fare-policy-assistant/tests/test_snapshots.py at main · ghzhost/fare-policy-assistant · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
"""Schema-2 source-complete corpus snapshot archives."""

from __future__ import annotations

import hashlib
import json
import threading
from concurrent.futures import ThreadPoolExecutor
from pathlib import Path

import pytest
import yaml

from assistant import config, snapshots
from assistant.ingest import Chunk
from assistant.snapshots import (
    ARCHIVE_SCHEMA,
    EVIDENCE_SCHEMA,
    SnapshotArchiveError,
    archive_snapshot,
    capture_source_material,
    collect_snapshot_material,
    list_snapshots,
    load_snapshot_chunks,
    validate_snapshot_archive,
)
from tests.conftest import make_chunk

_ARCHIVED_AT = "2026-07-30T12:34:56+00:00"
_BASELINE_SNAPSHOT = "b4834677a41abee7c6d1933ffd1f337e22fcadbc5f88442ccfc2b7ee3a84d59e"


def _manifest_document(chunk: Chunk, *, explicit_format: str | None = None) -> dict:
    document = {
        "id": chunk.doc_id,
        "agency": chunk.agency,
        "agency_full": chunk.agency_full,
        "title": chunk.doc_title,
        "url": chunk.url,
        "language": chunk.language,
    }
    if explicit_format is not None:
        document["format"] = explicit_format
    return document


def _metadata(chunk: Chunk, raw: bytes, *, effective_format: str = "html") -> dict:
    return {
        "doc_id": chunk.doc_id,
        "url": chunk.url,
        "final_url": chunk.url,
        "fetch_date": chunk.fetch_date,
        "http_status": 200,
        "format": effective_format,
        "sha256": hashlib.sha256(raw).hexdigest(),
        "bytes": len(raw),
    }


def _write_source(
    raw_dir: Path,
    chunk: Chunk,
    raw: bytes,
    *,
    effective_format: str = "html",
) -> bytes:
    raw_dir.mkdir(parents=True, exist_ok=True)
    (raw_dir / f"{chunk.doc_id}.{effective_format}").write_bytes(raw)
    metadata_bytes = yaml.safe_dump(
        _metadata(chunk, raw, effective_format=effective_format),
        sort_keys=False,
    ).encode("utf-8")
    (raw_dir / f"{chunk.doc_id}.meta.yaml").write_bytes(metadata_bytes)
    return metadata_bytes


def _rewrite_metadata(raw_dir: Path, source_doc_id: str, **changes: object) -> None:
    path = raw_dir / f"{source_doc_id}.meta.yaml"
    metadata = yaml.safe_load(path.read_bytes())
    metadata.update(changes)
    path.write_text(yaml.safe_dump(metadata, sort_keys=False), encoding="utf-8")


@pytest.fixture
def snapshot_case(tmp_path):
    mst = make_chunk(
        chunk_id="mst-fares#0",
        section="MST fares",
        text="MST's published single-ride fare is $2.00.",
    )
    sbmtd = make_chunk(
        chunk_id="sbmtd-fares-passes#0",
        doc_id="sbmtd-fares-passes",
        agency="SBMTD",
        agency_full="Santa Barbara Metropolitan Transit District",
        doc_title="Fares & Passes",
        url="https://sbmtd.gov/fares-passes/",
        section="Cash fares",
        text="SBMTD's published cash fare is $2.50.",
    )
    # Deliberately not doc-id sorted: chunk order affects retrieval tie-breaking
    # and must survive the archive round trip exactly.
    chunks = [sbmtd, mst]
    manifest = {
        "documents": [
            _manifest_document(sbmtd),
            _manifest_document(mst),
        ]
    }
    raw_dir = tmp_path / "raw"
    source_bytes = {
        mst.doc_id: b"<html><main>MST fare policy</main></html>",
        sbmtd.doc_id: b"<html><main>SBMTD fare policy</main></html>",
    }
    metadata_bytes = {
        chunk.doc_id: _write_source(raw_dir, chunk, source_bytes[chunk.doc_id]) for chunk in chunks
    }
    return {
        "chunks": chunks,
        "manifest": manifest,
        "raw_dir": raw_dir,
        "snapshots_dir": tmp_path / "snapshots",
        "source_bytes": source_bytes,
        "metadata_bytes": metadata_bytes,
    }


