Compare commits
50 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 70a0ab091a | |||
| 48f8304517 | |||
| f2b9fe0149 | |||
| 32d99858c1 | |||
| 060d67aba5 | |||
| d0a466977c | |||
| 6d35a2e567 | |||
| 52f9740d99 | |||
| 4e071ca0f1 | |||
| f97852b630 | |||
| bcb8b2ad7e | |||
| 95ff77d77f | |||
| 9da6a45ec0 | |||
| 7407f97ef2 | |||
| 31ff7d1626 | |||
| 8a7896d352 | |||
| 35ff09a463 | |||
| 28805462ee | |||
| c469785a64 | |||
| 2f445eddc8 | |||
| 2907c3774f | |||
| a980b1a1df | |||
| 3754545003 | |||
| e38b2f1ca8 | |||
| 8e548287e9 | |||
| e85902bf18 | |||
| e82abec4c7 | |||
| 7d9c693761 | |||
| 98ef415912 | |||
| 90fd789a17 | |||
| 2000d340fa | |||
| 12f95cff91 | |||
| 9864632774 | |||
| f982851560 | |||
| 88aeb4d2eb | |||
| 83f3ebd26b | |||
| 13536d6268 | |||
| 7e63035570 | |||
| 44f4146f3e | |||
| b0bd9b69dc | |||
| 09a2d4ed20 | |||
| c8db6118d4 | |||
| 1c64971a52 | |||
| ac14df0625 | |||
| afb3b939ca | |||
| 8c147f5e6e | |||
| 6961710e96 | |||
| f4be9de547 | |||
| 5bdf9f0fe5 | |||
| 564821d883 |
+123
-3
@@ -19,8 +19,9 @@ jobs:
|
||||
# Public repository archive: no owner credentials or runner SSH keys.
|
||||
url = 'https://gitea.entroia.com/Zayden-Jung/paper-atlas/archive/' + sha + '.tar.gz'
|
||||
with urllib.request.urlopen(url, timeout=60) as response:
|
||||
archive = response.read(16*1024*1024+1)
|
||||
if len(archive)>16*1024*1024:
|
||||
# Full publication-source fixtures retain original bibliographic evidence.
|
||||
archive = response.read(24*1024*1024+1)
|
||||
if len(archive)>24*1024*1024:
|
||||
raise SystemExit('Unexpectedly large source archive')
|
||||
with tempfile.TemporaryDirectory(prefix='paper-atlas-ci-') as tmp:
|
||||
root = pathlib.Path(tmp)
|
||||
@@ -34,7 +35,7 @@ jobs:
|
||||
venv = root / 'venv'
|
||||
subprocess.run(['python3','-B','-m','venv','--system-site-packages',str(venv)],check=True)
|
||||
python = str(venv / 'bin/python')
|
||||
subprocess.run([python,'-B','-m','pip','install','--only-binary=:all:','--no-compile','psycopg[binary]==3.3.8'],check=True)
|
||||
subprocess.run([python,'-B','-m','pip','install','--only-binary=:all:','--no-compile','psycopg[binary]==3.3.8','pyarrow==26.0.1'],check=True)
|
||||
testenv = os.environ.copy()
|
||||
service_file = pathlib.Path('/etc/paper-atlas/pg_service.conf')
|
||||
if service_file.is_file():
|
||||
@@ -46,6 +47,77 @@ jobs:
|
||||
subprocess.run([python,'-B','-m','pip','install','--no-deps',str(wheel)],check=True)
|
||||
outside = root / 'outside-checkout'
|
||||
outside.mkdir()
|
||||
cikm_snapshots = pathlib.Path('/data/paper-index/validation/cikm2026-source-snapshots-v1')
|
||||
if cikm_snapshots.is_dir():
|
||||
cikm_output = root / 'installed-cikm2026'
|
||||
command = [python, '-I', '-B', '-m', 'paper_atlas', 'collect', 'cikm2026-sources']
|
||||
subprocess.run([*command, '--snapshots', str(cikm_snapshots), '--out', str(cikm_output)], cwd=outside, check=True)
|
||||
report = json.loads((cikm_output/'collection-report.json').read_text())
|
||||
assert {name: entry['records'] for name, entry in report['files'].items()} == {
|
||||
'programme.jsonl': 1008, 'alternate-programme.jsonl': 1005,
|
||||
'supplemental-programme.jsonl': 73, 'crossref.jsonl': 1081}
|
||||
assert len(report['registry_outside_doi_programme']) == 73 and not report['database_written']
|
||||
programme = [json.loads(line) for line in (cikm_output/'programme.jsonl').read_text().splitlines()]
|
||||
assert all(row['abstract'] and row['author_display_names'] for row in programme)
|
||||
assert len({row['doi'] for row in programme}) == 1008
|
||||
broken = root/'tampered-cikm2026'
|
||||
broken.mkdir()
|
||||
(broken/'program.html').write_text('tampered')
|
||||
rejected = subprocess.run([*command, '--snapshots', str(broken), '--out', str(root/'rejected-cikm2026')], cwd=outside, capture_output=True, text=True)
|
||||
assert rejected.returncode != 0 and 'Snapshot hash mismatch' in rejected.stderr
|
||||
assert not (root/'rejected-cikm2026').exists()
|
||||
print('CIKM 2026 full frozen-source CLI check passed', flush=True)
|
||||
mm_fixture = json.loads(gzip.decompress((src/'tests/fixtures/mm2026-source.json.gz').read_bytes()))
|
||||
mm_snapshots = root/'mm2026-snapshots'
|
||||
for name, body in mm_fixture['files'].items():
|
||||
path = mm_snapshots/name
|
||||
path.parent.mkdir(parents=True,exist_ok=True)
|
||||
path.write_bytes(body.encode('utf-8'))
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','mm2026-program','--snapshots',str(mm_snapshots),'--out',str(root/'installed-mm2026')],cwd=outside,check=True)
|
||||
mm_report = json.loads((root/'installed-mm2026/collection-report.json').read_text())
|
||||
assert (mm_report['records'],mm_report['program_observations'],mm_report['extra_presentation_observations']) == (2143,4326,2183)
|
||||
assert len(mm_report['files']) == 28 and mm_report['missing_abstract_count'] == 216
|
||||
assert mm_report['session_count'] == 157 and not mm_report['database_written'] and mm_report['network_requests'] == 0 and mm_report['approved_for_import'] == 0
|
||||
mm_rows = {r['paper_id']:r for f in mm_report['files'] for r in map(json.loads,pathlib.Path(f['file']).read_text().splitlines())}
|
||||
import hashlib
|
||||
assert set(mm_rows) == {g['paper_id'] for g in mm_fixture['independent_inventory']['inventory']}
|
||||
for gold in mm_fixture['independent_inventory']['inventory']:
|
||||
row = mm_rows[gold['paper_id']]
|
||||
assert (row['title'],row['authors'],row['affiliations'],row['track']) == (gold['title'],gold['authors'],gold['affiliations'],gold['track_literal'])
|
||||
assert hashlib.sha256(row['abstract'].encode()).hexdigest() == gold['abstract_sha256']
|
||||
assert [c['source_pointer'] for c in row['contexts']] == gold['locations']
|
||||
assert not row['approved_for_import'] and not row['final_decision_verified'] and 'emails' not in row
|
||||
mm_pub_fixture = json.loads(gzip.decompress((src/'tests/fixtures/mm2026-publications.json.gz').read_bytes()))
|
||||
mm_pub_registry = root/'mm2026-publication-registry'
|
||||
for name,body in mm_pub_fixture['files'].items():
|
||||
path=mm_pub_registry/name
|
||||
path.parent.mkdir(parents=True,exist_ok=True)
|
||||
path.write_bytes(body.encode('utf-8'))
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','mm2026-publications','--registry-snapshots',str(mm_pub_registry),'--program-snapshots',str(mm_snapshots),'--parent-snapshot',str(mm_pub_registry/'parent.json'),'--out',str(root/'installed-mm2026-publications')],cwd=outside,check=True)
|
||||
mm_pub_report=json.loads((root/'installed-mm2026-publications/collection-report.json').read_text())
|
||||
assert (mm_pub_report['records'],mm_pub_report['unique_registered_dois'],mm_pub_report['abstracts_from_exact_program_correspondence'],mm_pub_report['output_missing_abstract_count'])==(1918,1918,1743,175)
|
||||
assert mm_pub_report['unmatched_program_count']==391 and not mm_pub_report['database_written'] and mm_pub_report['approved_for_import']==0
|
||||
mm_pub_rows={r['external_id']:r for f in mm_pub_report['files'] for r in map(json.loads,pathlib.Path(f['file']).read_text().splitlines())}
|
||||
assert len(mm_pub_rows)==1918
|
||||
mm_format_gold=json.loads((src/'tests/fixtures/mm2026-title-formatting-review.json').read_text())
|
||||
mm_format_overrides={g['doi']:g for g in [*mm_format_gold['gold'],mm_format_gold['held_gold']]}
|
||||
assert mm_pub_report['abstracts_from_reviewed_title_formatting']==26 and mm_pub_report['abstracts_from_literal_program_correspondence']==1717
|
||||
for original_gold in mm_pub_fixture['independent_gold']:
|
||||
gold=mm_format_overrides.get(original_gold['doi'],original_gold)
|
||||
row=mm_pub_rows[gold['doi']]
|
||||
assert (row['title'],row['authors'],row['track'])==(gold['title'],gold['authors'],gold['track'])
|
||||
assert (hashlib.sha256(row['abstract'].encode()).hexdigest() if row['abstract'] is not None else None)==gold['abstract_sha256']
|
||||
assert row['metadata']['program_match_kind']==gold['kind'] and not row['metadata']['scientific_work_identity_approved'] and not row['metadata']['research_type_classified']
|
||||
if service_file.is_file():
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','audit','source-identities','--files',*[f['file'] for f in mm_pub_report['files']],'--pg-service','foundry-research','--pg-schema','research','--out',str(root/'installed-mm2026-publication-preflight')],cwd=outside,env=testenv,check=True)
|
||||
mm_pub_preflight=json.loads((root/'installed-mm2026-publication-preflight/postgresql-preflight.json').read_text())
|
||||
assert mm_pub_preflight['records']==1918 and mm_pub_preflight['transaction_read_only'] and not mm_pub_preflight['database_written'] and mm_pub_preflight['approved_for_import']==0
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','audit','source-identities','--mm-program-report',str(root/'installed-mm2026/collection-report.json'),'--pg-service','foundry-research','--pg-schema','research','--out',str(root/'installed-mm2026-preflight')],cwd=outside,env=testenv,check=True)
|
||||
mm_preflight = json.loads((root/'installed-mm2026-preflight/postgresql-preflight.json').read_text())
|
||||
assert mm_preflight['records'] == 2143 and len(mm_preflight['inputs']) == 28
|
||||
assert mm_preflight['transaction_read_only'] and not mm_preflight['database_written'] and mm_preflight['approved_for_import'] == 0
|
||||
assert sum(len(r['program_contexts']) for r in mm_preflight['entries']) == 4326
|
||||
assert all(r['status'] == 'unknown' and not r['identifiers'] and not r['contribution_type_verified'] and not r['publication_identity_verified'] and not r['final_decision_verified'] for r in mm_preflight['entries'])
|
||||
subprocess.run([str(venv/'bin/paper-atlas'),'--help'],cwd=outside,check=True)
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','cvf','--help'],cwd=outside,check=True)
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','rss2026-proceedings','--help'],cwd=outside,check=True)
|
||||
