docs(harness): preserve private S3 Windows evidence and verifier
This commit is contained in:
@@ -0,0 +1 @@
|
||||
* -text
|
||||
@@ -0,0 +1,46 @@
|
||||
# Private S3 Windows evidence
|
||||
|
||||
Source `1b1c3e3abb6fc3d0934c9dbfbd21fddc8ae4f594`, following `90fae14818f684bf778640be2e613f34e3c7e999`. This checkpoint verifies the private typed index/rename slice and the rejected-depth syntax cleanup. **It does not establish current full-core runtime, public source8, integration, native UI or release acceptance.** Parent owns review/publication and integration.
|
||||
|
||||
| Recorded scope | Result |
|
||||
| --- | --- |
|
||||
| Full library listing | 167 test names; original 166 retained plus one cleanup test |
|
||||
| Typed source runtime | 59 passed, including all 9 index and 5 update tests |
|
||||
| Existing raw-token control | 1 passed |
|
||||
| Three focused 128 KiB controls | Original whole-flow, eight-parent cleanup and synthetic 256-frame Path tests passed |
|
||||
| Strict core, all targets | Clippy with `-D warnings`, exit0 |
|
||||
| Original90fae typed gate | Stack overflow retained; aborted run is not a passing full gate |
|
||||
| Temporary diagnostics/probes | Two dirty overflow reproductions; four dirty direct-parser malformed-input probes passed |
|
||||
|
||||
Actual platform was Windows x86_64-pc-windows-gnullvm, Rust/Cargo 1.98.1, one compiler job. Every clean gate binds 538 committed inputs before/after, exact commands and source. Raw source manifests, logs, receipts, helper scripts and diagnostic diffs remain byte-for-byte unchanged in [evidence.zip](evidence.zip). Its SHA256 is `4486f4c42b140ae4b9b3caf0aacdf151a2acb1fd85eda09a57764fd69e980574`; 658251 bytes, 61 members. No executables, caches or profiles are included.
|
||||
|
||||
The failure occurred when `Cursor::node` recursively dropped a rejected depth129 syntax tree below parser frames. Complete depth128 parsing/indexing/renaming/output/model equality/admission and earlier cleanup had finished. The source fix uses iterative rejected-kind disposal and preserves the existing stack request, depth limit and assertions. The 14 original S3 tests and the added mixed-shape cleanup test now have actual runtime evidence. Ordinary AST destruction and all possible parser-error paths are not claimed iterative or universally stack-safe.
|
||||
|
||||
## Reproduce evidence verification
|
||||
|
||||
With Python 3 standard library and Git, from any working directory:
|
||||
|
||||
```text
|
||||
python -B PATH/verify.py --repo PATH/TO/GIT/REPOSITORY
|
||||
python -B PATH/negative_controls.py --repo PATH/TO/GIT/REPOSITORY
|
||||
```
|
||||
|
||||
The repository needs immutable commits `1b1c3e3`, `90fae148` and `1f828323` (the publication commit contains that ancestry). Verification reads commit blobs; it neither depends on nor changes the current checkout. `--evidence-dir PATH` optionally selects another complete extracted payload directory. No archived helper, compiler or native artifact is executed. Optimized Python is refused.
|
||||
|
||||
The verifier checks both manifests/ZIP byte integrity, actual listing and exact selected runtime membership, 538 committed inputs, recorded commands/times/environment, strict completion, historical source-artifact manifests and the original aborted gate. It reconstructs the dirty source patches in memory and verifies the corresponding 538 test-input hashes. Broader tracked-file hashes also remain in the original captures; the verifier does not claim to audit unrelated tracked files outside that test-input scope. Binary identities are checked against raw executable-path receipts, not rehashed executable bytes, because binaries are deliberately absent. Initial focused tests and final restored typed/token/listing tests were separate builds of identical source and have their own binary receipts.
|
||||
|
||||
[checks.json](checks.json) records the positive working evidence audit and 13 discriminating negative controls. Those controls corrupt evidence only in memory; raw archives stay intact. Semantic cases reach their specific checks instead of failing only at the outer ZIP checksum. Package verification also runs against the saved commit and a fresh extraction; exact publication-commit readback is reported in the handoff to the parent, since embedding a commit's own hash in its contents is not possible.
|
||||
|
||||
## Review and historical evidence navigation
|
||||
|
||||
Extract the ZIP only if desired; the verifier itself inspects it in memory. Current authoritative raw closure is `closure.json` plus `raw/`. Relevant retained notes inside the ZIP:
|
||||
|
||||
- `notes/analysis-source8-s3-index-independent-review-20260913.md`: original S3 source review and required corrections.
|
||||
- `notes/analysis-source8-s3-correction-independent-review-20260922.md`: corrected Path/index source review before runtime.
|
||||
- `notes/analysis-source3-stack-cleanup-review-20260922.md`: independent review of iterative rejected-kind cleanup.
|
||||
- `notes/analysis-source3-stack-cleanup-handoff-20260922.md`: exact failure isolation, fix and completed Windows gates.
|
||||
- Original index/correction source manifests bind their respective immutable Git versions; their historical `unrun` labels are preserved and superseded only by the later actual receipts.