def _archive(case, *, archived_at: str = _ARCHIVED_AT):
    return archive_snapshot(
        case["chunks"],
        case["manifest"],
        raw_dir=case["raw_dir"],
        snapshots_dir=case["snapshots_dir"],
        archived_at=archived_at,
    )


def _visible_entries(root: Path) -> list[str]:
    if not root.exists():
        return []
    return sorted(path.name for path in root.iterdir() if not path.name.startswith("."))


def test_committed_baseline_snapshot_revalidates_after_git_checkout():
    identity = validate_snapshot_archive(config.SNAPSHOTS_DIR / _BASELINE_SNAPSHOT)

    assert identity.snapshot_version == _BASELINE_SNAPSHOT
    assert (
        identity.content_version
        == "58deec7dfdd250d2890c03510230c70618fb34e56d6b982ca2487af69379aa95"
    )


def test_archive_round_trip_retains_artifacts_raw_receipts_and_chunk_order(snapshot_case):
    identity = _archive(snapshot_case)
    archive = snapshot_case["snapshots_dir"] / identity.snapshot_version

    assert archive.name == identity.snapshot_version
    assert len(archive.name) == 64
    assert validate_snapshot_archive(archive) == identity
    assert (
        load_snapshot_chunks(identity.snapshot_version, snapshot_case["snapshots_dir"])
        == (snapshot_case["chunks"])
    )
    assert list_snapshots(snapshot_case["snapshots_dir"]) == [identity.snapshot_version]

    version = json.loads((archive / "version.json").read_text(encoding="utf-8"))
    assert version["archive_schema"] == ARCHIVE_SCHEMA
    assert version["scope"] == "web_policy"
    assert version["content_version"] == identity.content_version
    assert version["snapshot_version"] == identity.snapshot_version
    assert version["content_version_short"] == identity.content_version[:12]
    assert version["snapshot_version_short"] == identity.snapshot_version[:12]
    assert version["archived_at"] == _ARCHIVED_AT
    assert version["documents"] == 2
    assert version["chunks"] == 2
    assert version["evidence"] == "raw_and_receipts_verified"

    expected_artifacts = {
        "chunks.jsonl",
        "manifest.snapshot.yaml",
        "source-evidence.json",
        "raw/mst-fares.html",
        "raw/mst-fares.meta.yaml",
        "raw/sbmtd-fares-passes.html",
        "raw/sbmtd-fares-passes.meta.yaml",
    }
    assert set(version["artifacts"]) == expected_artifacts
    for relative, receipt in version["artifacts"].items():
        artifact = (archive / relative).read_bytes()
        assert receipt == {
            "sha256": hashlib.sha256(artifact).hexdigest(),
            "bytes": len(artifact),
        }

    for doc_id, expected in snapshot_case["source_bytes"].items():
        assert (archive / "raw" / f"{doc_id}.html").read_bytes() == expected
    for doc_id, expected in snapshot_case["metadata_bytes"].items():
        assert (archive / "raw" / f"{doc_id}.meta.yaml").read_bytes() == expected

    chunk_rows = [
        json.loads(line)
        for line in (archive / "chunks.jsonl").read_text(encoding="utf-8").splitlines()
    ]
    assert [row["chunk_id"] for row in chunk_rows] == [
        chunk.chunk_id for chunk in snapshot_case["chunks"]
    ]

    manifest = yaml.safe_load((archive / "manifest.snapshot.yaml").read_bytes())
    assert [document["id"] for document in manifest["documents"]] == [
        "mst-fares",
        "sbmtd-fares-passes",
    ]
    assert all(document["format"] == "html" for document in manifest["documents"])

    evidence = json.loads((archive / "source-evidence.json").read_text(encoding="utf-8"))
    assert evidence["schema"] == EVIDENCE_SCHEMA
    assert [row["doc_id"] for row in evidence["observations"]] == [
        "mst-fares",
        "sbmtd-fares-passes",
    ]