@@ -77,6 +149,22 @@ jobs:
|
||||
eccv_literal_row = next(row for row in eccv_rows if row['external_id'] == '2026/3844')
|
||||
assert eccv_literal_row['title'] == eccv_literal['gold']['visible_title']
|
||||
assert '&' in eccv_literal_row['title']
|
||||
icassp_fixture = json.loads(gzip.decompress((src/'tests/fixtures/icassp2026-source.json.gz').read_bytes()))
|
||||
icassp_snapshots = root/'icassp2026-snapshots'
|
||||
icassp_snapshots.mkdir()
|
||||
for name, encoded in icassp_fixture['files'].items():
|
||||
target = icassp_snapshots/name
|
||||
target.parent.mkdir(parents=True,exist_ok=True)
|
||||
target.write_bytes(base64.b64decode(encoded))
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','icassp2026-registry','--snapshots',str(icassp_snapshots),'--out',str(root/'installed-icassp2026')],cwd=outside,check=True)
|
||||
icassp_report = json.loads((root/'installed-icassp2026/collection-report.json').read_text())
|
||||
assert icassp_report['records'] == 4582 and icassp_report['missing_abstract_count'] == 4582
|
||||
assert icassp_report['single_name_author_fields'] == 23
|
||||
assert len(icassp_report['excluded_nonpaper_items']) == 7 and len(icassp_report['excluded_other_conference_dois']) == 56
|
||||
assert not icassp_report['database_written'] and icassp_report['network_requests'] == 0 and icassp_report['approved_for_import'] == 0
|
||||
icassp_rows = [json.loads(line) for line in pathlib.Path(icassp_report['files'][0]['file']).read_text().splitlines()]
|
||||
assert [{'doi':r['external_id'],'title':r['title'],'authors':r['authors']} for r in icassp_rows] == sorted(icassp_fixture['independent_gold'],key=lambda r:r['doi'])
|
||||
assert all(r['abstract'] is None and r['track'] == 'proceedings_unclassified' for r in icassp_rows)
|
||||
sdm_fixture = json.loads(gzip.decompress((src/'tests/fixtures/sdm2026-source.json.gz').read_bytes()))
|
||||
sdm_snapshots = root/'sdm2026-snapshots'
|
||||
for name, text in sdm_fixture['files'].items():
|
||||
@@ -150,6 +238,21 @@ jobs:
|
||||
recsys_rows = [json.loads(line) for manifest in recsys_report['files'] for line in pathlib.Path(manifest['file']).read_text().splitlines()]
|
||||
gold_keys = recsys_fixture['independent_gold'][0].keys()
|
||||
assert [{k:r['metadata']['program_fields'][k] for k in gold_keys} for r in recsys_rows] == recsys_fixture['independent_gold']
|
||||
ismir_fixture = json.loads(gzip.decompress((src/'tests/fixtures/ismir2026-source.json.gz').read_bytes()))
|
||||
ismir_snapshots = root/'ismir2026-snapshots'
|
||||
ismir_snapshots.mkdir()
|
||||
for name,encoded in ismir_fixture['files'].items():
|
||||
target = ismir_snapshots/name
|
||||
target.parent.mkdir(parents=True,exist_ok=True)
|
||||
target.write_bytes(base64.b64decode(encoded))
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','ismir2026-program','--snapshots',str(ismir_snapshots),'--out',str(root/'installed-ismir2026')],cwd=outside,check=True)
|
||||
ismir_report = json.loads((root/'installed-ismir2026/collection-report.json').read_text())
|
||||
assert (ismir_report['records'],ismir_report['missing_abstract_count'],ismir_report['review_optin_yes'],ismir_report['review_optin_no']) == (140,0,101,39)
|
||||
assert ismir_report['literal_title_version_differences'] == 2 and ismir_report['author_sequence_version_differences'] == 3
|
||||
assert not ismir_report['database_written'] and ismir_report['network_requests'] == 0 and ismir_report['approved_for_import'] == 0
|
||||
ismir_rows = [json.loads(line) for line in pathlib.Path(ismir_report['files'][0]['file']).read_text().splitlines()]
|
||||
assert [(r['external_id'],r['title'],r['authors'],r['abstract']) for r in ismir_rows] == [('2026/'+g['program_id'],g['title'],g['authors'],g['abstract']) for g in ismir_fixture['independent_gold']]
|
||||
assert all(not r['identifiers'] and not r['submission_identifiers'] and r['metadata']['paper_type'] is None for r in ismir_rows)
|
||||
corl_fixture = json.loads(gzip.decompress((src/'tests/fixtures/corl2026-source.json.gz').read_bytes()))
|
||||
corl_official = root/'corl2026-official'
|
||||
corl_archive = root/'corl2026-archive'
|
||||
@@ -190,6 +293,23 @@ jobs:
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','iclr-archive','--help'],cwd=outside,check=True)
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','acl','--help'],cwd=outside,check=True)
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','springer-selected','--help'],cwd=outside,check=True)
|
||||
publisher_fixture = "\n".join((
|
||||
"import pathlib, runpy, sys",
|
||||
"fixture = runpy.run_path(sys.argv[1])",
|
||||
"root = pathlib.Path(sys.argv[2])",
|
||||
"root.mkdir()",
|
||||
"fixture['snapshots'](root)",
|
||||
))
|
||||
publisher_input = root/'eccv-publisher-snapshots'
|
||||
subprocess.run([python,'-I','-B','-c',publisher_fixture,
|
||||
str(src/'tests/test_eccv2026_publications.py'),str(publisher_input)],cwd=outside,check=True)
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','eccv2026-publications',
|
||||
'--toc-snapshots',str(publisher_input/'toc'),
|
||||
'--chapter-snapshots',str(publisher_input/'chapters'),
|
||||
'--out',str(root/'installed-eccv-publisher')],cwd=outside,check=True)
|
||||
publisher_report = json.loads((root/'installed-eccv-publisher/collection-report.json').read_bytes())
|
||||
assert publisher_report['records'] == 1 and publisher_report['toc_chapters'] == 2835
|
||||
assert not publisher_report['database_written']
|
||||
springer = json.loads(gzip.decompress((src/'tests/fixtures/springer-aistats1993-fixture.json.gz').read_bytes()))
|
||||
for group in ('toc','chapters'):
|
||||
folder = root/'springer-snapshots'/group
|
||||
|
||||
@@ -0,0 +1,88 @@
|
||||
name: ReviewArena metadata projection
|
||||
on:
|
||||
workflow_dispatch:
|
||||
inputs:
|
||||
split:
|
||||
description: 'Pinned public mirror family; metadata columns only'
|
||||
required: true
|
||||
default: emnlp
|
||||
projection_name:
|
||||
description: 'Optional existing metadata output directory name for offline replay'
|
||||
required: false
|
||||
default: ''
|
||||
projection_sha256:
|
||||
description: 'Exact cached collection-report.json SHA256, required for offline replay'
|
||||
required: false
|
||||
default: ''
|
||||
jobs:
|
||||
project:
|
||||
runs-on: ces-control
|
||||
env:
|
||||
ARCHIVE_SHA: ${{ github.sha }}
|
||||
MIRROR_SPLIT: ${{ inputs.split }}
|
||||
MIRROR_PROJECTION: ${{ inputs.projection_name }}
|
||||
MIRROR_PROJECTION_SHA: ${{ inputs.projection_sha256 }}
|
||||
PYTHONDONTWRITEBYTECODE: '1'
|
||||
steps:
|
||||
- name: Project permitted metadata on the approved remote runner
|
||||
run: |
|
||||
python3 -B - <<'PY'
|
||||
import hashlib, io, json, os, pathlib, re, subprocess, tarfile, tempfile, urllib.request
|
||||
sha = os.environ['ARCHIVE_SHA']
|
||||
split = os.environ['MIRROR_SPLIT']
|
||||
cached = os.environ.get('MIRROR_PROJECTION','')
|
||||
cached_sha = os.environ.get('MIRROR_PROJECTION_SHA','')
|
||||
if not re.fullmatch(r'[0-9a-f]{40}', sha) or split not in ('colm','corl','emnlp','iclr','icml','neurips','tmlr'):
|
||||
raise SystemExit('Explicit approved source commit/split required')
|
||||
base = pathlib.Path('/data/paper-index')
|
||||
if not (base/'collected').is_dir() or not (base/'validation').is_dir():
|
||||
raise SystemExit('Confirmed ces-control host dataset paths unavailable; do not use another host')
|
||||
out = base/'collected'/('reviewarena-c2978add-'+split+'-'+sha[:8])
|
||||
if out.exists():
|
||||
raise SystemExit('Inspect existing output; no overwrite or automatic retry')
|
||||
extra=[]
|
||||
if cached:
|
||||
if not re.fullmatch('reviewarena-c2978add-'+split+r'-[0-9a-f]{8}',cached) or not re.fullmatch('[0-9a-f]{64}',cached_sha):
|
||||
raise SystemExit('Explicit permitted cached projection name and report hash required')
|
||||
extra=['--from-projection',str(base/'collected'/cached/'collection-report.json'),'--projection-sha256',cached_sha]
|
||||
elif cached_sha:
|
||||
raise SystemExit('Cached projection name required')
|
||||
url = 'https://gitea.entroia.com/Zayden-Jung/paper-atlas/archive/'+sha+'.tar.gz'
|
||||
with urllib.request.urlopen(url,timeout=60) as response:
|
||||
archive = response.read(24*1024*1024+1)
|
||||
if len(archive)>24*1024*1024:
|
||||
raise SystemExit('Oversized source archive')
|
||||
with tempfile.TemporaryDirectory(prefix='paper-atlas-mirror-') as tmp:
|
||||
root = pathlib.Path(tmp)
|
||||
with tarfile.open(fileobj=io.BytesIO(archive),mode='r:gz') as tar:
|
||||
tar.extractall(root,filter='data')
|
||||
candidates=list(root.glob('*/pyproject.toml'))
|
||||
if len(candidates)!=1:
|
||||
raise SystemExit('Unexpected repository archive')
|
||||
src=candidates[0].parent
|
||||
venv=root/'venv'
|
||||
subprocess.run(['python3','-B','-m','venv','--system-site-packages',str(venv)],check=True)
|
||||
python=str(venv/'bin/python')
|
||||
subprocess.run([python,'-B','-m','pip','install','--only-binary=:all:','--no-compile','pyarrow==26.0.1'],check=True)
|
||||
wheels=root/'wheels'
|
||||
subprocess.run([python,'-B','-m','pip','wheel','--no-deps','--wheel-dir',str(wheels),str(src)],check=True)
|
||||
wheel,=wheels.glob('paper_atlas-*.whl')
|
||||
subprocess.run([python,'-B','-m','pip','install','--no-deps',str(wheel)],check=True)
|
||||
outside=root/'outside';outside.mkdir()
|
||||
subprocess.run([python,'-I','-B','-m','paper_atlas','collect','reviewarena-metadata','--split',split,'--out',str(out),*extra],cwd=outside,check=True)
|
||||
report=json.loads((out/'collection-report.json').read_bytes())
|
||||
assert report['database_written'] is False and report['approved_for_import']==0
|
||||
receipt={'source_commit':sha,'source_archive_sha256':hashlib.sha256(archive).hexdigest(),'wheel_sha256':hashlib.sha256(wheel.read_bytes()).hexdigest(),'dataset_commit':report['commit'],'split':split,'records':report['records'],'metadata_bytes_read':report['metadata_bytes_read'],'cached_projection_report_sha256':report.get('cached_projection_report_sha256'),'raw_parquet_saved':False,'database_written':False}