|
||||
|
||||
The ZIP's initial `README.md` contains the earlier portable-verifier proposal. It is retained as historical package text; this outer README and actual `verify.py` describe the completed verifier. No peer should execute the archived local orchestration helpers to verify evidence.
|
||||
|
||||
Source-bound original-span export, mixed ordinary/typed composition, complete output recompile/admission/intended-Project equality and public dispatch/migration/save integration remain separate work. This evidence does not close those requirements.
|
||||
@@ -0,0 +1,107 @@
|
||||
{
|
||||
"source": "1b1c3e3abb6fc3d0934c9dbfbd21fddc8ae4f594",
|
||||
"positive_audit": {
|
||||
"source": "1b1c3e3abb6fc3d0934c9dbfbd21fddc8ae4f594",
|
||||
"verified_git_inputs": 538,
|
||||
"listing": 167,
|
||||
"typed_pass": 59,
|
||||
"index_pass": 9,
|
||||
"update_pass": 5,
|
||||
"token_pass": 1,
|
||||
"strict_exit": 0,
|
||||
"historical_overflow_retained": true,
|
||||
"dirty_diagnostics": 2,
|
||||
"dirty_direct_parser_probes_pass": 4,
|
||||
"recorded_test_artifacts": {
|
||||
"stack-fixed-original-1b1c3e3": {
|
||||
"path": "C:\\Users\\User\\Documents\\Git\\ARCADIA-Studio\\.local\\analysis-checked-save-target-20260913\\debug\\deps\\oa_core-73987e0d7089bf1e.exe",
|
||||
"bytes": 39014400,
|
||||
"sha256": "56c0e5cd44ac5cb96b814d25ea403356195271a1aa9b6c6d1165fed4ffd09ddf"
|
||||
},
|
||||
"stack-fixed-shapes-1b1c3e3": {
|
||||
"path": "C:\\Users\\User\\Documents\\Git\\ARCADIA-Studio\\.local\\analysis-checked-save-target-20260913\\debug\\deps\\oa_core-73987e0d7089bf1e.exe",
|
||||
"bytes": 39014400,
|
||||
"sha256": "56c0e5cd44ac5cb96b814d25ea403356195271a1aa9b6c6d1165fed4ffd09ddf"
|
||||
},
|
||||
"stack-fixed-paths-1b1c3e3": {
|
||||
"path": "C:\\Users\\User\\Documents\\Git\\ARCADIA-Studio\\.local\\analysis-checked-save-target-20260913\\debug\\deps\\oa_core-73987e0d7089bf1e.exe",
|
||||
"bytes": 39014400,
|
||||
"sha256": "56c0e5cd44ac5cb96b814d25ea403356195271a1aa9b6c6d1165fed4ffd09ddf"
|
||||
},
|
||||
"typed-fixed-1b1c3e3": {
|
||||
"path": "C:\\Users\\User\\Documents\\Git\\ARCADIA-Studio\\.local\\analysis-checked-save-target-20260913\\debug\\deps\\oa_core-73987e0d7089bf1e.exe",
|
||||
"bytes": 39014400,
|
||||
"sha256": "7b38b54ad94678171bd8b5964b1e00dd7da8346d96276c1ad790d1a6eb335208"
|
||||
},
|
||||
"token-fixed-1b1c3e3": {
|
||||
"path": "C:\\Users\\User\\Documents\\Git\\ARCADIA-Studio\\.local\\analysis-checked-save-target-20260913\\debug\\deps\\oa_core-73987e0d7089bf1e.exe",
|
||||
"bytes": 39014400,
|
||||
"sha256": "7b38b54ad94678171bd8b5964b1e00dd7da8346d96276c1ad790d1a6eb335208"
|
||||
},
|
||||
"list-fixed-1b1c3e3": {
|
||||
"path": "C:\\Users\\User\\Documents\\Git\\ARCADIA-Studio\\.local\\analysis-checked-save-target-20260913\\debug\\deps\\oa_core-73987e0d7089bf1e.exe",
|
||||
"bytes": 39014400,
|
||||
"sha256": "7b38b54ad94678171bd8b5964b1e00dd7da8346d96276c1ad790d1a6eb335208"
|
||||
}
|
||||
},
|
||||
"scope": "Private S3 only; no full-core runtime/public8/integration/native/release acceptance. Recorded binary identities verified against receipts, not rehashed executable bytes; binaries intentionally absent."