@pytest.mark.parametrize(
    ("mutation", "message"),
    [
        (
            lambda manifest, chunks: manifest["documents"].append(
                _manifest_document(
                    make_chunk(
                        chunk_id="extra-policy#0",
                        doc_id="extra-policy",
                        agency="Extra",
                        agency_full="Extra Transit",
                        doc_title="Extra policy",
                        url="https://example.org/extra/",
                    )
                )
            ),
            "zero chunks",
        ),
        (
            lambda manifest, chunks: manifest["documents"].pop(),
            "chunks absent from manifest",
        ),
        (
            lambda manifest, chunks: manifest["documents"][0].update(
                {"title": "A conflicting title"}
            ),
            "does not match chunk field doc_title",
        ),
        (
            lambda manifest, chunks: manifest["documents"].append(dict(manifest["documents"][0])),
            "duplicate document id",
        ),
    ],
)
def test_manifest_mismatches_fail_before_any_archive_is_visible(
    snapshot_case,
    mutation,
    message,
):
    mutation(snapshot_case["manifest"], snapshot_case["chunks"])

    with pytest.raises(SnapshotArchiveError, match=message):
        _archive(snapshot_case)

    assert _visible_entries(snapshot_case["snapshots_dir"]) == []


@pytest.mark.parametrize(
    ("changes", "message"),
    [
        ({"doc_id": "other-policy"}, "doc_id does not match manifest"),
        ({"url": "https://example.org/other/"}, "requested URL does not match manifest"),
        ({"fetch_date": "2026-07-30"}, "fetch date does not match chunks"),
        ({"sha256": "0" * 64}, "sha256 does not match"),
        ({"bytes": 999_999}, "bytes does not match"),
    ],
)
def test_receipt_mismatches_fail_before_any_archive_is_visible(
    snapshot_case,
    changes,
    message,
):
    doc_id = snapshot_case["chunks"][0].doc_id
    _rewrite_metadata(snapshot_case["raw_dir"], doc_id, **changes)

    with pytest.raises(SnapshotArchiveError, match=message):
        _archive(snapshot_case)

    assert _visible_entries(snapshot_case["snapshots_dir"]) == []


def test_missing_receipt_format_fails_closed(snapshot_case):
    doc_id = snapshot_case["chunks"][0].doc_id
    path = snapshot_case["raw_dir"] / f"{doc_id}.meta.yaml"
    metadata = yaml.safe_load(path.read_bytes())
    del metadata["format"]
    path.write_text(yaml.safe_dump(metadata, sort_keys=False), encoding="utf-8")

    with pytest.raises(SnapshotArchiveError, match="unsupported or missing format"):
        _archive(snapshot_case)

    assert _visible_entries(snapshot_case["snapshots_dir"]) == []


def test_manifest_and_receipt_format_disagreement_fails_closed(snapshot_case):
    snapshot_case["manifest"]["documents"][0]["format"] = "pdf"

    with pytest.raises(SnapshotArchiveError, match="formats disagree"):
        _archive(snapshot_case)

    assert _visible_entries(snapshot_case["snapshots_dir"]) == []


def test_both_raw_formats_for_one_document_are_rejected_as_ambiguous(snapshot_case):
    doc_id = snapshot_case["chunks"][0].doc_id
    (snapshot_case["raw_dir"] / f"{doc_id}.pdf").write_bytes(b"%PDF-1.7 stale snapshot")

    with pytest.raises(SnapshotArchiveError, match="format is ambiguous"):
        _archive(snapshot_case)

