921 | assert result.exit_code != 0 |
| 1922 | assert "\x1b[31m" not in result.output |
| 1923 | |
| 1924 | def test_export_text_output_no_ansi(self, tmp_path: pathlib.Path) -> None: |
| 1925 | """Normal text output from export contains no ANSI escape sequences.""" |
| 1926 | _init_repo(tmp_path) |
| 1927 | _create_files(tmp_path, 1) |
| 1928 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1929 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1930 | out_file = tmp_path / "clean.tar.gz" |
| 1931 | result = _invoke( |
| 1932 | ["snapshot", "export", snap_id, "--output", str(out_file)], |
| 1933 | env=_env(tmp_path), |
| 1934 | ) |
| 1935 | assert result.exit_code == 0 |
| 1936 | assert "\x1b[" not in result.output |
| 1937 | |
| 1938 | def test_export_zip_slip_dotdot_skipped(self, tmp_path: pathlib.Path) -> None: |
| 1939 | """A manifest entry with '..' segments is skipped (zip-slip guard).""" |
| 1940 | _init_repo(tmp_path) |
| 1941 | malicious_path = "../../../etc/passwd" |
| 1942 | obj_data = b"malicious content" |
| 1943 | obj_id = blob_id(obj_data) |
| 1944 | write_object(tmp_path, obj_id, obj_data) |
| 1945 | manifest = {malicious_path: obj_id} |
| 1946 | snap_id = hash_snapshot(manifest) |
| 1947 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 1948 | out_file = tmp_path / "slip.tar.gz" |
| 1949 | result = _invoke( |
| 1950 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1951 | env=_env(tmp_path), |
| 1952 | ) |
| 1953 | assert result.exit_code == 0 |
| 1954 | assert json.loads(result.output)["file_count"] == 0 |
| 1955 | |
| 1956 | def test_export_zip_slip_absolute_skipped(self, tmp_path: pathlib.Path) -> None: |
| 1957 | """A manifest entry with an absolute path is skipped (zip-slip guard).""" |
| 1958 | _init_repo(tmp_path) |
| 1959 | obj_data = b"absolute malicious" |
| 1960 | obj_id = blob_id(obj_data) |
| 1961 | write_object(tmp_path, obj_id, obj_data) |
| 1962 | manifest = {"/etc/passwd": obj_id} |
| 1963 | snap_id = hash_snapshot(manifest) |
| 1964 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 1965 | out_file = tmp_path / "abs.tar.gz" |
| 1966 | result = _invoke( |
| 1967 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1968 | env=_env(tmp_path), |
| 1969 | ) |
| 1970 | assert result.exit_code == 0 |
| 1971 | assert json.loads(result.output)["file_count"] == 0 |
| 1972 | |
| 1973 | def test_export_prefix_dotdot_skipped(self, tmp_path: pathlib.Path) -> None: |
| 1974 | """A --prefix containing '..' causes all entries to be skipped.""" |
| 1975 | _init_repo(tmp_path) |
| 1976 | _create_files(tmp_path, 1) |
| 1977 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 1978 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 1979 | out_file = tmp_path / "dotdot.tar.gz" |
| 1980 | result = _invoke( |
| 1981 | ["snapshot", "export", snap_id, "--output", str(out_file), |
| 1982 | "--prefix", "../escape", "--json"], |
| 1983 | env=_env(tmp_path), |
| 1984 | ) |
| 1985 | assert result.exit_code == 0 |
| 1986 | assert json.loads(result.output)["file_count"] == 0 |
| 1987 | |
| 1988 | def test_export_missing_object_skipped(self, tmp_path: pathlib.Path) -> None: |
| 1989 | """A manifest entry whose object is missing from the store is skipped.""" |
| 1990 | _init_repo(tmp_path) |
| 1991 | ghost_id = blob_id(b"ghost") |
| 1992 | manifest = {"ghost.txt": ghost_id} |
| 1993 | snap_id = hash_snapshot(manifest) |
| 1994 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 1995 | out_file = tmp_path / "ghost.tar.gz" |
| 1996 | result = _invoke( |
| 1997 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 1998 | env=_env(tmp_path), |
| 1999 | ) |
| 2000 | assert result.exit_code == 0 |
| 2001 | assert json.loads(result.output)["file_count"] == 0 |
| 2002 | |
| 2003 | |
| 2004 | class TestSnapshotExportStress: |
| 2005 | """Stress tests for ``muse snapshot export``.""" |
| 2006 | |
| 2007 | def test_export_500_file_tar_gz(self, tmp_path: pathlib.Path) -> None: |
| 2008 | """Export of a 500-file snapshot produces a valid tar.gz with all files.""" |