|
||||
},
|
||||
"negative_control_count": 13,
|
||||
"negative_controls": [
|
||||
{
|
||||
"name": "changed_zip_bytes",
|
||||
"refusal": "Evidence ZIP SHA256 differs"
|
||||
},
|
||||
{
|
||||
"name": "duplicate_json_key",
|
||||
"refusal": "Duplicate JSON key: source"
|
||||
},
|
||||
{
|
||||
"name": "duplicate_zip_member_with_test_only_pin",
|
||||
"refusal": "Exact unique ZIP membership"
|
||||
},
|
||||
{
|
||||
"name": "unsafe_zip_member_with_test_only_pin",
|
||||
"refusal": "Unsafe ZIP path"
|
||||
},
|
||||
{
|
||||
"name": "truncated_strict_completion",
|
||||
"refusal": "Closed clean strict log"
|
||||
},
|
||||
{
|
||||
"name": "missing_typed_runtime_member",
|
||||
"refusal": "Exact ordered runtime results: typed-fixed-1b1c3e3"
|
||||
},
|
||||
{
|
||||
"name": "altered_gate_command",
|
||||
"refusal": "Exact command: token-fixed-1b1c3e3"
|
||||
},
|
||||
{
|
||||
"name": "altered_committed_input_hash",
|
||||
"refusal": "Exact538 source bindings: typed-fixed-1b1c3e3"
|
||||
},
|
||||
{
|
||||
"name": "dirty_run_mislabeled_as_fixed_source",
|
||||
"refusal": "Dirty capture identity/stability"
|
||||
},
|
||||
{
|
||||
"name": "historical_abort_mislabeled_success",
|
||||
"refusal": "Historical aborted gate"
|
||||
},
|
||||
{
|
||||
"name": "diagnostic_patch_disagrees_with_captured_hash",
|
||||
"refusal": "Reconstructed diagnostic input: crates/oa-core/src/language/analysis_source/index/tests.rs"
|
||||
},
|
||||
{
|
||||
"name": "probe_case_mislabeled",
|
||||
"refusal": "Dirty probe result/case"
|
||||
},
|
||||
{
|
||||
"name": "optimized_python_refused",
|
||||
"refusal": "Optimized Python refused"
|
||||
}
|
||||
],
|
||||
"verifier_sha256": "4da79ba0c6fe9dc0ae9b81da6589756d3a215620f2e2fd6cbefaa621f646ebc9",
|
||||
"controls_sha256": "a969264c98ab2fb95a0585b81c9d27a277ebe629488f20f45216d4483c26b396",
|
||||
"method": "Raw archive untouched. Semantic corruptions pass directly to audit() after authentic archive validation; ZIP structural negatives temporarily substitute only the test module pin. No compiler, artifact execution or worktree write."
|
||||
}
|
||||
Binary file not shown.
@@ -0,0 +1,36 @@
|
||||
{
|
||||
"source": "1b1c3e3abb6fc3d0934c9dbfbd21fddc8ae4f594",
|
||||
"archive_sha256": "4486f4c42b140ae4b9b3caf0aacdf151a2acb1fd85eda09a57764fd69e980574",
|
||||
"files": [
|
||||
{
|
||||
"path": ".gitattributes",
|
||||
"bytes": 8,
|
||||
"sha256": "705fd4d6451a31d36b3df7de96f83f30ac976c9b4a6d1e51671d8e2f33e2d0da"
|
||||
},
|
||||
{
|
||||
"path": "README.md",
|
||||
"bytes": 5456,
|
||||
"sha256": "8059aa1794544272e3cdaa622ff48a9d0fdf4c257a4a1ca0daf9cc1adcc8b061"
|
||||
},
|
||||
{
|
||||
"path": "verify.py",
|
||||
"bytes": 17589,
|
||||
"sha256": "4da79ba0c6fe9dc0ae9b81da6589756d3a215620f2e2fd6cbefaa621f646ebc9"
|
||||
},
|
||||
{
|
||||
"path": "negative_controls.py",
|
||||
"bytes": 5767,
|
||||
"sha256": "a969264c98ab2fb95a0585b81c9d27a277ebe629488f20f45216d4483c26b396"
|
||||
},
|
||||
{
|
||||
"path": "checks.json",
|
||||
"bytes": 4506,
|
||||
"sha256": "295acaadd2b1fe0f55c08c71c66c193635f70c2f7c63311ea42b039035065139"
|
||||
},
|
||||
{
|
||||
"path": "evidence.zip",
|
||||
"bytes": 658251,
|
||||
"sha256": "4486f4c42b140ae4b9b3caf0aacdf151a2acb1fd85eda09a57764fd69e980574"
|
||||
}
|
||||
]
|
||||
}
|
||||
@@ -0,0 +1,98 @@
|
||||
"""Read-only in-memory falsifiers for verify.py; does not modify archived evidence."""
|
||||
import argparse
|
||||
import importlib.util
|
||||
import io
|
||||
import json
|
||||
from pathlib import Path
|
||||
import subprocess
|
||||
import sys
|
||||
import warnings
|
||||
import zipfile
|
||||
|
||||
|
||||
def run(repo, directory):
|
||||
spec = importlib.util.spec_from_file_location('s3_evidence_verifier', directory / 'verify.py')
|
||||
verifier = importlib.util.module_from_spec(spec)
|
||||
spec.loader.exec_module(verifier)
|
||||
verifier.require(__debug__, 'Optimized Python refused')
|
||||
raw = (directory / 'evidence.zip').read_bytes()
|
||||
payload = verifier.archive_payload(raw)
|
||||
positive = verifier.audit(repo, payload)
|
||||
cases = []
|
||||
|
||||
def refused(name, expected, action):
|
||||
try:
|
||||
action()
|
||||
except RuntimeError as error:
|
||||
verifier.require(expected in str(error), f'{name}: wrong refusal: {error}')
|
||||
cases.append({'name': name, 'refusal': str(error)})
|
||||
else:
|
||||
raise RuntimeError(name + ': invalid evidence was accepted')
|
||||
|
||||
def changed_json(name, change):
|
||||
candidate = dict(payload)
|
||||
data = verifier.document(candidate[name])
|
||||
change(data)
|
||||
candidate[name] = json.dumps(data).encode()
|
||||
return candidate
|
||||
|
||||
refused('changed_zip_bytes', 'Evidence ZIP SHA256 differs', lambda: verifier.archive_payload(raw + b'tampered'))
|
||||
refused('duplicate_json_key', 'Duplicate JSON key', lambda: verifier.document(b'{"source":1,"source":2}'))
|
||||
|
||||
def changed_zip(mode):
|
||||
buffer = io.BytesIO()
|
||||
with zipfile.ZipFile(buffer, 'w') as archive:
|
||||
for name, content in payload.items():
|
||||
archive.writestr('../README.md' if mode == 'unsafe' and name == 'README.md' else name, content)
|
||||
if mode == 'duplicate':
|
||||
with warnings.catch_warnings():
|
||||
warnings.simplefilter('ignore', UserWarning)
|
||||
archive.writestr('README.md', payload['README.md'])
|
||||
malformed = buffer.getvalue()