    assert _visible_entries(snapshot_case["snapshots_dir"]) == []


def test_rearchiving_identical_snapshot_preserves_first_archive(snapshot_case):
    first = _archive(snapshot_case)
    archive = snapshot_case["snapshots_dir"] / first.snapshot_version
    before = {
        path.relative_to(archive).as_posix(): path.read_bytes()
        for path in archive.rglob("*")
        if path.is_file()
    }

    second = _archive(snapshot_case, archived_at="2099-01-01T00:00:00+00:00")
    after = {
        path.relative_to(archive).as_posix(): path.read_bytes()
        for path in archive.rglob("*")
        if path.is_file()
    }

    assert second == first
    assert after == before
    version = json.loads((archive / "version.json").read_text(encoding="utf-8"))
    assert version["archived_at"] == _ARCHIVED_AT


def test_captured_sources_bind_archive_to_bytes_used_before_working_tree_changes(
    snapshot_case,
):
    captured = capture_source_material(
        snapshot_case["manifest"],
        snapshot_case["raw_dir"],
    )
    original = {source.observation.doc_id: source.raw_bytes for source in captured}
    changed_doc = snapshot_case["chunks"][0]
    replacement = b"<html><main>concurrent replacement</main></html>"
    _write_source(
        snapshot_case["raw_dir"],
        changed_doc,
        replacement,
    )

    identity = archive_snapshot(
        snapshot_case["chunks"],
        snapshot_case["manifest"],
        raw_dir=snapshot_case["raw_dir"],
        snapshots_dir=snapshot_case["snapshots_dir"],
        archived_at=_ARCHIVED_AT,
        sources=captured,
    )
    archive = snapshot_case["snapshots_dir"] / identity.snapshot_version

    assert (archive / "raw" / f"{changed_doc.doc_id}.html").read_bytes() == original[
        changed_doc.doc_id
    ]
    assert (snapshot_case["raw_dir"] / f"{changed_doc.doc_id}.html").read_bytes() == replacement
    assert validate_snapshot_archive(archive) == identity


def test_concurrent_identical_writers_converge_on_one_verified_archive(
    snapshot_case,
    monkeypatch,
):
    original = snapshots._write_staged_archive
    staged = threading.Barrier(2)

    def synchronize_after_staging(*args, **kwargs):
        original(*args, **kwargs)
        staged.wait(timeout=5)

    monkeypatch.setattr(snapshots, "_write_staged_archive", synchronize_after_staging)
    with ThreadPoolExecutor(max_workers=2) as executor:
        futures = [executor.submit(_archive, snapshot_case) for _ in range(2)]
        identities = [future.result(timeout=10) for future in futures]

    assert identities[0] == identities[1]
    assert _visible_entries(snapshot_case["snapshots_dir"]) == [identities[0].snapshot_version]
    assert (
        validate_snapshot_archive(snapshot_case["snapshots_dir"] / identities[0].snapshot_version)
        == identities[0]
    )


def test_existing_corrupt_destination_is_rejected_and_never_overwritten(snapshot_case):
    material = collect_snapshot_material(
        snapshot_case["chunks"],
        snapshot_case["manifest"],
        snapshot_case["raw_dir"],
    )
    final = snapshot_case["snapshots_dir"] / material.identity.snapshot_version
    final.mkdir(parents=True)
    sentinel = b'{"archive_schema": 2, "sentinel": "do not overwrite"}\n'
    (final / "version.json").write_bytes(sentinel)

    with pytest.raises(SnapshotArchiveError):
        _archive(snapshot_case)

    assert (final / "version.json").read_bytes() == sentinel
    assert list(final.iterdir()) == [final / "version.json"]