| 2009 | _init_repo(tmp_path) |
| 2010 | manifest: Manifest = {} |
| 2011 | for i in range(500): |
| 2012 | data = f"content-{i}".encode() |
| 2013 | obj_id = blob_id(data) |
| 2014 | write_object(tmp_path, obj_id, data) |
| 2015 | manifest[f"f{i:04d}.dat"] = obj_id |
| 2016 | snap_id = hash_snapshot(manifest) |
| 2017 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 2018 | out_file = tmp_path / "big.tar.gz" |
| 2019 | result = _invoke( |
| 2020 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 2021 | env=_env(tmp_path), |
| 2022 | ) |
| 2023 | assert result.exit_code == 0 |
| 2024 | data_out: _ExportOut = json.loads(result.output) |
| 2025 | assert data_out["file_count"] == 500 |
| 2026 | assert data_out["size_bytes"] > 0 |
| 2027 | assert tarfile.is_tarfile(str(out_file)) |
| 2028 | |
| 2029 | def test_export_500_file_zip(self, tmp_path: pathlib.Path) -> None: |
| 2030 | """Export of a 500-file snapshot produces a valid zip with all files.""" |
| 2031 | _init_repo(tmp_path) |
| 2032 | manifest: Manifest = {} |
| 2033 | for i in range(500): |
| 2034 | data = f"zip-content-{i}".encode() |
| 2035 | obj_id = blob_id(data) |
| 2036 | write_object(tmp_path, obj_id, data) |
| 2037 | manifest[f"z{i:04d}.dat"] = obj_id |
| 2038 | snap_id = hash_snapshot(manifest) |
| 2039 | write_snapshot(tmp_path, SnapshotRecord(snapshot_id=snap_id, manifest=manifest)) |
| 2040 | out_file = tmp_path / "big.zip" |
| 2041 | result = _invoke( |
| 2042 | ["snapshot", "export", snap_id, "--format", "zip", "--output", str(out_file), "--json"], |
| 2043 | env=_env(tmp_path), |
| 2044 | ) |
| 2045 | assert result.exit_code == 0 |
| 2046 | data_out: _ExportOut = json.loads(result.output) |
| 2047 | assert data_out["file_count"] == 500 |
| 2048 | assert zipfile.is_zipfile(str(out_file)) |
| 2049 | |
| 2050 | def test_export_10_consecutive_exports_same_snapshot(self, tmp_path: pathlib.Path) -> None: |
| 2051 | """10 consecutive exports of the same snapshot all succeed with consistent results.""" |
| 2052 | _init_repo(tmp_path) |
| 2053 | _create_files(tmp_path, 5) |
| 2054 | create_res = _invoke(["snapshot", "create", "--json"], env=_env(tmp_path)) |
| 2055 | snap_id: str = json.loads(create_res.output)["snapshot_id"] |
| 2056 | for i in range(10): |
| 2057 | out_file = tmp_path / f"repeat_{i}.tar.gz" |
| 2058 | result = _invoke( |
| 2059 | ["snapshot", "export", snap_id, "--output", str(out_file), "--json"], |
| 2060 | env=_env(tmp_path), |
| 2061 | ) |
| 2062 | assert result.exit_code == 0, f"Iteration {i} failed: {result.output}" |
| 2063 | data_out: _ExportOut = json.loads(result.output) |
| 2064 | assert data_out["snapshot_id"] == snap_id |
| 2065 | assert data_out["file_count"] >= 5 |
| 2066 | |
| 2067 | |
| 2068 | # --------------------------------------------------------------------------- |
| 2069 | # Flag registration tests |
| 2070 | # --------------------------------------------------------------------------- |
| 2071 | |
| 2072 | |
| 2073 | class TestRegisterFlags: |
| 2074 | def _parser(self) -> "argparse.ArgumentParser": |
| 2075 | import argparse |
| 2076 | from muse.cli.commands.snapshot_cmd import register |
| 2077 | |
| 2078 | p = argparse.ArgumentParser() |
| 2079 | subs = p.add_subparsers() |
| 2080 | register(subs) |
| 2081 | return p |
| 2082 | |
| 2083 | def test_default_json_out_is_false(self) -> None: |
| 2084 | args = self._parser().parse_args(["snapshot", "create"]) |
| 2085 | assert args.json_out is False |
| 2086 | |
| 2087 | def test_json_flag_sets_json_out(self) -> None: |
| 2088 | args = self._parser().parse_args(["snapshot", "create", "--json"]) |
| 2089 | assert args.json_out is True |
| 2090 | |
| 2091 | def test_j_shorthand_sets_json_out(self) -> None: |
| 2092 | args = self._parser().parse_args(["snapshot", "create", "-j"]) |
| 2093 | assert args.json_out is True |
File History
1 commit
sha256:84df9126d09aeec0b8f1b908f0b06c10913feec28f3514b382efb1ba6d619385
refactor: rename StructuredMergePlugin to AddressedMergePlu…
Sonnet 4.6
minor
⚠
23 days ago