|
||||
(out/'runner-receipt.json').write_text(json.dumps(receipt,indent=2)+'\n')
|
||||
backup=base/'validation'/(out.name+'-recovery.tar.gz')
|
||||
if backup.exists():
|
||||
raise SystemExit('Recovery archive already exists; inspect')
|
||||
with tarfile.open(backup,'w:gz') as tar:
|
||||
tar.add(out,arcname=str(out.relative_to(base)))
|
||||
backup.chmod(0o444)
|
||||
for path in out.iterdir():
|
||||
if path.is_file():path.chmod(0o444)
|
||||
result={'status':'metadata_projection_backed_up_not_imported','archive':str(backup),'bytes':backup.stat().st_size,'sha256':hashlib.sha256(backup.read_bytes()).hexdigest(),**receipt}
|
||||
(base/'validation'/(out.name+'-recovery.json')).write_text(json.dumps(result,indent=2)+'\n')
|
||||
print(json.dumps(result),flush=True)
|
||||
PY
|
||||
@@ -2,6 +2,8 @@
|
||||
|
||||
**长期维护、按需运行的论文元数据与公开投稿历史工具。**
|
||||
|
||||
其他会议的公开评审镜像可按 [ReviewArena 元数据投影流程](docs/reviewarena-metadata.md) 核查:仅取固定版本的标题、摘要、作者及决定等索引字段,保留镜像声明和稳定行定位,不下载正文/评审数据页,也不将镜像 ID 自动提升为官方论坛身份。
|
||||
|
||||
Paper Atlas 将 arXiv、AI 会议论文集及公开投稿决定整理到可追溯的数据库中,用于查询和后续研究分析。生产数据使用 PostgreSQL,支持从既有 SQLite 索引完整迁移;SQLite 保留离线核验和小型样本用途。它按批次运行,不需要常驻 Web 服务;采集标题、摘要、标识、标签及决定证据,不下载论文全文。
|
||||
|
||||
项目仍在建设中。候选范围包含 43 个会议(含 ACM MM、MLSys),尚未完成全部历史数据和最终去重复核。实际进度以数据库、来源清单和覆盖报告为准,而不是仓库文件数量。
|
||||
@@ -117,6 +119,8 @@ CI 在已授权的远程 runner 上运行测试、构建 wheel,并从源码目
|
||||
|
||||
已导入整卷的逐条轨道修正见 [edition-tracks 维护流程](docs/edition-tracks.md):按完整证据生成只读计划,保留原投稿和决定 ID,事务性修正 edition、标识与覆盖,并防止旧整卷刷新覆盖分类。
|
||||
|
||||
完整出版来源的新旧版本导致分轨或摘要变化时,使用 [source-revisions 流程](docs/source-revisions.md):核对同一 DOI 全集,保存完整修正前数据与版本链,支持已有记录的目标 track,按新版本重算分区覆盖数,并阻止旧版元数据刷新。普通增量导入仍禁止缩小预期数量。
|
||||
|
||||
入库前可使用通用的[来源身份只读预检](docs/source-identity-audit.md),导出候选论文、稳定标识和既有投稿决定证据;候选结果需要审核,不会自动批准导入。
|
||||
|
||||
RSS 2026 已独立适配官方论文集与录用表;使用方法和与 2025 demo 分类的区别见 [RSS 2026 来源说明](docs/rss2026-proceedings.md)。
|
||||
@@ -131,8 +135,20 @@ ICAPS 2026 也提供独立的 `collect icaps2026-registry`,保留 Crossref 卷
|
||||
|
||||
ECCV 2026 的[暂定节目采集器](docs/eccv2026-program.md)合并重复节目 ID,并以不可变缓存断点采集公开元数据;保留摘要原文、来源关键词及未分类 track,始终待身份审核,不自动入库。
|
||||
|
||||
`collect ismir2026-program` 核验 140 项 ISMIR 官方录用节目及原始摘要,保留两处标题、三处作者名单差异和 101 组自愿公开评审文本;不推断原始 OpenReview ID 或拒稿。见 [ISMIR 2026 使用与恢复说明](docs/ismir2026-program.md)。
|
||||
|
||||
`collect icassp2026-registry` 从冻结的 10 页 Crossref 元数据核验 4,582 条论文,明确排除 7 条非论文及 56 条其他会议记录;保留摘要缺失和单字段作者姓名,不自动融合节目表。见 [ICASSP 2026 采集与恢复说明](docs/icassp2026-registry.md)。
|
||||
|
||||
`collect sdm2026-proceedings` 核验 53 个 Crossref 论文章节,并单独保留 53 项日程观察及 52 页摘要。作者差异、取消展示和摘要缺失分别保留,不自动拼接来源或绑定作品;见 [SDM 2026 使用与恢复说明](docs/sdm2026-proceedings.md)。
|
||||
|
||||
`collect cikm2026-sources` 核验并分开导出两份官方节目、附属议程和出版登记。见 [CIKM 2026 来源采集与恢复](docs/cikm2026-sources.md)。
|
||||
|
||||
`collect icdm2026-program` 离线核验 324 条 ICDM2026 官方线下日程,按明确的研究/应用标签输出 241/83 条;展示长短、缺失摘要及未核实出版身份分别保留,不自动入库。见 [ICDM2026 使用与恢复说明](docs/icdm2026-program.md)。
|
||||
|
||||
`collect recsys2026-program` 核验九个 track 的 263 项官方录用贡献观察(260 个不同 DOI),保留摘要、原始作者和重复 DOI 的来源版本。见 [RecSys 2026 使用与恢复说明](docs/recsys2026-program.md)。
|
||||
|
||||
`collect mm2026-program` 从已冻结官方快照恢复全部 28 个来源 track 的 2,143 个节目 ID,保留 4,326 次展示的来源位置与场次上下文,单列 216 个摘要缺失项;输出供贡献类型、出版及作品身份核查使用,不是可直接导入的论文/决定队列。见 [MM 2026 使用、来源和恢复说明](docs/mm2026-source-status.md)。
|
||||
|
||||
`collect mm2026-publications` 核验完整论文集的 1,918 个注册出版 DOI,并单独关联节目分轨、展示上下文和摘要。1,743 份节目摘要明确标注来源(其中 26 份来自已复核的标题排版对应),166 项标题/作者版本差异保留为待核;所有节目及未匹配项均保留。注册文章类型不等于研究类型,输出尚需数据库身份/历史审查。见 [MM 2026 出版条目采集与恢复](docs/mm2026-publications.md)。
|
||||
|
||||
ECCV 2026 已归档 Springer 目录和章节可通过 [正式出版采集流程](docs/eccv2026-publications.md)整理为带 DOI 和可复核证据的 JSONL。
|
||||
|
||||
@@ -0,0 +1,22 @@
|
||||
# CIKM 2026 来源采集与恢复
|
||||
|
||||
使用固定节目和登记快照复现 CIKM 2026 来源盘点。采集命令生成四份独立 JSONL,输入逐文件校验 SHA-256。
|
||||
|
||||
```sh
|
||||
paper-atlas collect cikm2026-sources \
|
||||
--snapshots /data/paper-index/validation/cikm2026-source-snapshots-v1 \
|
||||
--out /data/paper-index/collected/cikm2026-sources-v1
|
||||
```
|
||||
|
||||
输入包括两版官方节目 HTML、Crossref 卷登记、五页逐篇登记及分页回执。服务器验证目录保留输入原文。输出目录必须是新目录。先检查 `collection-report.json`,报告在四份 JSONL 写完后发布。
|
||||
|
||||
| 文件 | 内容 |
|
||||
| --- | --- |
|
||||
| `programme.jsonl` | 带 DOI 的 1,008 条节目记录,包括标题、作者显示名、摘要和五类节目标签 |
|
||||
| `alternate-programme.jsonl` | 另一版节目 1,005 条记录及各自的字段 |
|
||||
| `supplemental-programme.jsonl` | 73 条演讲、教程、工作坊及其他附属活动观察 |
|
||||
| `crossref.jsonl` | 1,081 条 ACM 卷内登记,每条保留登记字段及分页文件位置 |
|
||||
|
||||
节目 DOI 指向 Crossref 的出版登记。确认作者和标题差异以及历史投稿绑定后,再为来源观察准备入库队列。报告另列 73 个未出现在带 DOI 节目区的登记,附属活动仍保留各自的活动类型。公开拒稿与撤回决定仍需独立来源核查。
|
||||
|
||||
官方原始来源见[带 DOI 的节目](https://cikm2026.diag.uniroma1.it/cikm-program/program.html)、[另一版节目](https://cikm2026.diag.uniroma1.it/cikm-program/CIKM2026_program.html)和 [Crossref 卷登记](https://api.crossref.org/works/10.1145/3799682)。
|
||||
+10
-2
@@ -1,6 +1,6 @@
|
||||
# 会议覆盖与公开决定核查
|
||||
|
||||
最新生产进展和剩余工作见 [2026-11-17 补漏状态](backfill-status-2026-11-17.md):固定 278 个会议年份槽位中,270 个有记录、8 个仍为空;这不代表完整覆盖,第 14 步仍未完成。
|
||||
最新生产验收与剩余工作见 [2026-11-17 补漏状态](backfill-status-2026-11-17.md):当日 `14:06:28 UTC` 固定范围快照中,278 个会议年份槽位有 272 个含记录、6 个仍为空;这不代表完整覆盖,第 14 步仍未完成。后续 MM 规范化、只读候选审计与出版核查不等于生产导入。
|
||||
|
||||
**当前工作范围:2020 年起主动补漏。** 2000 年起已发现资料保留,恢复工作从 [补漏范围与恢复清单(2026-11-10)](backfill-handoff-2026-11-10.md) 开始;下文旧日期快照不改写,较早年份排除不算补齐。
|
||||
|
||||
@@ -72,7 +72,7 @@ PGSERVICEFILE=/etc/paper-atlas/pg_service.conf paper-atlas coverage \
|
||||
|
||||
## 优先缺口
|
||||
|
||||
- **UAI 2024–2026**:官方均允许作者选择公开拒稿,当前主库拒稿仍为零。原 notes 接口要求 Challenge 验证,公开镜像尚无已核验来源,必须保留这个真实缺口;不能根据 PMLR 录用名单推断拒稿。参见 [2025 投稿指南](https://www.auai.org/uai2025/submission_instructions)、JSON 内其他年份官方证据。
|
||||
- **UAI 2024–2026**:官方均允许作者选择公开拒稿,首份生产快照拒稿为零。原 notes 接口要求 Challenge 验证;新核验镜像的 2022–2025 年 902 个论坛来自录用子集,不能填补公开拒稿缺口。2024 年的元评审也不能直接转成最终决定,见 [2026-11-17 原始来源与回复结构核查](uai-public-negative-source-status-2026-11-17.md)。不能根据 PMLR 或镜像录用名单反推拒稿;政策参见 [2025 投稿指南](https://www.auai.org/uai2025/submission_instructions)、JSON 内其他年份官方证据。
|
||||
- **EMNLP 2023**:官方允许公开 opt-in 拒稿,目前只导入 9 条。需继续核验实际公开全集。ARR 评审、conference commitment 与最终会议决定分开;[ARR 作者说明](https://aclrollingreview.org/authors) 明确其自愿捐赠数据与 EMNLP 2023 直接投稿发布的范围不同。
|
||||
- **ICLR**:固定快照已观察 2017 与 2020–2026;快照之后,2013/2018/2019 的 1,654 条来源记录已完成独立 PostgreSQL 验收(报告时间 `2026-11-07T19:52:28.378Z`),不回写旧快照计数,也不据此宣称逐届全集已齐。快照中的 10,208 次 unknown 不等于 rejected,近期补采与冲突核对仍在进行;评审推荐不能当最终决定。
|
||||
- **MM / MLSys**:MM 11,200 条书目记录保留 `proceedings_unspecified`;MLSys 的 448 次官方投稿与 313 条 DBLP 记录重叠。2026 MLSys 的研究/工业轨道在 [官方 CFP](https://mlsys.org/Conferences/2026/CallForResearchPapers) 中分开,仍须核对逐篇轨道证据,不能仅凭会议名全部归 main。
|
||||
@@ -189,6 +189,8 @@ ICAPS官方共同历史明确列出两个前身系列,现有DBLP映射只覆
|
||||
|
||||
MM 2026 的[官方 CFP](https://2026.acmmm.org/site/cfp-guidelines.html)说明会后公开最终评审及适用的作者回复,但未明确包含全部被拒投稿。2026-11-08 对官方组的匿名 API 查询返回200,对该组 `venueid` 的投稿查询返回403;[证据](mm-editions-2026-11-08.json)保存查询、时间和响应哈希。公开拒稿可用性仍为 unknown,403不能作为零记录或不存在公开拒稿的证据。
|
||||
|
||||
2026-11-17后续:完整节目JSON与按声明的负面venue ID进行的四条匿名403核查已冻结,见 [MM最新来源状态](../mm2026-source-status.md);没有据此关闭公开拒稿缺口。
|
||||
|
||||
WWW早期逐来源核查见 `www-early-proceedings-2026-11-08.json`:1995春季论文三目录均为38个相同条目ID,海报摘要53个条目与作者链接47个条目不齐;1994春季50个Title字段与主页49场报告说法待逐项核对。上述为源目录观察,不是已导入或全部录用数量;主会、海报、教程等继续独立核验。
|
||||
|
||||
WWW事件标识生产迁移已独立验收(报告SHA `f5dbb8a8`):保留1399个原届次字段及IDs、59个Fall投稿和完整全库历史哈希。服务器已缓存46个官方HTML来源(约223KB),只作为后续维护式采集器的输入;尚未补导早期WWW论文,也未启用可选注解表。
|
||||
@@ -198,3 +200,9 @@ IJCAI 联办届次的逐条后续核查见 [2018/2022 分区与目录差集](ijc
|
||||
2026年11月9日的[当前历史年度缺口核对](historical-gaps-2026-11-09.md)单列135个历史会议—年份、14个2026年度及85个经范围修正的初步候选来源年度;它使用当前生产edition/别名快照,保留原固定快照不变。年度有记录不等于逐篇、逐track或公开拒稿历史完整。
|
||||
|
||||
[早期NLP来源范围复核](historical-source-scope-2026-11-09.md)保留3份官方网页的响应哈希,排除4个COLING卷对ACL缺口的误匹配,并区分会议举办记录、目录可获取性与论文级覆盖;尚未导入或关闭这些年度缺口。
|
||||
|
||||
- [MM 2026 官方节目来源状态](../mm2026-source-status.md):完整官方日程 JSON、重复展示盘点、28 个来源 track、缺失作者/摘要和四条匿名决定路径的 403 证据;这些不是已导入或公开拒稿全集。
|
||||
|
||||
- [2026-11-18 公开负面来源核查](public-negative-source-audit-2026-11-18.md):ResearchArcade 930 条非 ICLR 声明的当前库比对、5 条未绑定声明及原始接口访问证据;不把镜像差集算作新增拒稿。
|
||||
|
||||
- [ECCV 2026 出版目录与节目单对账](eccv2026-publisher-crosswalk-2026-11-18.md)记录 Springer 78 卷的 DOI、节目单差异及 PostgreSQL 只读绑定核查。
|
||||
|
||||
@@ -1493,7 +1493,7 @@
|
||||
},
|
||||
"TMLR_current20261108": {
|
||||
"url": "https://www.jmlr.org/tmlr/faq.html",
|
||||
"finding": "Additionalnonconferencevenuealreadyinregistry:accepted/rejected/withdrawn/retractedallpublic; deskrejectedneverpublic. Authorshiddenonnotyetacceptedpapers; donotinferanonymousidentities. Continuousjournal; eventcertification/journal-to-conferencepresentationnotanotherindependentreviewacceptance."
|
||||
"finding": "Additionalnonconferencepolicyreference; TMLR is not in the current 43-conference registry:accepted/rejected/withdrawn/retractedallpublic; deskrejectedneverpublic. Authorshiddenonnotyetacceptedpapers; donotinferanonymousidentities. Continuousjournal; eventcertification/journal-to-conferencepresentationnotanotherindependentreviewacceptance."
|
||||
}
|
||||
},
|
||||
"public_negative_recovery_notes": {
|
||||
@@ -1507,6 +1507,12 @@
|
||||
"original_rejection_notes_obtained": false,
|
||||
"prior_decisions_inserted": 0,
|
||||
"requires": "Per-paper primary original identifiers/decisions or an explicitly distinguished policy-supported relation; never manufacture an original note."
|
||||
},
|
||||
"researcharcade_20261118": {
|
||||
"audit": "public-negative-source-audit-2026-11-18.json",
|
||||
"summary": "930 non-ICLR 2020+ mirror claims:925 scoped ID matches all have rejected status;5 held,3 with equal title+source abstract but distinct unverified note IDs. No writes or new rejections approved.",
|
||||
"access": "NeurIPS2021 and EMNLP2023 original note queries403; stopped, no bypass.",
|
||||
"next": "Seek independent primary evidence for note.id/forum relationships and final decisions; no title-only merge or duplicate-rejection import. No full-public-set coverage claim."