|
||||
# Test-only pin substitution reaches ZIP structure checks. Production
|
||||
# archive_payload always requires its checked-in immutable ZIP digest.
|
||||
original_pin = verifier.ZIP_SHA
|
||||
try:
|
||||
verifier.ZIP_SHA = verifier.sha(malformed)
|
||||
verifier.archive_payload(malformed)
|
||||
finally:
|
||||
verifier.ZIP_SHA = original_pin
|
||||
|
||||
refused('duplicate_zip_member_with_test_only_pin', 'Exact unique ZIP membership', lambda: changed_zip('duplicate'))
|
||||
refused('unsafe_zip_member_with_test_only_pin', 'Unsafe ZIP path', lambda: changed_zip('unsafe'))
|
||||
|
||||
changed = dict(payload)
|
||||
changed['raw/strict-fixed-1b1c3e3.log'] = b' Compiling oa-core\n'
|
||||
refused('truncated_strict_completion', 'Closed clean strict log', lambda: verifier.audit(repo, changed))
|
||||
changed = dict(payload)
|
||||
name = 'raw/typed-fixed-1b1c3e3.log'
|
||||
changed[name] = b'\n'.join(line for line in changed[name].split(b'\n') if verifier.NEW.encode() not in line)
|
||||
refused('missing_typed_runtime_member', 'Exact ordered runtime results', lambda: verifier.audit(repo, changed))
|
||||
changed = changed_json('raw/token-fixed-1b1c3e3.result.json', lambda data: data['command'].remove('--offline'))
|
||||
refused('altered_gate_command', 'Exact command', lambda: verifier.audit(repo, changed))
|
||||
changed = changed_json('raw/typed-fixed-1b1c3e3.inputs.json', lambda data: data['inputs'][0].update(sha256='0' * 64))
|
||||
refused('altered_committed_input_hash', 'Exact538 source bindings', lambda: verifier.audit(repo, changed))
|
||||
changed = changed_json('raw/stack-diag-02-dirty.inputs.json', lambda data: data.update(head=verifier.SOURCE))
|
||||
refused('dirty_run_mislabeled_as_fixed_source', 'Dirty capture identity/stability', lambda: verifier.audit(repo, changed))
|
||||
changed = changed_json('raw/typed-90fae148.result.json', lambda data: data.update(exit=0))
|
||||
refused('historical_abort_mislabeled_success', 'Historical aborted gate', lambda: verifier.audit(repo, changed))
|
||||
changed = changed_json('raw/stack-diag-01-dirty.inputs.json', lambda data: data.update(dirty=data['dirty'].replace('STAGE entered', 'STAGE changed')))
|
||||
refused('diagnostic_patch_disagrees_with_captured_hash', 'Reconstructed diagnostic input', lambda: verifier.audit(repo, changed))
|
||||
changed = dict(payload)
|
||||
changed['raw/stack-probe-if-01-dirty.log'] = changed['raw/stack-probe-if-01-dirty.log'].replace(b'PROBE after malformed case 1:', b'PROBE after malformed case 0:')
|
||||
refused('probe_case_mislabeled', 'Dirty probe result/case', lambda: verifier.audit(repo, changed))
|
||||
result = subprocess.run([sys.executable, '-B', '-O', str(directory / 'verify.py'), '--repo', str(repo)], capture_output=True, text=True)
|
||||
verifier.require(result.returncode != 0 and 'Optimized Python refused' in result.stderr, 'Optimized verifier must refuse before audit')
|
||||
cases.append({'name': 'optimized_python_refused', 'refusal': 'Optimized Python refused'})
|
||||
return {'source': verifier.SOURCE, 'positive_audit': positive, 'negative_control_count': len(cases), 'negative_controls': cases, 'verifier_sha256': verifier.sha((directory / 'verify.py').read_bytes()), 'controls_sha256': verifier.sha(Path(__file__).read_bytes()), 'method': 'Raw archive untouched. Semantic corruptions pass directly to audit() after authentic archive validation; ZIP structural negatives temporarily substitute only the test module pin. No compiler, artifact execution or worktree write.'}
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument('--repo', type=Path, required=True)
|
||||
args = parser.parse_args()
|
||||
print(json.dumps(run(args.repo.resolve(), Path(__file__).resolve().parent), indent=2))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
@@ -0,0 +1,257 @@
|
||||
"""Read-only portable S3 evidence verification; never compiles or runs artifacts."""