@pytest.mark.parametrize("artifact", ["raw", "metadata", "chunks", "evidence", "version"])
def test_any_published_artifact_tampering_is_detected(snapshot_case, artifact):
    identity = _archive(snapshot_case)
    archive = snapshot_case["snapshots_dir"] / identity.snapshot_version
    doc_id = snapshot_case["chunks"][0].doc_id

    if artifact == "raw":
        path = archive / "raw" / f"{doc_id}.html"
        path.write_bytes(path.read_bytes() + b"tampered")
    elif artifact == "metadata":
        path = archive / "raw" / f"{doc_id}.meta.yaml"
        metadata = yaml.safe_load(path.read_bytes())
        metadata["sha256"] = "0" * 64
        path.write_text(yaml.safe_dump(metadata, sort_keys=False), encoding="utf-8")
    elif artifact == "chunks":
        path = archive / "chunks.jsonl"
        rows = [json.loads(line) for line in path.read_text(encoding="utf-8").splitlines()]
        rows[0]["text"] += " Tampered."
        path.write_text(
            "\n".join(json.dumps(row, sort_keys=True) for row in rows) + "\n",
            encoding="utf-8",
        )
    elif artifact == "evidence":
        path = archive / "source-evidence.json"
        evidence = json.loads(path.read_text(encoding="utf-8"))
        evidence["observations"][0]["final_url"] = "https://example.org/tampered/"
        path.write_text(json.dumps(evidence, indent=2) + "\n", encoding="utf-8")
    else:
        path = archive / "version.json"
        version = json.loads(path.read_text(encoding="utf-8"))
        version["artifacts"]["chunks.jsonl"]["bytes"] += 1
        path.write_text(json.dumps(version, indent=2) + "\n", encoding="utf-8")

    with pytest.raises(SnapshotArchiveError):
        validate_snapshot_archive(archive)
    with pytest.raises(SnapshotArchiveError):
        load_snapshot_chunks(identity.snapshot_version, snapshot_case["snapshots_dir"])


def test_truncated_chunks_raise_the_public_archive_error(snapshot_case):
    identity = _archive(snapshot_case)
    archive = snapshot_case["snapshots_dir"] / identity.snapshot_version
    (archive / "chunks.jsonl").write_text('{"truncated":', encoding="utf-8")

    with pytest.raises(SnapshotArchiveError, match="chunks are malformed"):
        validate_snapshot_archive(archive)


def test_added_symbolic_link_directory_is_detected_as_tampering(
    snapshot_case,
    tmp_path,
):
    identity = _archive(snapshot_case)
    archive = snapshot_case["snapshots_dir"] / identity.snapshot_version
    outside = tmp_path / "outside"
    outside.mkdir()
    (archive / "linked-artifacts").symlink_to(outside, target_is_directory=True)

    with pytest.raises(SnapshotArchiveError, match="symbolic link"):
        validate_snapshot_archive(archive)


def test_listing_ignores_hidden_staging_directories(snapshot_case):
    root = snapshot_case["snapshots_dir"]
    hidden = root / ".abandoned-stage"
    hidden.mkdir(parents=True)
    (hidden / "garbage").write_text("incomplete", encoding="utf-8")
    (root / "not-a-snapshot").mkdir()

    identity = _archive(snapshot_case)

    assert list_snapshots(root) == [identity.snapshot_version]
    assert hidden.exists()


@pytest.mark.parametrize("fail_on_write", [1, 3, 8])
def test_staged_write_failure_cleans_up_and_publishes_nothing(
    snapshot_case,
    monkeypatch,
    fail_on_write,
):
    original = snapshots._write_file
    calls = 0

    def fail_at_selected_write(path: Path, content: bytes) -> None:
        nonlocal calls
        calls += 1
        if calls == fail_on_write:
            raise OSError("injected staged write failure")
        original(path, content)

    monkeypatch.setattr(snapshots, "_write_file", fail_at_selected_write)

    with pytest.raises(OSError, match="injected staged write failure"):
        _archive(snapshot_case)

    root = snapshot_case["snapshots_dir"]
    assert all(path.name == ".archive.lock" for path in root.iterdir())
    assert _visible_entries(root) == []