|
||||
}
|
||||
},
|
||||
"restore_rules": [
|
||||
|
||||
@@ -1,47 +1,50 @@
|
||||
# 2020 起补漏状态:2026-11-17
|
||||
|
||||
**总体仍为 13/17,第 14 步未完成。** 本页补充新的生产结果,不回写 [2026-11-10 恢复清单](backfill-handoff-2026-11-10.md) 或更早快照。用户确定的范围不变:2020 年起主动补漏;43 个更早历史缺口延期而非补齐,2000–2019 已发现的名单和身份线索继续保留。
|
||||
**总体仍为 13/17,第 14 步未完成。** 本页更新当日最新生产结果;[2026-11-10 恢复清单](backfill-handoff-2026-11-10.md)和更早快照继续保留。用户确定的范围不变:2020 年起主动补漏;43 个更早历史缺口延期而非补齐,2000–2019 已发现的名单和身份线索继续保留。
|
||||
|
||||
## 最新数据库盘点
|
||||
|
||||
只读快照 `2026-11-17 09:38:31.677743+00`,固定 43 个目标会议、278 个应有会议年份槽位,其中 270 个至少有一条投稿记录,8 个仍为空,均为 2026 年:**CIKM、ECCV、ICASSP、ICDM、IROS、ISMIR、MM、RecSys**。本次 SDM2026 新增记录使空槽从 9 减到 8。
|
||||
只读快照 `2026-11-17 14:06:28.803415+00`,固定 43 个目标会议、278 个应有会议年份槽位,其中 **272 个至少有一条投稿记录,6 个仍完全为空**,均为 2026 年:**CIKM、ECCV、ICDM、IROS、ISMIR、MM**。SDM25、ICASSP112、RecSys47 三批分别使空槽从 9 降至 8、7、6。
|
||||
|
||||
这只说明有无记录,**不表示 270 个槽位已经逐篇、逐 track、摘要或公开拒稿覆盖完整**;不能据此宣布第 14 步完成。2020–2025 没有整年空槽,仍有逐来源完整性和公开负面决定缺口。SDM2026 仅导入 25/53 条,状态为 partial,同样不能称为整届补齐。
|
||||
这些是会议—年份的存在性统计,**不表示 272 个槽位已逐篇、逐 track、摘要或公开拒稿覆盖完整**。2020–2025 没有整年空槽,仍有来源完整性和公开负面决定缺口。RecSys 九轨、ICASSP 和 SDM 的本批来源均仍为 partial;不能据此宣布第 14 步完成。
|
||||
|
||||
| 生产文件(相对 `/data/paper-index/`) | SHA-256 |
|
||||
| 生产文件(相对 `/data/paper-index/`) | SHA256 |
|
||||
| --- | --- |
|
||||
| collected/active-scope-census-20261117-v17/summary.json | bfb1db8b50619c3756abb91ad0f73fa58ff998c7db58a216661342f12047749f |
|
||||
| collected/active-scope-census-20261117-v17/snapshot.json | b5389205f8b34f7ed57946416f49c253e0aed4a05e45eb818801f18fca6a8cb4 |
|
||||
| collected/active-scope-census-20261117-v19/summary.json | cf0c5a013b6efe7d232187c9c76d9820bc8b33e2dd5674cab53d78cd3369ab4c |
|
||||
| collected/active-scope-census-20261117-v19/snapshot.json | 38d09e3782372edbdd2840f5922d65fb497c425b1eaf2d1fb10342956a7a82f2 |
|
||||
| validation/active-scope-fixed278-v1.json | a835bef5b9c3a0f76069d840cecdb9a6bcf65d3a38796d70dbfdfd4b49242264 |
|
||||
|
||||
## 已推进的两批来源
|
||||
原始快照、范围定义与生成脚本已备份,本机已按固定范围和别名逐条重算 272/278,未用 `max(id)` 推算数量。此前 v17/v18 快照保留,不能将其旧数字当作当前状态。
|
||||
|
||||
- **SDM2026**:53 个正式注册章节、53 个独立日程观察和 52 个公开摘要页已冻结并规范化。25 条发表记录已在 PostgreSQL 独立验收,新增论文实体为 0;28 条待身份/版本核查。完整旧历史和 5,653 篇受保护论文的资料、来源、标签和既有标识保留。见 [生产验收与恢复](../sdm2026-identity-review.md)。这批队列已经消费,禁止重放。
|
||||
- **ECCV2026**:2,834 个唯一初步日程条目及公开摘要已全量采集并独立验收;新的只读身份预检和 1,997 个候选历史证据包已保存,尚未入库。初步日程不等于已核验出版社最终论文集;负面决定元数据仍受匿名访问限制。见 [来源状态](../eccv2026-source-status.md)。
|
||||
## 已独立接受的近期补录
|
||||
|
||||
SDM 新增事件使 SeBA 和 AdamG 同时具有较早 ICLR 拒稿与后来的 SDM 发表事实。旧的两次 ICLR 投稿及四条决定来源观察仍保留,没有被较晚 accepted 覆盖;多个来源观察不直接累加成多次拒稿。
|
||||
- **RecSys2026**:官方九轨 263 项观察、260 个不同 DOI 均已冻结并规范化。47 条现有作品关联通过独立验收,新增论文实体为 0;新增一个来源、九个 edition、47 次投稿/47 条录用观察、47 个发表 DOI 投稿标识及47条来源类型标签证据。原始摘要单独保存,规范摘要不覆盖。5,812 篇受保护旧作品及全部旧投稿/决定历史保持不变,foundry 就绪检查200、SQLite备份只读。216项待审查,包括三组重复 DOI 的六项观察。见 [九轨验收与恢复](../recsys2026-program.md)。
|
||||
- **ICASSP2026**:Crossref 冻结的 4,645 个条目中排除 56 个其他会议条目和 7 个目录/封面等非论文条目,剩余 4,582 条注册论文;全部注册摘要缺失。112 条已有 DOI 的来源/录用历史已独立验收,新增论文实体为 0,coverage 为 partial112/4582;4,470 条仍待身份核查。正式节目标题表与注册元数据保持独立,不自动融合作者或摘要。见 [来源边界与恢复](../icassp2026-registry.md)。
|
||||
- **SDM2026**:53 个注册章节、53 项独立日程观察及 52 页摘要已冻结。25 条来源关联已独立接受,新增论文实体为 0,28 条待审查;见 [生产验收与恢复](../sdm2026-identity-review.md)。
|
||||
|
||||
## 其余七个会议的最新入口核查
|
||||
三批已经消费的输入队列均**禁止重放**。最新已接受生产结果为 RecSys47:论文实体仍为 **3,567,888**,来源记录 **3,814,931**,投稿 **531,620**,决定来源观察 **571,172**。观察数不等于不同论文或不同投稿的数量;后续须使用当前基线和新的增量队列。
|
||||
|
||||
已对下表七个会议的 14 个明确官方 URL 保存原始响应和回执:13 个 HTTP 200,IROS 官方论文索引 HTTP 403。所有请求均匿名,不绕过访问限制,没有下载论文 PDF;保存入口不是完成逐篇采集或负面决定普查。
|
||||
SDM 新增事件使 SeBA 和 AdamG 同时保有较早 ICLR 拒稿与后来 SDM 发表事实。原来的两次 ICLR 投稿及四条决定来源观察均保留,没有被 accepted 覆盖;四条观察不能直接累加成四次拒稿。本批 RecSys47 本身没有既往负面观察,但全部旧全库历史同样通过保护验收。
|
||||
|
||||
| 会议 | 已保存的明确官方入口 | 尚未完成 |
|
||||
## 已采集、尚待关联或规范化的来源
|
||||
|
||||
| 会议 | 已取得的来源与状态 | 下一步 |
|
||||
| --- | --- | --- |
|
||||
| ICDM | [详细日程](https://icdm2026.neu.edu.cn/DetailedSessions/list.htm)、[研究轨道政策](https://icdm2026.neu.edu.cn/CallforResearchTrack/list.htm)、[应用轨道政策](https://icdm2026.neu.edu.cn/CallforAppliedTrack/list.htm) | 官方页面称为线下展示列表;需要核对逐篇范围、研究/应用轨道及演讲长短标签,不能把展示长短自动等同于论文类型;摘要和正式出版身份仍待核查 |
|
||||
| ICASSP | [录用标题清单](https://cmsworkshops.com/ICASSP2026/papers/accepted_papers.php)、[作者指南](https://2026.ieeeicassp.org/author-guidelines/) | 取得逐篇完整作者/摘要/出版记录,区别会议投稿与期刊展示,不能只导入标题或从缺席推断拒稿 |
|
||||
| IROS | [官方论文索引](https://2026.ieee-iros.org/program/paper_index/) | 当前服务器返回 403,保留缺口并停止该入口;第三方浏览器页面不能替代独立核验的官方原文 |
|
||||
| ISMIR | [官网](https://ismir2026.ismir.net/)、[官方虚拟平台](https://ismir2026program.ismir.net/)、[作者指南](https://ismir2026.ismir.net/authors/author-guidelines) | 核查平台公开静态数据与正式开放论文集的逐篇映射;首页可读取不代表论文目录已经取得 |
|
||||
| CIKM | [录用页](https://cikm2026.diag.uniroma1.it/accepted-papers/)、[日程页](https://cikm2026.diag.uniroma1.it/program/) | 网页提取的录用页仍显示 TBD;需要检查原文引用的节目数据与独立正式出版目录,不能当成没有论文 |
|
||||
| RecSys | [官网](https://recsys.acm.org/recsys26/)、[投稿说明](https://recsys.acm.org/recsys26/call/) | 继续取得可核验的逐篇正式记录,并分开主轨、工业、复现、demo、challenge/workshop |
|
||||
| MM | [官网](https://2026.acmmm.org/) | 继续核对官方完整日程和出版目录,分开研究论文与 grand challenge、workshop 等;首页响应不证明覆盖完整 |
|
||||
| ECCV | 2,834 个不同初步节目 ID 及公开摘要已独立采集;1,997 个只读候选历史证据包已保存,尚未入库 | 初步日程不等于 Springer 最终论文集;继续作品/版本和出版核查。见 [来源状态](../eccv2026-source-status.md) |
|
||||
| ICDM | 324 项线下节目观察,明确研究241/应用83;演讲长168/短156,摘要均缺失;规范化和80个候选历史包已核验 | 展示长短不能当成论文类型;需审查作品与正式出版版本,尚未导入主库。见 [采集器说明](../icdm2026-program.md) |
|
||||
| ISMIR | 140 项节目和正式名单均已核验,140份摘要,101组自愿公开评审;两处标题、三处作者名单差异保留,78个候选历史包已核验 | 数字 miniconf ID 不是原始 OpenReview ID;评审意见不转成最终拒稿。继续作品身份/历史关联,尚未入库。采集器见 [PR #63](https://gitea.entroia.com/Zayden-Jung/paper-atlas/pulls/63) |
|
||||
| MM | 全28轨2143个节目ID/4326展示位置已规范化;814个只读候选历史证据包已核验,24篇候选涉及24次负面投稿/37条来源观察;67项原始arXiv DOI中66项注册元数据一致、1项404暂缓 | 不是2143篇已核验科研论文。继续逐轨贡献/出版依据与作品审查,0身份批准、0入库。ISBN查询与父XML均只返回父记录,不作逐篇清单。见 [来源、审计与恢复证据](../mm2026-source-status.md) |
|
||||
| CIKM | 官方录用/日程入口与五个分轨页面已保存;五个分轨链接实际是CFP/政策,不是逐篇录用名单 | 查找实际节目或出版社完整目录;TBD和没有静态表不能当作没有论文 |
|
||||
| IROS | 官方论文索引入口的原始HTTP403已保存 | 保留访问限制,不重试绕过;寻找可合法读取、能独立核验的出版元数据 |
|
||||
|
||||
冻结目录:`/data/paper-index/collected/remaining2026-official-leads-v1`。manifest SHA-256:`0e13313d94e3bff17f071d3a1384f6ed83ece156239fe5bc4633cc7a34fb8e04`。恢复包:`/data/paper-index/validation/remaining2026-official-leads-v1-recovery.tar.gz`,SHA-256:`0be8c84bb5c591e5afaaa5f95dd5334bf13a9be1671cb942e46e998f502495bb`。HTTP 成功仅是入口可读性,不能作为可导入记录数量。
|
||||
早期 14 个入口快照保持在 `/data/paper-index/collected/remaining2026-official-leads-v1`,后续新增原文另建冻结目录,不覆盖早期回执。HTTP200只证明入口可读,不直接作为可导入记录或全部论文数量。
|
||||
|
||||
## 第 14 步仍需做什么
|
||||
## 公开拒稿与后续工作
|
||||
|
||||
1. 取得并规范化其余可公开取得的完整逐篇来源,记录应有范围、排除项和版本;对缺失、受限或尚未公开的入口保存明确证据,不能伪造“已补齐”。优先推进当前可读的 ICDM 详细日程、ICASSP 清单及其他会议的正式目录。
|
||||
2. 完成来源与年度、track、摘要完整性的对应核查;已经有若干记录的年份也要看目录差集,不能只消除空槽。
|
||||
3. 继续近期和历史年度实际公开的 rejected、desk_rejected、withdrawn 来源核查,特别是 UAI2024–2026 的 opt-in 拒稿、EMNLP2023 的实际公开范围以及其他会议的逐年限定政策。公开评审和公开决定不同;没有找到不等于零条或未曾拒稿。
|
||||
4. 通过身份核查的记录再执行新快照守护、恢复备份和增量导入,保留较早决定。深层作品版本、跨会历史及类型/标签协调继续列为第 15 步;不能为减少空槽创建重复论文,也不能只盯住少量身份疑难项而停止其他来源采集。
|
||||
MM 官方群组元数据可匿名读取,但根据其给出的 submission invitation 和 rejected/withdrawn/desk-rejected venue ID,四次限1条的匿名 notes 核查均为403。查询、时间、原文和哈希已保留,没有认证重试或绕过权限。访问限制不等于零条负面记录;详见 [MM来源证据](../mm2026-source-status.md)。ISMIR公开的101组评审来自录用节目,不能记成101条拒稿。
|
||||
|
||||
第 16 步仍是所有导入结束后的独立全库去重与完整审计,第 17 步是最终长期运行/恢复和交付整理。上述任务均未因单批验收或 CI 成功而完成;总体 goal 继续保持 active。
|
||||
UAI新增固定镜像已核对2022–2025的902个原始论坛及3053个回复的元数据/字段结构。整理版均标录用,2022保存230条明确Accept决定;2024仅有评审、作者回复、元评审,2023/2025未保存回复,不能据此补拒稿。来源限制和恢复见 [逐年原始核查](uai-public-negative-source-status-2026-11-17.md)。这不是UAI没有公开拒稿的结论,缺口保持未补齐。
|
||||
|
||||
第14步仍需逐会议、年份、track完成公开拒稿/desk-reject/withdrawn来源核查和可得数据采集;继续核查UAI近年自愿公开范围、EMNLP实际覆盖范围及其余已列明缺口。不能从未见录用、评分、评审推荐、缺少摘要或接口不可访问制造负面决定。
|
||||
|
||||
之后仍需完成跨会作品/历次投稿与类型标签关联(第15步),所有导入结束后的独立全库去重、标识冲突、近似标题作者、摘要/约束/查询及覆盖审计(第16步),以及长期运行与恢复文档最终整理(第17步)。阶段性来源入库、测试通过、某年度有记录均不能替代这些验收。
|
||||
|
||||
@@ -0,0 +1,35 @@
|
||||
# 2020 起补漏状态:2026-11-18
|
||||
|
||||
总体仍为 **13/17,第 14 步未完成**。主动补漏从 2020 年起,2000–2019 年已发现线索仍保留,43 个早期历史缺口仍为延期,没有据此算作补齐。更早状态页保留为历史快照。
|
||||
|
||||
## 当前生产结果
|
||||
|
||||
ISMIR 两批累计31项已关联到现有论文,新增论文实体为0。第二批新增来源记录、独立投稿、录用观察各21条及173条来源标签证据;两批累计267条标签证据。覆盖 `partial31/140`,109项仍待核验。详见[来源说明](../ismir2026-program.md)、[首批验收](ismir2026-import10-2026-11-18.json)和[第二批逐条验收](ismir2026-import21-2026-11-18.json)。
|
||||
|
||||
最新精确计数:论文实体 **3,568,022**,来源记录 **3,815,534**,投稿 **532,223**,决定来源观察 **571,775**。实体数量不是已经完成最终科学工作去重的数量;同一次投稿可能有多条决定来源观察。