|
||||
import argparse
|
||||
from datetime import datetime
|
||||
import hashlib
|
||||
import io
|
||||
import json
|
||||
from pathlib import Path, PurePosixPath, PureWindowsPath
|
||||
import re
|
||||
import subprocess
|
||||
import zipfile
|
||||
|
||||
SOURCE = '1b1c3e3abb6fc3d0934c9dbfbd21fddc8ae4f594'
|
||||
PRIOR = '90fae14818f684bf778640be2e613f34e3c7e999'
|
||||
ZIP_SHA = '4486f4c42b140ae4b9b3caf0aacdf151a2acb1fd85eda09a57764fd69e980574'
|
||||
RECORDER = '328970b226d9f4c3b8d3feb7d64478fb9756310ae5dfd3caa43b23c8b9d6f245'
|
||||
PREFIX = ['cargo', 'test', '--locked', '--offline', '-p', 'oa-core', '--lib']
|
||||
SCOPES = ['Cargo.toml', 'Cargo.lock', 'rust-toolchain.toml', '.cargo', 'build', 'crates', 'assets', 'docs', 'examples', 'harness/fixtures', 'vendor']
|
||||
STACK = 'deep_parse_index_plan_and_failed_build_cleanup_on_128k_stack'
|
||||
NEW = 'every_parent_shape_rejects_depth129_and_cleans_children_on_128k_stack'
|
||||
PATHS = 'maximum_index_paths_compare_lookup_debug_and_drop_on_128k_stack'
|
||||
OVERFLOW = 3221225725
|
||||
|
||||
|
||||
def require(ok, message):
|
||||
if not ok:
|
||||
raise RuntimeError(message)
|
||||
|
||||
|
||||
def sha(raw):
|
||||
return hashlib.sha256(raw).hexdigest()
|
||||
|
||||
|
||||
def unique(pairs):
|
||||
value = {}
|
||||
for key, item in pairs:
|
||||
require(key not in value, 'Duplicate JSON key: ' + key)
|
||||
value[key] = item
|
||||
return value
|
||||
|
||||
|
||||
def document(raw):
|
||||
return json.loads(raw, object_pairs_hook=unique)
|
||||
|
||||
|
||||
def git(repo, *args, data=None):
|
||||
return subprocess.check_output(['git', *args], cwd=repo, input=data)
|
||||
|
||||
|
||||
def source_inputs(repo, commit):
|
||||
paths = git(repo, 'ls-tree', '-r', '--name-only', commit, '--', *SCOPES).decode().splitlines()
|
||||
require(len(paths) == len(set(paths)) == 538, 'Exactly538 committed test inputs')
|
||||
batch = git(repo, 'cat-file', '--batch', data=''.join(f'{commit}:{p}\n' for p in paths).encode())
|
||||
offset, inputs, blobs = 0, [], {}
|
||||
for path in paths:
|
||||
end = batch.index(b'\n', offset)
|
||||
_, kind, size = batch[offset:end].split()
|
||||
size = int(size)
|
||||
raw = batch[end + 1:end + 1 + size]
|
||||
require(kind == b'blob' and len(raw) == size and batch[end + 1 + size:end + 2 + size] == b'\n', 'Git blob framing: ' + path)
|
||||
inputs.append({'path': path, 'bytes': size, 'sha256': sha(raw)})
|
||||
blobs[path] = raw
|
||||
offset = end + size + 2
|
||||
require(offset == len(batch), 'Complete Git blob stream')
|
||||
return inputs, blobs
|
||||
|
||||
|
||||
def archive_payload(raw):
|
||||
require(sha(raw) == ZIP_SHA, 'Evidence ZIP SHA256 differs')
|
||||
with zipfile.ZipFile(io.BytesIO(raw)) as archive:
|
||||
names = archive.namelist()
|
||||
require(len(names) == len(set(names)) == 61, 'Exact unique ZIP membership')
|
||||
for name in names:
|
||||
path = PurePosixPath(name)
|
||||
require(not path.is_absolute() and '..' not in path.parts and str(path) == name and '\\' not in name and ':' not in name, 'Unsafe ZIP path')
|
||||
payload = {name: archive.read(name) for name in names}
|
||||
manifest = document(payload['manifest.json'])
|
||||
require(manifest['source'] == SOURCE, 'Archive manifest source')
|
||||
listed = [item['path'] for item in manifest['files']]
|
||||
require(len(listed) == len(set(listed)) and set(listed) | {'manifest.json'} == set(payload), 'Archive manifest membership')
|
||||
for item in manifest['files']:
|
||||
require(len(payload[item['path']]) == item['bytes'] and sha(payload[item['path']]) == item['sha256'], 'Archive member binding: ' + item['path'])
|
||||
return payload
|
||||
|
||||
|
||||
def apply_recorded_diff(blobs, diff):
|
||||
"""Reconstruct diagnostic source without touching an index or worktree."""