|
||||
|
||||
paper `2757497` 保留 ICLR2026 一次拒稿投稿下的两条来源观察,并新增 ISMIR2026 录用投稿;实际 CLI 查询显示已知录用,旧拒稿历史仍可查询。不将两条来源观察计作两次拒稿,也不将来源采集时间当作决定时间。
|
||||
|
||||
## 固定范围存在性盘点
|
||||
|
||||
只读快照 `2026-11-18 06:46:44.099675+00`,范围仍为 43 会、278 个目标年度槽位:**274 个至少有一条投稿,4 个完全为空**,均为 2026 年的 **CIKM、ECCV、ICDM、IROS**。首批ISMIR使空槽从5降至4;第二批只增加同届来源覆盖,空槽数不变。
|
||||
|
||||
这只统计“该年度至少有数据”,不是逐篇、逐 track、摘要或公开拒稿覆盖完整。2020–2025 没有整年空槽,仍有来源完整性和公开负面历史缺口。
|
||||
|
||||
|证据(相对服务器 `/data/paper-index/`)|SHA256|
|
||||
|---|---|
|
||||
|`collected/active-scope-census-20261118-v22/summary.json`|`ee4ec2e6d6eaed55ce3475e735bfba9bd94e512893cd3c3024035b48b744f475`|
|
||||
|`collected/active-scope-census-20261118-v22/snapshot.json`|`5aabc91217ad787946ba50d6a670296b1e3152a54f5ecaa0e3a9b3a952b63fa7`|
|
||||
|`validation/active-scope-fixed278-v1.json`|`a835bef5b9c3a0f76069d840cecdb9a6bcf65d3a38796d70dbfdfd4b49242264`|
|
||||
|
||||
本机使用固定分母、会议别名和逐 edition 快照独立重算 274/278;没有使用 `max(id)` 代替计数。
|
||||
|
||||
## 尚待完成
|
||||
|
||||
- ISMIR 仍有109项;62项无精确标题候选不等于新作品,6项作者部分匹配仍待审查。31项摘要逐字一致候选已分两批审查并消费;其余候选不能仅凭标题、作者或摘要检索结果算作已批准。
|
||||
- MM 已验收572项来源观察,注册清单1918项中仍有1346项待处理。已有134个暂建索引实体仍须后续身份审计;H2AL 来源及轨道纠正、版本化来源覆盖方案尚未用于生产,不能把不同版本轨道预期数相加当完整分母。
|
||||
- ECCV 的初步节目、ICDM 的研究/应用来源和缺失摘要仍需作品及出版核查;CIKM 缺实际逐篇目录;IROS 官方访问403仍保留限制,不重试绕过。各来源的采集、批准和导入状态必须分开。
|
||||
- 公开 rejected、desk-rejected、withdrawn 继续按会议、年份、track核查。ISMIR 的101组自愿公开评审来自录用节目,不是101条拒稿;UAI accepted-only 镜像与 EMNLP/NeurIPS 原始接口403均不能关闭拒稿缺口。近期镜像差集核查见 [PR #72](https://gitea.entroia.com/Zayden-Jung/paper-atlas/pulls/72)。
|
||||
- 后续仍需完成跨会作品与历史、独立类型和标签关系(第15步);全部导入结束后的全库去重、标识冲突、近似标题作者、摘要/约束/查询及覆盖审计(第16步);长期运行与恢复文档最终整理(第17步)。
|
||||
|
||||
已消费输入不得重放。下次从[ISMIR 当前增量父状态和恢复说明](../ismir2026-program.md)及 `work-state.json` 的本批已接受回执继续;保留旧 SQLite 只读备份,不因查询超时重新执行导入。
|
||||
File diff suppressed because it is too large
Load Diff
@@ -0,0 +1,16 @@
|
||||
# CIKM 2026 节目与出版登记核对
|
||||
|
||||
官方节目列有 1,008 项带 DOI 的条目,逐篇均有摘要。五个标签的数量分别为 Full Research 596、Short Research 234、Applied Research 86、Resources 46、Demo 46。另一份官方节目快照少三个节目编号,其余 1,005 个编号重合。两份原始快照和响应哈希保存在本机的 `cikm2026-official-detailed-program-v1` 工作目录。[带 DOI 的节目](https://cikm2026.diag.uniroma1.it/cikm-program/program.html) · [另一份官方节目](https://cikm2026.diag.uniroma1.it/cikm-program/CIKM2026_program.html)
|
||||
|
||||
正式卷 DOI 为 `10.1145/3799682`,ISBN 为 `9798400725395`。ISBN 查询只返回卷级登记。按卷名及出版日期分页取得 1,081 条登记,逐条核对卷 DOI 前缀、完整容器标题及来源类型。1,008 个节目 DOI 全部匹配,另有 73 条登记位于此前带 DOI 节目分区之外。附属议程核对结果见下文。该次 Crossref 响应中摘要字段均缺失。[ACM 出版条目](https://doi.org/10.1145/3799682.3840615) · [Crossref 卷登记](https://api.crossref.org/works/10.1145/3799682)
|
||||
|
||||
按已知排版标记、Unicode 和标点变化比较,同 DOI 的条目有 4 处标题差异、11 处有序作者列表差异。差异条目保留两份原文,等后续核对后再决定如何关联。当前 PostgreSQL 快照中,97 个节目 DOI 已绑定论文,304 个节目标题有候选,其中 95 个 DOI 绑定与标题候选重合。仍需核查已有作品与历次投稿历史,当前尚未写入。
|
||||
|
||||
[逐条差集、字段差异及证据哈希](cikm2026-source-reconciliation-2026-11-18.json)。先将出版登记与节目摘要作为各自来源采集,再核查两者的作品关系和轨道。公开拒稿、撤回及桌拒的可得性继续单独核查。
|
||||
|
||||
|
||||
## 附属议程与已有 DOI 的进一步核查
|
||||
|
||||
在官方节目页面的附属议程及主题演讲中,73 条登记有 72 条找到唯一规范标题对应,70 条同时通过有序作者比较。对应项包括 Industry Day 20 条、博士论坛 19 条、教程 14 条、工作坊条目 15 条、主题演讲 3 条及 AnalytiCup 1 条。工作坊条目记录该场活动对应的出版材料,独立 workshop 论文仍需各自目录。两条工作坊条目有作者写法差异,另有一条 LASS 标题带不同前后缀,三条均保留待核对。[官方完整节目](https://cikm2026.diag.uniroma1.it/cikm-program/program.html)
|
||||
|
||||
全卷 1,081 个 DOI 的 PostgreSQL 只读复查找到 98 个已有绑定,其中 97 个来自带 DOI 节目分区。89 项在现有论文、Crossref 和节目中有一致的规范标题、有序作者及明确节目标签,可优先准备来源关联。其余 9 项保留待核查。98 项的现有历史中无拒稿、桌拒或撤回记录,这一观察仅覆盖已有绑定的作品。此次仍未写入,入库时须重新确认绑定及保护既有历史。
|
||||
File diff suppressed because it is too large
Load Diff
@@ -0,0 +1,20 @@
|
||||
# ECCV 2026 出版目录与节目单对账
|
||||
|
||||
2026-11-18 核验了 Springer LNCS 17001–17078 的全部 78 卷。会议索引显示 **2,835 条**,逐卷目录合计也有 **2,835 条带 DOI 的章节**,另有 156 项卷首或卷末资料。出版方卷册简介及大会节目单均为 **2,834 条**。两个来源仍有逐条差异,汇总数量相差一条的原因待核。逐卷数字、目录页哈希和只读数据库回执见[机器可读清单](eccv2026-publisher-crosswalk-2026-11-18.json)。[Springer 会议索引](https://link.springer.com/conference/eccv) · [第一卷](https://link.springer.com/book/10.1007/978-3-032-36984-0) · [大会节目单](https://eccv.ecva.net/Conferences/2026/AcceptedPapers)
|
||||
|
||||
## 逐篇对账
|
||||
|
||||
|核对对象|唯一标题且完整作者顺序一致|唯一标题且目录所显示作者顺序一致|标题对得上、作者有差异|标题待核|
|
||||
|---|---:|---:|---:|---:|
|
||||
|全部 2,834 条节目记录|1,332|709|535|258|
|
||||
|首批已入库的 30 条节目记录|23|3|4|0|
|
||||
|本次只读核验通过的 248 条既有论文候选|173|72|2|1|
|
||||
|20 篇有既往拒稿、桌拒或撤回的优先论文|12|7|1|0|
|
||||
|
||||
目录作者栏有时只显示前六人。作者差异包括缩写、姓名文字变化与排列变化。原始名单、顺序和来源保留,前缀匹配与作者差异需查章节作者页。节目单与出版目录的摘要、分轨及公开负面决定覆盖仍待分别核查。
|
||||
|
||||
## 当前库核对
|
||||
|
||||
只读核对了全部 2,835 个出版 DOI,68 个已有论文绑定。对账后另检查 271 个已知论文关联,涵盖 245 个本次候选和 26 个已入库节目记录。其中 14 个 DOI 已指向同一论文,257 个 DOI 尚未绑定。245 个候选内有 10 篇带既往负面决定。
|
||||
|
||||
下一批准备以出版 DOI、正式分轨和完整作者证据复核关联。首批 30 条节目记录保留原投稿、决定和来源。确认同届同一投稿后,再补充出版证据,避免重复创建 ECCV 投稿。本次核查未写入数据库。
|
||||
@@ -0,0 +1,339 @@
|
||||
{
|
||||
"sources": {
|
||||
"poster.html": {
|
||||
"url": "https://www3.cs.stonybrook.edu/~icdm2026/PosterwDemo/list.htm",
|
||||
"final_url": "https://www3.cs.stonybrook.edu/~icdm2026/PosterwDemo/list.htm",
|
||||
"http_status": 200,
|
||||
"bytes": 132977,
|
||||
"sha256": "b02083ca7574a188fe876851a700a12c131c262353db557bd43a8320fad5f747",
|
||||
"at": "2026-11-18T07:06:11.984184+00:00"
|
||||
},
|
||||
"schedule.html": {
|
||||
"url": "https://icdm2026.neu.edu.cn/DetailedSessions/list.htm",
|
||||
"final_url": "https://icdm2026.neu.edu.cn/DetailedSessions/list.htm",
|
||||
"http_status": 200,
|
||||
"bytes": 213830,
|
||||
"sha256": "c1294b89bae8c2dfcf652d8a25556e733ef42dc3ac3387070abe08419b961b0f",
|
||||
"at": "2026-11-18T07:06:12.349258+00:00"
|
||||
},
|
||||
"deepsync.html": {
|
||||
"url": "https://dm.cs.univie.ac.at/news/story/deepsync-paper-accepted-at-icdm-2026",
|
||||
"final_url": "https://dm.cs.univie.ac.at/news/story/deepsync-paper-accepted-at-icdm-2026",
|
||||
"http_status": 200,
|
||||
"bytes": 90333,
|
||||
"sha256": "e243196943434225acec2841b53c77a2b1459184e7c185e1e1cbb055118fe282",
|
||||
"at": "2026-11-18T07:06:20.561308+00:00"
|
||||
}
|
||||
},
|
||||
"poster_count": 343,
|
||||
"poster_groups": {
|
||||
"long-posters": 177,
|
||||
"short-posters": 166
|
||||
},
|
||||
"oral_count": 324,
|
||||
"shared_count": 324,
|
||||
"poster_only": {
|
||||
"DM1022": {
|
||||
"group": "short-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area C",
|
||||
"slot": "P2-3C-085",
|
||||
"session": "Applied Track: Real-world AI in finance, healthcare and industry II",
|
||||
"title": "Beyond Bidding Agents: A Dual-Role Multi-Agent Framework with Dynamic Reserve Pricing for Auto-Bidding Auctions",
|
||||
"line": 683
|
||||
},
|
||||
"DM1189": {
|
||||
"group": "long-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area A",
|
||||
"slot": "P1-3A-032",
|
||||
"session": "Heterogeneous data mining: spatiotemporal and representation learning",
|
||||
"title": "Traj-Transformer: Diffusion Models with Transformer for GPS Trajectory Generation",
|
||||
"line": 456
|
||||
},
|
||||
"DM1234": {
|
||||
"group": "long-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area A",
|
||||
"slot": "P1-2A-007",
|
||||
"session": "Foundations, causality and learning theory",
|
||||
"title": "Prediction-Powered Corpus Mining: Coverage-Valid Statistics from Scarce Labels",
|
||||
"line": 354
|
||||
},
|
||||
"DM1246": {
|
||||
"group": "short-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-3B-074",
|
||||
"session": "LLM-driven data mining and novel applications II",
|
||||
"title": "Feature Selection for Trajectory Inference",
|
||||
"line": 672
|
||||
},
|
||||
"DM1345": {
|
||||
"group": "short-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-3B-060",
|
||||
"session": "LLM-driven data mining and heterogeneous learning",
|
||||
"title": "TRACER: An LLM-based Coarse-to-Fine Framework for Adaptive Network Traffic Classification",
|
||||
"line": 658
|
||||
},
|
||||
"DM1408": {
|
||||
"group": "long-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P1-2B-043",
|
||||
"session": "Applied Track: Industrial, financial and decision intelligence",
|
||||
"title": "A Generation-to-Application Framework for Semantic ID Collision in Industrial Generative Recommendation",
|
||||
"line": 390
|
||||
},
|
||||
"DM1644": {
|
||||
"group": "long-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P1-2B-072",
|
||||
"session": "Recommendation, adaptation and spatiotemporal modeling II",
|
||||
"title": "LQ-Rec: Late Quantization for Sequential Recommendation",