|
||||
changed = {}
|
||||
sections = re.split(r'(?m)^diff --git ', diff)[1:]
|
||||
require(sections, 'Missing dirty diff')
|
||||
for section in sections:
|
||||
lines = section.splitlines(keepends=True)
|
||||
names = re.fullmatch(r'a/(\S+) b/(\S+)\n', lines[0])
|
||||
require(names is not None and names[1] == names[2] and names[1] in blobs, 'Unexpected diagnostic diff path')
|
||||
path = names[1]
|
||||
old = blobs[path].decode().splitlines(keepends=True)
|
||||
output, cursor, index = [], 0, 1
|
||||
while index < len(lines) and not lines[index].startswith('@@ '):
|
||||
index += 1
|
||||
require(index < len(lines), 'Missing diagnostic hunk')
|
||||
while index < len(lines):
|
||||
match = re.match(r'@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@', lines[index])
|
||||
require(match is not None, 'Invalid diagnostic hunk header')
|
||||
old_start, old_count = int(match[1]), int(match[2] or 1)
|
||||
new_start, new_count = int(match[3]), int(match[4] or 1)
|
||||
start = old_start - 1 if old_count else old_start
|
||||
require(cursor <= start <= len(old), 'Diagnostic hunk position')
|
||||
output.extend(old[cursor:start])
|
||||
cursor = start
|
||||
require(len(output) == (new_start - 1 if new_count else new_start), 'Diagnostic new hunk position')
|
||||
consumed = emitted = 0
|
||||
index += 1
|
||||
while index < len(lines) and not lines[index].startswith('@@ '):
|
||||
line = lines[index]
|
||||
require(line[:1] in [' ', '+', '-'], 'Unsupported diagnostic patch line')
|
||||
if line[0] in ' -':
|
||||
require(cursor < len(old) and old[cursor] == line[1:], 'Diagnostic old source mismatch')
|
||||
cursor += 1
|
||||
consumed += 1
|
||||
if line[0] in ' +':
|
||||
output.append(line[1:])
|
||||
emitted += 1
|
||||
index += 1
|
||||
require((consumed, emitted) == (old_count, new_count), 'Diagnostic hunk counts')
|
||||
output.extend(old[cursor:])
|
||||
require(path not in changed, 'Repeated diagnostic path')
|
||||
changed[path] = ''.join(output).encode()
|
||||
return changed
|
||||
|
||||
|
||||
def audit(repo, payload):
|
||||
read = lambda name: payload['raw/' + name]
|
||||
doc = lambda name: document(read(name))
|
||||
closure = document(payload['closure.json'])
|
||||
expected, blobs = source_inputs(repo, SOURCE)
|
||||
prior_inputs, prior_blobs = source_inputs(repo, PRIOR)
|
||||
require(git(repo, 'rev-parse', SOURCE + '^').decode().strip() == PRIOR, 'Correction parent')
|
||||
require(sha(payload['helpers/run-analysis-source3-gate-20260922.py']) == RECORDER, 'Recorder bytes')
|
||||
for manifest_name, commit, count in [
|
||||
('analysis-source8-s3-index-source-manifest-20260913.json', '1f8283236f8ccdefca13814003e7f046860b4f22', 7),
|
||||
('analysis-source8-s3-index-correction-source-manifest-20260913.json', PRIOR, 2),
|
||||
]:
|
||||
manifest = document(payload['notes/' + manifest_name])
|
||||
require(manifest['source'] == commit and len(manifest['files']) == count, 'Historical source artifact manifest')
|
||||
for item in manifest['files']:
|
||||
raw = git(repo, 'show', commit + ':' + item['path'])
|
||||
require(len(raw) == item['bytes'] and sha(raw) == item['sha256'] and git(repo, 'rev-parse', commit + ':' + item['path']).decode().strip() == item['git_blob'], 'Historical source artifact: ' + item['path'])
|
||||
names = re.findall(r'^([^\r\n]+): test\r?$', read('list-fixed-1b1c3e3.log').decode(), re.M)
|
||||
old_names = re.findall(r'^([^\r\n]+): test\r?$', read('list-90fae148.log').decode(), re.M)
|
||||
require(len(names) == len(set(names)) == 167 and len(old_names) == len(set(old_names)) == 166, 'Actual listing sizes')
|
||||
qualified_new = 'language::analysis_source::tests::' + NEW
|
||||
require(set(names) == set(old_names) | {qualified_new} and qualified_new not in old_names, 'Retained166 plus one new case')
|
||||
require(len([n for n in names if 'language::analysis_source' in n]) == 59 and len([n for n in names if 'language::raw_tokens' in n]) == 1, 'Typed/token listing counts')
|
||||
for module, count in [('index', 9), ('update', 5)]:
|
||||
path = f'crates/oa-core/src/language/analysis_source/{module}/tests.rs'
|
||||
authored = [f'language::analysis_source::{module}::tests::{name}' for name in re.findall(r'#\[test\]\s*fn (\w+)\(', blobs[path].decode())]
|
||||
require(len(authored) == count and set(authored) == {n for n in names if n.startswith(f'language::analysis_source::{module}::tests::')}, 'Source-derived membership: ' + module)
|
||||