|
||||
"line": 419
|
||||
},
|
||||
"DM1751": {
|
||||
"group": "short-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area A",
|
||||
"slot": "P2-3A-014",
|
||||
"session": "ML/DL and statistical methods",
|
||||
"title": "MissGNN: Missingness-Guided Graph Neural Network for Feature- and Sample-Aware Imputation",
|
||||
"line": 612
|
||||
},
|
||||
"DM1995": {
|
||||
"group": "long-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P1-2B-053",
|
||||
"session": "Applied Track: Recommendation, security and operational intelligence II",
|
||||
"title": "Multi-Modal Causal Disentanglement for Multi-Domain Sequential Audio Recommendation",
|
||||
"line": 400
|
||||
},
|
||||
"DM2002": {
|
||||
"group": "long-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area B",
|
||||
"slot": "P1-3B-048",
|
||||
"session": "Heterogeneous data mining and systems",
|
||||
"title": "HiMamba: A Hierarchical State Space Framework for Asymmetric Multimodal Data Mining in Clinical Assessment",
|
||||
"line": 472
|
||||
},
|
||||
"DM2055": {
|
||||
"group": "short-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area A",
|
||||
"slot": "P2-3A-038",
|
||||
"session": "Applied Track: Real-world AI in finance, healthcare and industry I",
|
||||
"title": "CGSBR: Adaptive Cross-Source Gating for Cold-Start Sequential Bundle Recommendation",
|
||||
"line": 636
|
||||
},
|
||||
"DM2140": {
|
||||
"group": "long-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area A",
|
||||
"slot": "P1-3A-008",
|
||||
"session": "Machine learning and statistical methods",
|
||||
"title": "Reliability-Guided Pseudo-Label Learning and Prototype Alignment for Domain-Adaptive Colorectal Segmentation",
|
||||
"line": 432
|
||||
},
|
||||
"DM252": {
|
||||
"group": "short-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-2B-046",
|
||||
"session": "LLM-driven data mining and novel applications I",
|
||||
"title": "Trust-But-Verify: Poisoning-Resilient Locally Private Graph Learning Protocols",
|
||||
"line": 572
|
||||
},
|
||||
"DM383": {
|
||||
"group": "long-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area A",
|
||||
"slot": "P1-2A-022",
|
||||
"session": "Heterogeneous graphs and multimodal intelligence",
|
||||
"title": "Subgraph Meta-Learning for Direction and Time Aware Transaction Anomaly Detection",
|
||||
"line": 369
|
||||
},
|
||||
"DM463": {
|
||||
"group": "short-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-2B-050",
|
||||
"session": "Modeling, recommendation and LLM methods",
|
||||
"title": "CaST-POI: Candidate-Conditioned Spatiotemporal Ranking for Next POI Recommendation",
|
||||
"line": 576
|
||||
},
|
||||
"DM483": {
|
||||
"group": "short-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-2B-051",
|
||||
"session": "Modeling, recommendation and LLM methods",
|
||||
"title": "Causal Masking Lowers Intrinsic Dimension in Full-Vocabulary Sequential Recommendation: Evidence from a Single-Variable Attention Ablation",
|
||||
"line": 577
|
||||
},
|
||||
"DM713": {
|
||||
"group": "short-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-2B-060",
|
||||
"session": "Modeling, recommendation and LLM methods",
|
||||
"title": "Hint-Conditioned Knowledge Tracing: A Causal-Inspired View of Context-Dependent Hint Usage",
|
||||
"line": 586
|
||||
},
|
||||
"DM736": {
|
||||
"group": "long-posters",
|
||||
"floor": "3F",
|
||||
"area": "Area B",
|
||||
"slot": "P1-3B-050",
|
||||
"session": "LLM-driven data mining and retrieval I",
|
||||
"title": "From Natural-Language Hypotheses to Verified Data-Mining Discoveries",
|
||||
"line": 474
|
||||
},
|
||||
"DM803": {
|
||||
"group": "short-posters",
|
||||
"floor": "2F",
|
||||
"area": "Area B",
|
||||
"slot": "P2-2B-032",
|
||||
"session": "Heterogeneous data mining and multimodal learning I",
|
||||
"title": "Graph-Guided LLM Reasoning for Incident Prediction Over Multi-Source Data Streams",
|
||||
"line": 558
|
||||
}
|
||||
},
|
||||
"oral_only": [],
|
||||
"title_differences": [
|
||||
{
|
||||
"paper_id": "DM1319",
|
||||
"poster_title": "FFE-FRCN: A Frequency Feature Enhanced Few-shot Learning Approach for Object Detection in Microscopic Images",
|
||||
"oral_title": "An FFE-FRCN-Based Data Platform for Environmental Microorganism Image Retrieval and Knowledge Platform"
|
||||
}
|
||||
],
|
||||
"nonpaper_table_rows": [],
|
||||
"database_written": false,
|
||||
"scope": "Official presentation metadata. Shared programme IDs identify repeated appearances within this event. Poster-only IDs are scheduling observations pending identity and publication review.",
|
||||
"pg_readonly": {
|
||||
"snapshot": {
|
||||
"at": "2026-11-18 07:08:21.138511+00",
|
||||
"snapshot": "1095530:1095530:"
|
||||
},
|
||||
"report_sha256": "04ad5547342d7e3c2939046f290ecff647878c202c867e47acbdc438d4e10ad4",
|
||||
"existing_icdm2026_editions": [],
|
||||
"existing_icdm_program_source": false,
|
||||
"reviewed_links": [
|
||||
{
|
||||
"external_id": "2026/DM1125",
|
||||
"paper_id": 2783475,
|
||||
"track": "research",
|
||||
"source_row_sha256": "4e981605cf471a5ea29bb3ba3a676784cea55f7777a681eb5fe267ff9c83fa4a",
|
||||
"source_abstract": null,
|
||||
"work_anchor": {
|
||||
"scheme": "arxiv",
|
||||
"value": "2608.29635"
|
||||
},
|
||||
"independent_source": "https://arxiv.org/abs/2608.29635",
|
||||
"existing_history": [
|
||||
{
|
||||
"id": 132984,
|
||||
"venue": "ICLR",
|
||||
"year": 2026,
|
||||
"observations": [
|
||||
{
|
||||
"id": 134279,
|
||||
"status": "rejected"
|
||||
},
|
||||
{
|
||||
"id": 539402,
|
||||
"status": "rejected"
|
||||
}
|
||||
]
|
||||
}
|
||||
],
|
||||
"original_icdm_decision_verified": false,
|
||||
"imported": false
|
||||
},
|
||||
{
|
||||
"external_id": "2026/DM790",
|
||||
"paper_id": 2485633,
|
||||
"track": "applied",
|
||||
"source_row_sha256": "14e03ca3833b267d6f677325f4e4b9295d1fa42cc77be6707b9bca745444888b",
|
||||
"source_abstract": null,
|
||||
"work_anchor": {
|
||||
"scheme": "arxiv",
|
||||
"value": "2509.19774"
|
||||
},
|
||||
"independent_source": "https://arxiv.org/abs/2509.19774",
|
||||
"existing_history": [
|
||||
{
|
||||
"id": 138980,
|
||||
"venue": "ICLR",
|
||||
"year": 2026,
|
||||
"observations": [
|
||||
{
|
||||
"id": 545219,
|
||||
"status": "withdrawn"
|
||||
}
|
||||
]
|
||||
}
|
||||
],
|
||||
"original_icdm_decision_verified": false,
|
||||
"imported": false
|
||||
},
|
||||
{
|
||||
"external_id": "2026/DM1157",
|
||||
"paper_id": 3286364,
|
||||
"track": "research",
|
||||
"source_row_sha256": "22ed027f29a161bc978b3c33e921b42c67d999d520eed49ec46ce5935a6f2f2a",
|
||||
"source_abstract": null,
|
||||
"work_anchor": {
|
||||
"scheme": "openreview_snapshot",
|
||||
"value": "2026/1dGzRTSLqL"
|
||||
},
|
||||
"independent_source": "https://dm.cs.univie.ac.at/news/story/deepsync-paper-accepted-at-icdm-2026",
|
||||
"existing_history": [
|
||||
{
|
||||
"id": 127709,
|
||||
"venue": "ICLR",
|
||||
"year": 2026,
|
||||
"observations": [
|
||||
{
|
||||
"id": 129094,
|
||||
"status": "rejected"
|
||||
},
|
||||
{
|
||||
"id": 534293,
|
||||
"status": "rejected"
|
||||
}
|
||||
]
|
||||
}
|
||||
],
|
||||
"original_icdm_decision_verified": false,
|
||||
"imported": false
|
||||
}
|
||||
],
|
||||
"held": [
|
||||
{
|
||||
"external_id": "2026/DM1791",
|
||||
"paper_id": 3288747,
|
||||
"reason": "DomainBridging has name-order/format differences and no arxiv self-accepted link; token-bag retrieval is not author or work identity approval."
|
||||
},
|
||||
{
|
||||
"external_id": "2026/DM1801",
|
||||
"paper_id": 3288866,
|
||||
"reason": "LightGFC author lists substantially differ; hold version/work review, no automatic author aliases or merge."
|
||||
},
|
||||
{
|
||||
"external_id": "2026/DM1413",
|
||||
"paper_id": 3292171,
|
||||
"reason": "OFS author lists differ including JianchengLv vs JindiLv; not treated as aliases; hold independent version/work verification."
|
||||
}
|
||||
]
|
||||
},
|
||||
"publisher_catalog_status": "No 2026 IEEE publisher catalogue verified in this check",
|
||||
"public_negative_availability": "unverified",
|
||||
"next": "Use the existing oral source denominator by track. Compare the three reviewed candidates with the fresh import baseline and preserve their earlier attempts. Keep the poster-only rows as a separate source inventory, and resolve the title disagreement before linking that poster."