selectors = {'stack-fixed-original-1b1c3e3': STACK, 'stack-fixed-shapes-1b1c3e3': NEW, 'stack-fixed-paths-1b1c3e3': PATHS, 'typed-fixed-1b1c3e3': 'language::analysis_source', 'token-fixed-1b1c3e3': 'language::raw_tokens', 'list-fixed-1b1c3e3': None, 'strict-fixed-1b1c3e3': 'clippy'}
|
||||
environment, artifacts, last_end = None, {}, None
|
||||
for stem, selector in selectors.items():
|
||||
record, log = doc(stem + '.result.json'), read(stem + '.log').decode()
|
||||
require(record['source'] == record['head_after'] == SOURCE and type(record['exit']) is int and record['exit'] == 0 and record['changed_inputs'] == [], 'Clean source/result: ' + stem)
|
||||
require(record['inputs'] == stem + '.inputs.json' and doc(record['inputs']) == {'source': SOURCE, 'inputs': expected}, 'Exact538 source bindings: ' + stem)
|
||||
require(record['launcher_sha256'] == RECORDER, 'Recorded launcher')
|
||||
start, end = (datetime.fromisoformat(record[k]) for k in ['start', 'end'])
|
||||
require(start.tzinfo is not None and end.tzinfo is not None and start <= end and (last_end is None or last_end <= start), 'Closed ordered gate times')
|
||||
last_end = end
|
||||
if environment is None:
|
||||
environment = record['environment']
|
||||
require(record['environment'] == environment and environment['CARGO_BUILD_JOBS'] == '1' and PureWindowsPath(environment['CARGO_TARGET_DIR']).name == 'analysis-checked-save-target-20260913', 'Recorded environment')
|
||||
if selector == 'clippy':
|
||||
command = ['cargo', 'clippy', '--locked', '--offline', '-p', 'oa-core', '--all-targets', '--', '-D', 'warnings']
|
||||
require(record['artifacts'] == [] and 'Finished `dev` profile [unoptimized]' in log and not re.search(r'^\s*(warning|error)(\[|:)', log, re.M), 'Closed clean strict log')
|
||||
selected = []
|
||||
else:
|
||||
require(len(record['artifacts']) == 1, 'One recorded test artifact')
|
||||
artifact = record['artifacts'][0]
|
||||
emitted = re.findall(r'Running unittests [^\r\n]+ \(([^\r\n]+\.exe)\)', log)
|
||||
require(len(emitted) == 1 and PureWindowsPath(emitted[0]) == PureWindowsPath(artifact['path']) and artifact['bytes'] > 0 and re.fullmatch('[0-9a-f]{64}', artifact['sha256']), 'Recorded executable/log binding')
|
||||
artifacts[stem] = artifact
|
||||
if selector is None:
|
||||
command = PREFIX + ['--', '--list']
|
||||
selected = names
|
||||
require(record['listing_members'] == names and record['listing_count'] == 167 and re.findall(r'^(\d+) tests, (\d+) benchmarks\s*$', log, re.M) == [('167', '0')], 'Listing receipt and summary')
|
||||
else:
|
||||
command = PREFIX + [selector, '--', '--test-threads=1']
|
||||
selected = [name for name in names if selector in name]
|
||||
rows = re.findall(r'^test ([^\r\n]+) \.\.\. (ok|FAILED|ignored)\r?$', log, re.M)
|
||||
require(rows == [(name, 'ok') for name in selected], 'Exact ordered runtime results: ' + stem)
|
||||
require(re.findall(r'^test result: ok\. (\d+) passed; (\d+) failed; (\d+) ignored; (\d+) measured; (\d+) filtered out;', log, re.M) == [(str(len(selected)), '0', '0', '0', str(167-len(selected)))], 'Runtime summary')
|
||||
require(record['command'] == command, 'Exact command: ' + stem)
|
||||
if selector is not None:
|
||||
require(record['listing_count'] == 0 and record['listing_members'] == [], 'Non-list metadata')
|
||||
saved = closure['gates'][stem]
|
||||
require([saved[k] for k in ['command', 'exit', 'start', 'end', 'names']] == [command, 0, record['start'], record['end'], selected], 'Closure gate facts')
|
||||
require(saved['pins'] == {stem+s: sha(read(stem+s)) for s in ['.inputs.json', '.log', '.result.json']}, 'Closure raw pins')
|
||||
require(artifacts['stack-fixed-original-1b1c3e3'] == artifacts['stack-fixed-shapes-1b1c3e3'] == artifacts['stack-fixed-paths-1b1c3e3'], 'First focused build receipt')
|
||||
require(artifacts['typed-fixed-1b1c3e3'] == artifacts['token-fixed-1b1c3e3'] == artifacts['list-fixed-1b1c3e3'], 'Final restored clean build receipt')
|
||||
require([closure[k] for k in ['source', 'parent', 'verified_inputs', 'listing', 'typed_pass', 'index_pass', 'update_pass', 'token_pass', 'strict_exit']] == [SOURCE, PRIOR, 538, 167, 59, 9, 5, 1, 0], 'Closure counts')
|
||||
for stem, mode in [('list-90fae148', 'list'), ('typed-90fae148', 'failed')]:
|
||||
record, log = doc(stem + '.result.json'), read(stem + '.log').decode()
|
||||
require(record['source'] == record['head_after'] == PRIOR and record['changed_inputs'] == [] and record['launcher_sha256'] == RECORDER, 'Historical frozen source')
|
||||
require(doc(stem + '.inputs.json') == {'source': PRIOR, 'inputs': prior_inputs}, 'Historical538 inputs')
|
||||
if mode == 'list':
|
||||