|
||||
}
|
||||
@@ -0,0 +1,204 @@
|
||||
{
|
||||
"checked_on": "2026-11-18",
|
||||
"venue": "IROS",
|
||||
"year": 2026,
|
||||
"conference_program": {
|
||||
"url": "https://2026.ieee-iros.org/program/paper_index/",
|
||||
"status": "unavailable_in_checked_environment",
|
||||
"paper_level_full_index_verified": false
|
||||
},
|
||||
"publisher_registration_check": {
|
||||
"status": "no_target_container_in_first_25_relevance_results",
|
||||
"rows_checked": 25,
|
||||
"observed_container": "2026 IEEE 13th International Conference on Intelligent Systems (IS)",
|
||||
"complete_registry_search": false
|
||||
},
|
||||
"journal_presentations": {
|
||||
"journal": "IEEE Transactions on Robotics",
|
||||
"scope": "IROS 2026 conference presentation of journal papers",
|
||||
"rendered_index_and_embedded_entries_match": true,
|
||||
"count": 30,
|
||||
"xplore_document_ids": [
|
||||
"10989573",
|
||||
"11251175",
|
||||
"11488837",
|
||||
"11268969",
|
||||
"11098164",
|
||||
"11219220",
|
||||
"11304178",
|
||||
"11267071",
|
||||
"11304172",
|
||||
"11455384",
|
||||
"11397863",
|
||||
"11288096",
|
||||
"11544114",
|
||||
"11534108",
|
||||
"11488912",
|
||||
"11535315",
|
||||
"11342351",
|
||||
"11353077",
|
||||
"11429535",
|
||||
"11364144",
|
||||
"11427341",
|
||||
"11427344",
|
||||
"11455389",
|
||||
"11488904",
|
||||
"11535317",
|
||||
"11554436",
|
||||
"11495270",
|
||||
"11575122",
|
||||
"11488843",
|
||||
"11581355"
|
||||
],
|
||||
"conference_submission_decision_verified": false,
|
||||
"publication_dois_verified": false
|
||||
},
|
||||
"author_reported_conference_acceptance_candidates": [
|
||||
{
|
||||
"arxiv": "2607.06499",
|
||||
"acceptance_evidence": [
|
||||
"uw-roundup.html",
|
||||
"uw-faculty.html"
|
||||
],
|
||||
"author_count_current": 3,
|
||||
"work_identity_status": "existing_work_binding_verified_read_only",
|
||||
"database_paper_id": 2731624,
|
||||
"ordered_authors_match": true,
|
||||
"arxiv_versions_stored": [
|
||||
"v1"
|
||||
]
|
||||
},
|
||||
{
|
||||
"arxiv": "2505.04999",
|
||||
"acceptance_evidence": [
|
||||
"usc-acceptance.html"
|
||||
],
|
||||
"author_list_evidence": "usc-clam-project.html",
|
||||
"author_count_current": 7,
|
||||
"work_identity_status": "existing_work_binding_verified_read_only",
|
||||
"database_paper_id": 2381843,
|
||||
"ordered_authors_match": true,
|
||||
"arxiv_versions_stored": [
|
||||
"v1",
|
||||
"v2"
|
||||
]
|
||||
},
|
||||
{
|
||||
"acceptance_evidence": [
|
||||
"ntt-contributed.html"
|
||||
],
|
||||
"author_count_current": 6,
|
||||
"work_identity_status": "stable_identifier_and_track_pending"
|
||||
}
|
||||
],
|
||||
"purdue_scope_note": {
|
||||
"evidence": "purdue-mars.html",
|
||||
"reported_combined_count": 10,
|
||||
"reported_tro_subset": 2
|
||||
},
|
||||
"public_rejected_desk_rejected_withdrawn": "not_located_in_checked_public_sources",
|
||||
"coverage_status": "partial_source_audit",
|
||||
"database_written": false,
|
||||
"receipts": [
|
||||
{
|
||||
"name": "ieee-tro-list.html",
|
||||
"url": "https://www.ieee-ras.org/publications/t-ro/t-ro-list-of-conference-papers/",
|
||||
"final_url": "https://www.ieee-ras.org/publications/t-ro/t-ro-list-of-conference-papers/",
|
||||
"status": 200,
|
||||
"bytes": 610713,
|
||||
"sha256": "2f2bb466446c62c37dc0927f8df45b161a096cbf8ab1aeffc4f43fa5d27e6358"
|
||||
},
|
||||
{
|
||||
"name": "ieee-tro-policy.html",
|
||||
"url": "https://www.ieee-ras.org/publications/t-ro/",
|
||||
"final_url": "https://www.ieee-ras.org/publications/t-ro/",
|
||||
"status": 200,
|
||||
"bytes": 566001,
|
||||
"sha256": "37b8877babb62e69d61e8b32193378afad4372117025a139cf70ffc26fcada5c"
|
||||
},
|
||||
{
|
||||
"name": "ntt-contributed.html",
|
||||
"url": "https://group.ntt/en/topics/2026/09/28/iros2026.html",
|
||||
"final_url": "https://group.ntt/en/topics/2026/09/28/iros2026.html",
|
||||
"status": 200,
|
||||
"bytes": 221293,
|
||||
"sha256": "ade0a0320a83e1d4504ff5da24c9f495aed581cffa01aa38b1e4cbec331f6dbc"
|
||||
},
|
||||
{
|
||||
"name": "crossref-proceedings-query.json",
|
||||
"url": "https://api.crossref.org/works?query.container-title=2026+IEEE%2FRSJ+International+Conference+on+Intelligent+Robots+and+Systems+%28IROS%29&filter=prefix%3A10.1109%2Ctype%3Aproceedings-article%2Cfrom-pub-date%3A2026-01-01%2Cuntil-pub-date%3A2026-12-31&rows=25&select=DOI%2Ctitle%2Cevent%2Ccontainer-title%2Cpublisher%2Cpublished%2CURL",
|
||||
"final_url": "https://api.crossref.org/works?query.container-title=2026+IEEE%2FRSJ+International+Conference+on+Intelligent+Robots+and+Systems+%28IROS%29&filter=prefix%3A10.1109%2Ctype%3Aproceedings-article%2Cfrom-pub-date%3A2026-01-01%2Cuntil-pub-date%3A2026-12-31&rows=25&select=DOI%2Ctitle%2Cevent%2Ccontainer-title%2Cpublisher%2Cpublished%2CURL",
|
||||
"status": 200,
|
||||
"bytes": 13634,
|
||||
"sha256": "415bf105de8006a42d3f2f8e6eb0078eb54c51bd65315296c0735e853431b228"
|
||||
},
|
||||
{
|
||||
"name": "purdue-mars.html",
|
||||
"url": "https://engineering.purdue.edu/IE/news/2026/she-iros-mars-lab",
|
||||
"final_url": "https://engineering.purdue.edu/IE/news/2026/she-iros-mars-lab",
|
||||
"status": 200,
|
||||
"bytes": 60386,
|
||||
"sha256": "fae6ba9637ea141ff3fda93643b97ddfb65d6ea3689f1bca3580956272b2d373"
|
||||
},
|
||||
{
|
||||
"name": "uw-acceptance.html",
|
||||
"url": "https://faculty.washington.edu/chx/post/2026-06-17-iros-mppi/",
|
||||
"status": 404,
|
||||
"error": "HTTP Error 404: "
|
||||
},
|
||||
{
|
||||
"name": "usc-acceptance.html",
|
||||
"url": "https://liralab.usc.edu/news/",
|
||||
"final_url": "https://liralab.usc.edu/news/",
|
||||
"status": 200,
|
||||
"bytes": 25381,
|
||||
"sha256": "9fa6b272b7995448a4318dd91cc94b93eeb514ddaa92d7df6880b3d93d91fceb"
|
||||
},
|
||||
{
|
||||
"name": "arxiv-2607.06499.html",
|
||||
"url": "https://arxiv.org/abs/2607.06499",
|
||||
"final_url": "https://arxiv.org/abs/2607.06499",
|
||||
"status": 200,
|
||||
"bytes": 43362,
|
||||
"sha256": "98fcbcaf39832f3002fa2910f69ca650481bb473cdb450290e5f9ca9245646a2"
|
||||
},
|
||||
{
|
||||
"name": "arxiv-2505.04999.html",
|
||||
"url": "https://arxiv.org/abs/2505.04999",
|
||||
"final_url": "https://arxiv.org/abs/2505.04999",
|
||||
"status": 200,
|
||||
"bytes": 42506,
|
||||
"sha256": "26ea0fd9ac8753b911ff97f1aaf29ea495f50b5c1e106b33c1336ea87c03b0df"
|
||||
},
|
||||
{
|
||||
"name": "uw-faculty.html",
|
||||
"url": "https://www.me.washington.edu/facultyfinder/xu-chen",
|
||||
"final_url": "https://www.me.washington.edu/facultyfinder/xu-chen",
|
||||
"status": 200,
|
||||
"bytes": 99274,
|
||||
"sha256": "877af368b564ba26a62b1ce4f947b439cceada2490dfb446b8730ba132c72c78"
|
||||
},
|
||||
{
|
||||
"name": "uw-roundup.html",
|
||||
"url": "https://faculty.washington.edu/chx/post/2026-09-07-new-papers-2026/",
|
||||
"final_url": "https://faculty.washington.edu/chx/post/2026-09-07-new-papers-2026/",
|
||||
"status": 200,
|
||||
"bytes": 180954,
|
||||
"sha256": "bc9ef19a07ea2875f8cfffe9afd62bcc2998fbe7c4bc8a4440c1bc87bae3856b"
|
||||
},
|
||||
{
|
||||
"name": "usc-clam-project.html",
|
||||
"url": "https://clamrobot.github.io/",
|
||||
"final_url": "https://clamrobot.github.io/",
|
||||
"status": 200,
|
||||
"bytes": 20098,
|
||||
"sha256": "3b3ea3f817ce6483807ace9b5a42ccf9835b811837346b63411190316fe2819c"
|
||||
}
|
||||
],
|
||||
"work_identity_review": {
|
||||
"snapshot_at": "2026-11-18 09:26:04.582744+00",
|
||||
"evidence_sha256": "5e5cfb57525a369e8c81e1a9ea6edeb4f5e83a33e1961eeead8a014e3d59ad6d",
|
||||
"existing_work_matches": 2,
|
||||
"database_written": false
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,21 @@
|
||||
# IROS 2026 来源核查
|
||||
|
||||
2026-11-18 核查了会议节目、期刊报告名单、作者机构公布的录用信息和 DOI 登记线索。逐项来源、响应哈希与文献定位符见[机器可读清单](iros2026-source-audit-2026-11-18.json)。本次来源核查未写入数据库。
|
||||
|
||||
## 期刊论文在会上报告
|
||||
|
||||
IEEE RAS 的 T-RO 名单在 IROS 2026 一节列出 30 个不同的 IEEE Xplore 文献编号。页面渲染内容和内嵌数据的编号及顺序完全一致。T-RO 政策允许已录用的期刊文章到 IROS 报告。后续应保留期刊出版与会议报告两种关系,并在确认期刊 DOI 后核对现有论文身份。本清单没有逐条验证 DOI 或会议投稿决定。[T-RO 报告名单](https://www.ieee-ras.org/publications/t-ro/t-ro-list-of-conference-papers/) · [T-RO 报告政策](https://www.ieee-ras.org/publications/t-ro/)
|
||||
|
||||
普渡大学的通告将团队的 10 项成果一同介绍,其中 2 项属于 T-RO。这个团队样本可以帮助核查来源口径,暂不用于生成逐篇录用记录。[普渡大学通告](https://engineering.purdue.edu/IE/news/2026/she-iros-mars-lab)
|
||||
|
||||
## 作者机构公布的录用线索
|
||||
|
||||
- 华盛顿大学的作者通告列出 IROS 2026 待出版论文,标题与 arXiv 2607.06499 对应,三位作者一致。只读查询确认对应论文 2731624,库内标题与三位作者的顺序一致。[作者通告](https://faculty.washington.edu/chx/post/2026-09-07-new-papers-2026/) · [arXiv](https://arxiv.org/abs/2607.06499)
|
||||
- USC 实验室确认 CLAM 在 IROS 2026 录用。项目页指向 arXiv 2505.04999,当前作者名单为七人。库内 arXiv 保留 v1、v2,现有论文 2381843 的标题和七位作者顺序与当前详情页一致。[实验室通告](https://liralab.usc.edu/news/) · [作者项目页](https://clamrobot.github.io/) · [arXiv](https://arxiv.org/abs/2505.04999)
|
||||
- NTT 公布一条题名及六位作者,待核实稳定标识和会议 track。[NTT 通告](https://group.ntt/en/topics/2026/09/28/iros2026.html)
|
||||
|
||||
这两条工作身份已核实,尚未录入 IROS 录用证据。
|
||||
|
||||
## 待补证据
|
||||
|
||||
[会议论文索引](https://2026.ieee-iros.org/program/paper_index/)在受检环境未能读取。一次限定 IEEE、2026 年及会议论文类型的 Crossref 相关性查询返回的前 25 条属于另一会议。实际会议论文集 DOI、完整出版清单、正式分轨和逐篇覆盖仍待核查。已检查的公开来源中未找到逐篇拒稿、桌拒或撤回清单,继续保持这些决定的公开可得性待查。
|
||||
@@ -0,0 +1,247 @@
|
||||
{
|
||||
"as_of": "2026-11-18",
|
||||
"goal_progress": "13/17; stage14 incomplete",
|
||||
"source": "ismir_program",
|
||||
"scope": "First10 reviewed existing-work source links, partial10/140, not public-rejection completeness or final deduplication",
|
||||
"database_acceptance": {
|
||||
"status": "passed",
|
||||
"at": "2026-11-18 05:15:27.565655+00:00",
|
||||
"writer_commit": "12f95cff91813a2143bc8fab5bc3e89df1bd0fdb",
|
||||
"before_counts": {
|
||||
"papers": {
|
||||
"rows": 3568022,
|
||||
"max_id": 3568595
|
||||
},
|
||||
"source_records": {
|
||||
"rows": 3815503,
|
||||
"max_id": 3815504
|
||||
},
|
||||
"submissions": {
|
||||
"rows": 532192,
|
||||
"max_id": 532193
|
||||
},
|
||||
"decisions": {
|
||||
"rows": 571744,
|
||||
"max_id": 571744
|
||||
}
|
||||
},
|
||||
"after_counts": {
|
||||
"papers": {
|
||||
"rows": 3568022,
|
||||
"max_id": 3568595
|
||||
},
|
||||
"source_records": {
|
||||
"rows": 3815513,
|
||||
"max_id": 3815514
|
||||
},
|
||||
"submissions": {
|
||||
"rows": 532202,
|
||||
"max_id": 532203
|
||||
},
|
||||
"decisions": {
|
||||
"rows": 571754,
|
||||
"max_id": 571754
|
||||
}
|
||||
},
|
||||
"new_papers": 0,
|
||||
"new_sources": 1,
|
||||
"new_editions": 1,
|
||||
"new_source_records": 10,
|
||||
"new_attempts": 10,
|
||||
"new_decisions": 10,
|
||||
"source_id": 123160,
|
||||
"source_tag_evidence_links": 94,
|
||||
"protected_papers": 6394,
|
||||
"all_protected_old_profiles_sources_tags_identifiers_preserved": true,
|
||||
"source_observations": 10,
|
||||
"source_expected": 140,
|
||||
"source_remaining": 130,
|
||||
"coverage": "partial",
|
||||
"foundry_readyz": [
|
||||
200
|
||||
],
|
||||
"sqlite_frozen": true,
|
||||
"imported": [
|
||||
{
|
||||
"external_id": "2026/3",
|
||||
"paper_id": 2813483,
|
||||
"record_id": 3815505,
|
||||
"submission_id": 532194,
|
||||
"decision_id": 571745,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/12",
|
||||
"paper_id": 2785459,
|
||||
"record_id": 3815506,
|