require(record['exit'] == 0 and record['listing_members'] == old_names and record['listing_count'] == 166 and record['command'] == PREFIX + ['--', '--list'], 'Historical listing')
|
||||
else:
|
||||
require(record['exit'] == OVERFLOW and record['command'] == PREFIX + ['language::analysis_source', '--', '--test-threads=1'] and 'STATUS_STACK_OVERFLOW' in log and STACK + ' ...' in log and 'test result:' not in log, 'Historical aborted gate')
|
||||
selected = [n for n in old_names if 'language::analysis_source' in n]
|
||||
failing = 'language::analysis_source::index::tests::' + STACK
|
||||
rows = re.findall(r'^test ([^\r\n]+) \.\.\. (ok|FAILED|ignored)\r?$', log, re.M)
|
||||
require(rows == [(n, 'ok') for n in selected[:selected.index(failing)]] and 'running 58 tests' in log, 'Historical completed prefix before abort')
|
||||
dirty = ['stack-diag-01-dirty', 'stack-diag-02-dirty', 'stack-probe-sum-01-dirty', 'stack-probe-if-01-dirty', 'stack-probe-power-01-dirty', 'stack-probe-trailing-01-dirty']
|
||||
for index, stem in enumerate(dirty):
|
||||
capture, result, log = doc(stem + '.inputs.json'), doc(stem + '.result.json'), read(stem + '.log').decode()
|
||||
diagnostic = index < 2
|
||||
commit, original = (PRIOR, prior_blobs) if diagnostic else (SOURCE, blobs)
|
||||
require(capture['head'] == commit and result['changed'] == [], 'Dirty capture identity/stability')
|
||||
patched = apply_recorded_diff(original, capture['dirty'])
|
||||
expected_paths = {'crates/oa-core/src/language/analysis_source/index/tests.rs'} if diagnostic else {'crates/oa-core/src/language/analysis_source/tests.rs'}
|
||||
if index == 1:
|
||||
expected_paths.add('crates/oa-core/src/language/analysis_source/syntax.rs')
|
||||
require(set(patched) == expected_paths, 'Bounded diagnostic paths')
|
||||
for path, raw in original.items():
|
||||
require(capture['inputs'][path] == sha(patched.get(path, raw)), 'Reconstructed diagnostic input: ' + path)
|
||||
expected_filter = STACK if diagnostic else 'diagnostic_malformed_cleanup_on_128k_stack'
|
||||
require(capture['command'] == PREFIX + [expected_filter, '--', '--test-threads=1', '--nocapture'], 'Diagnostic command')
|
||||
if diagnostic:
|
||||
require(result['exit'] == OVERFLOW and 'STATUS_STACK_OVERFLOW' in log and 'STAGE failed builds dropped' in log and 'STAGE finished' not in log, 'Diagnostic abort')
|
||||
if index == 1:
|
||||
require('STAGE node rejecting depth 129 before kind drop' in log and 'STAGE node rejected kind dropped' not in log, 'Isolated rejected-kind cleanup')
|
||||
else:
|
||||
case = index - 2
|
||||
require(result['exit'] == 0 and f'PROBE before malformed case {case}' in log and f'PROBE after malformed case {case}:' in log and '1 passed; 0 failed; 0 ignored' in log, 'Dirty probe result/case')
|
||||
return {'source': SOURCE, 'verified_git_inputs': 538, 'listing': 167, 'typed_pass': 59, 'index_pass': 9, 'update_pass': 5, 'token_pass': 1, 'strict_exit': 0, 'historical_overflow_retained': True, 'dirty_diagnostics': 2, 'dirty_direct_parser_probes_pass': 4, 'recorded_test_artifacts': artifacts, 'scope': 'Private S3 only; no full-core runtime/public8/integration/native/release acceptance. Recorded binary identities verified against receipts, not rehashed executable bytes; binaries intentionally absent.'}
|
||||
|
||||
|
||||
def verify_package(directory):
|
||||
manifest = document((directory / 'manifest.json').read_bytes())
|
||||
require(manifest['source'] == SOURCE and manifest['archive_sha256'] == ZIP_SHA, 'Outer manifest identity')
|
||||
names = [item['path'] for item in manifest['files']]
|
||||
require(set(names) == {'.gitattributes', 'README.md', 'verify.py', 'negative_controls.py', 'checks.json', 'evidence.zip'} and len(names) == len(set(names)), 'Outer manifest membership')
|
||||
for item in manifest['files']:
|
||||
raw = (directory / item['path']).read_bytes()
|
||||
require(len(raw) == item['bytes'] and sha(raw) == item['sha256'], 'Outer manifest bytes: ' + item['path'])
|
||||
return archive_payload((directory / 'evidence.zip').read_bytes())
|
||||
|
||||
|
||||
def main():
|
||||
require(__debug__, 'Optimized Python refused')
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument('--repo', type=Path, required=True, help='Git repo containing immutable source commits; current checkout is not used')
|
||||
parser.add_argument('--evidence-dir', type=Path, default=Path(__file__).resolve().parent)
|
||||
args = parser.parse_args()
|
||||
print(json.dumps(audit(args.repo, verify_package(args.evidence_dir)), indent=2))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
Reference in New Issue
Block a user