||||
"submission_id": 532195,
|
||||
"decision_id": 571746,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/35",
|
||||
"paper_id": 2757859,
|
||||
"record_id": 3815507,
|
||||
"submission_id": 532196,
|
||||
"decision_id": 571747,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/51",
|
||||
"paper_id": 2752643,
|
||||
"record_id": 3815508,
|
||||
"submission_id": 532197,
|
||||
"decision_id": 571748,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/74",
|
||||
"paper_id": 2762340,
|
||||
"record_id": 3815509,
|
||||
"submission_id": 532198,
|
||||
"decision_id": 571749,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/75",
|
||||
"paper_id": 2757480,
|
||||
"record_id": 3815510,
|
||||
"submission_id": 532199,
|
||||
"decision_id": 571750,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/78",
|
||||
"paper_id": 2768260,
|
||||
"record_id": 3815511,
|
||||
"submission_id": 532200,
|
||||
"decision_id": 571751,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/84",
|
||||
"paper_id": 2740614,
|
||||
"record_id": 3815512,
|
||||
"submission_id": 532201,
|
||||
"decision_id": 571752,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/87",
|
||||
"paper_id": 2752271,
|
||||
"record_id": 3815513,
|
||||
"submission_id": 532202,
|
||||
"decision_id": 571753,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/363",
|
||||
"paper_id": 2757497,
|
||||
"record_id": 3815514,
|
||||
"submission_id": 532203,
|
||||
"decision_id": 571754,
|
||||
"edition_id": 104044
|
||||
}
|
||||
]
|
||||
},
|
||||
"history_protection": {
|
||||
"submissions": {
|
||||
"rows": 532192,
|
||||
"sha256": "34e2fe27cccf89e6e80228318212a846369fdd315c4cfe67d7d535013eac8298"
|
||||
},
|
||||
"decisions": {
|
||||
"rows": 571744,
|
||||
"sha256": "efbda4e91950c5c7298f536bc5967f7fee57d8b20a7843e00dc69871b1553bd1"
|
||||
},
|
||||
"decision_sources": {
|
||||
"rows": 571744,
|
||||
"sha256": "1d8899b16b40e9f37ac84dc601efc3e7906af5c06bca49946100697f22c96174"
|
||||
},
|
||||
"submission_sources": {
|
||||
"rows": 581976,
|
||||
"sha256": "ab3ab200c4abbfef39bc5a2a757550191441c703ae505372d38d9afa9a4daf65"
|
||||
},
|
||||
"submission_identifiers": {
|
||||
"rows": 383956,
|
||||
"sha256": "0e8f514e186e08bd00a65f8e14fc7f63db6add76722271e0111e14f5d27b9dbe"
|
||||
}
|
||||
},
|
||||
"evidence": {
|
||||
"accepted_report": "/data/paper-index/collected/ismir2026-import10-v1/postgresql-postcheck-v1.json",
|
||||
"accepted_report_sha256": "8a08015bbd725b400de53477837e850ca92a50479d7f86f2cfed56d50bb22577",
|
||||
"baseline": "/data/paper-index/collected/ismir2026-baseline10-v1/baseline.json",
|
||||
"baseline_sha256": "a0b5b8c2c21ee567cf80152b16f66643309bbe33ee9cd67d19da33d6b6337656",
|
||||
"baseline_archive_sha256": "eb94f99cb53c9a30f63d34ea11b23c11759e45b571b7efce2dcbe8cb4fe52fb8",
|
||||
"preparation_sha256": "0d9fad74640ab57fc8bea8aaf72ba07202a40b174d05253db7348b14455a0462",
|
||||
"acceptance_archive": "/data/paper-index/validation/ismir2026-import10-acceptance-recovery-v1.tar.gz",
|
||||
"acceptance_archive_sha256": "138e1fb986f5aa6a42794eed982fdc91ae994b8be82f8bf25360db2f7ba6c70c",
|
||||
"fresh_packets_sha256": "88e423aede0ea83b9f736891c09b12d0cb24f2f8b70508e4863cd07760de5e0d",
|
||||
"local_acceptance_check_sha256": "42d31337c76b96f93505011db7c797123b67ce2eec7defd0175ea0e9c1df2089",
|
||||
"census_summary_sha256": "2157035b72d351e2fe91f2bc0c6ebb305f91d95f8772ab9a088076ce4b86e1c2"
|
||||
},
|
||||
"native_query_example": {
|
||||
"paper_id": 2757497,
|
||||
"publication_outcome": "accepted",
|
||||
"prior_venue": "ICLR",
|
||||
"prior_year": 2026,
|
||||
"prior_submission_id": 131721,
|
||||
"prior_negative_attempts": 1,
|
||||
"prior_negative_observations": 2,
|
||||
"prior_decision_ids": [
|
||||
133086,
|
||||
538172
|
||||
],
|
||||
"later_observed_venue": "ISMIR",
|
||||
"later_observed_year": 2026,
|
||||
"original_ismir_signed_decision_verified": false,
|
||||
"ismir_decision_date": null,
|
||||
"observation_time_is_not_decision_time": true
|
||||
},
|
||||
"presence_census": {
|
||||
"snapshot_at": "2026-11-18 05:20:48.641315+00",
|
||||
"scope": "Fixed278 venue-year presence only, not paper/track/abstract/public-negative completeness or final deduplication. Legacy43 gaps deferred, not filled.",
|
||||
"target_venues": 43,
|
||||
"expected_slots": 278,
|
||||
"populated_slots": 274,
|
||||
"missing_slots": 4,
|
||||
"missing": [
|
||||
{
|
||||
"venue": "CIKM",
|
||||
"year": 2026
|
||||
},
|
||||
{
|
||||
"venue": "ECCV",
|
||||
"year": 2026
|
||||
},
|
||||
{
|
||||
"venue": "ICDM",
|
||||
"year": 2026
|
||||
},
|
||||
{
|
||||
"venue": "IROS",
|
||||
"year": 2026
|
||||
}
|
||||
],
|
||||
"delta_from_v20": {
|
||||
"newly_present": [
|
||||
{
|
||||
"venue": "ISMIR",
|
||||
"year": 2026
|
||||
}
|
||||
],
|
||||
"previous_missing": 5,
|
||||
"current_missing": 4
|
||||
},
|
||||
"snapshot_sha256": "c14bab883c5fcab33b450fcf240dec7225f2f17bc04c3e728b5583657b4ad23b",
|
||||
"fixed_scope_sha256": "a835bef5b9c3a0f76069d840cecdb9a6bcf65d3a38796d70dbfdfd4b49242264",
|
||||
"accepted_import_sha256": "8a08015bbd725b400de53477837e850ca92a50479d7f86f2cfed56d50bb22577",
|
||||
"database_written": false
|
||||
},
|
||||
"consumed_queue": "/data/paper-index/collected/ismir2026-prepared10-v1: NEVER REPLAY",
|
||||
"next": [
|
||||
"Review130 remaining source observations, including31 total verbatim-abstract candidates of which10 consumed; do not count a title/author screen as identity approval.",
|
||||
"Fresh incremental parent must be partial10/140; old preparation0/140 is consumed.",
|
||||
"Continue all venue/year/track public negative availability and formal source coverage.",
|
||||
"Finish cross-venue identities/types, then final whole-DB dedup/identifier/abstract/constraints/query audit after imports."
|
||||
]
|
||||
}
|
||||
@@ -0,0 +1,330 @@
|
||||
{
|
||||
"as_of": "2026-11-18",
|
||||
"goal_progress": "13/17; stage14 incomplete",
|
||||
"source": "ismir_program",
|
||||
"scope": "21 additional reviewed existing-work source links; cumulativepartial31/140, not public-rejection completeness or final deduplication",
|
||||
"database_acceptance": {
|
||||
"status": "passed",
|
||||
"at": "2026-11-18 06:38:55.148539+00:00",
|
||||
"writer_commit": "3754545003e1f96b88c6533cc3a69dc557f07305",
|
||||
"before_counts": {
|
||||
"papers": {
|
||||
"rows": 3568022,
|
||||
"max_id": 3568595
|
||||
},
|
||||
"source_records": {
|
||||
"rows": 3815513,
|
||||
"max_id": 3815514
|
||||
},
|
||||
"submissions": {
|
||||
"rows": 532202,
|
||||
"max_id": 532203
|
||||
},
|
||||
"decisions": {
|
||||
"rows": 571754,
|
||||
"max_id": 571754
|
||||
}
|
||||
},
|
||||
"after_counts": {
|
||||
"papers": {
|
||||
"rows": 3568022,
|
||||
"max_id": 3568595
|
||||
},
|
||||
"source_records": {
|
||||
"rows": 3815534,
|
||||
"max_id": 3815535
|
||||
},
|
||||
"submissions": {
|
||||
"rows": 532223,
|
||||
"max_id": 532224
|
||||
},
|
||||
"decisions": {
|
||||
"rows": 571775,
|
||||
"max_id": 571775
|
||||
}
|
||||
},
|
||||
"new_papers": 0,
|
||||
"new_sources": 0,
|
||||
"new_editions": 0,
|
||||
"new_source_records": 21,
|
||||
"new_attempts": 21,
|
||||
"new_decisions": 21,
|
||||
"source_id": 123160,
|
||||
"source_tag_evidence_links": 173,
|
||||
"protected_papers": 6415,
|
||||
"all_protected_old_profiles_sources_tags_identifiers_preserved": true,
|
||||
"source_observations": 31,
|
||||
"source_expected": 140,
|
||||
"source_remaining": 109,
|
||||
"coverage": "partial",
|
||||
"foundry_readyz": [
|
||||
200
|
||||
],
|
||||
"sqlite_frozen": true,
|
||||
"imported": [
|
||||
{
|
||||
"external_id": "2026/148",
|
||||
"paper_id": 2755057,
|
||||
"record_id": 3815515,
|
||||
"submission_id": 532204,
|
||||
"decision_id": 571755,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/184",
|
||||
"paper_id": 2752723,
|
||||
"record_id": 3815516,
|
||||
"submission_id": 532205,
|
||||
"decision_id": 571756,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/196",
|
||||
"paper_id": 2784571,
|
||||
"record_id": 3815517,
|
||||
"submission_id": 532206,
|
||||
"decision_id": 571757,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/240",
|
||||
"paper_id": 2738896,
|
||||
"record_id": 3815518,
|
||||
"submission_id": 532207,
|
||||
"decision_id": 571758,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/251",
|
||||
"paper_id": 2753889,
|
||||
"record_id": 3815519,
|
||||
"submission_id": 532208,
|
||||
"decision_id": 571759,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/258",
|
||||
"paper_id": 2662030,
|
||||
"record_id": 3815520,
|
||||
"submission_id": 532209,
|
||||
"decision_id": 571760,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/270",
|
||||
"paper_id": 2769109,
|
||||
"record_id": 3815521,
|
||||
"submission_id": 532210,
|
||||
"decision_id": 571761,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/298",
|
||||
"paper_id": 2752114,
|
||||
"record_id": 3815522,
|
||||
"submission_id": 532211,
|
||||
"decision_id": 571762,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/317",
|
||||
"paper_id": 2765378,
|
||||
"record_id": 3815523,
|
||||
"submission_id": 532212,
|
||||
"decision_id": 571763,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/318",
|
||||
"paper_id": 2764935,
|
||||
"record_id": 3815524,
|
||||
"submission_id": 532213,
|
||||
"decision_id": 571764,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/342",
|
||||
"paper_id": 2738585,
|
||||
"record_id": 3815525,
|
||||
"submission_id": 532214,
|
||||
"decision_id": 571765,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/370",
|
||||
"paper_id": 2830871,
|
||||
"record_id": 3815526,
|
||||
"submission_id": 532215,
|
||||
"decision_id": 571766,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/371",
|
||||
"paper_id": 2858445,
|
||||
"record_id": 3815527,
|
||||
"submission_id": 532216,
|
||||
"decision_id": 571767,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/373",
|
||||
"paper_id": 2820562,
|
||||
"record_id": 3815528,
|
||||
"submission_id": 532217,
|
||||
"decision_id": 571768,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/409",
|
||||
"paper_id": 2522863,
|
||||
"record_id": 3815529,
|
||||
"submission_id": 532218,
|
||||
"decision_id": 571769,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/418",
|
||||
"paper_id": 2750749,
|
||||
"record_id": 3815530,
|
||||
"submission_id": 532219,
|
||||
"decision_id": 571770,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/468",
|
||||
"paper_id": 2846824,
|
||||
"record_id": 3815531,
|
||||
"submission_id": 532220,
|
||||
"decision_id": 571771,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/518",
|
||||
"paper_id": 2761686,
|
||||
"record_id": 3815532,
|
||||
"submission_id": 532221,
|
||||
"decision_id": 571772,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/524",
|
||||
"paper_id": 2784543,
|
||||
"record_id": 3815533,
|
||||
"submission_id": 532222,
|
||||
"decision_id": 571773,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/539",
|
||||
"paper_id": 2761760,
|
||||
"record_id": 3815534,
|
||||
"submission_id": 532223,
|
||||
"decision_id": 571774,
|
||||
"edition_id": 104044
|
||||
},
|
||||
{
|
||||
"external_id": "2026/552",
|
||||
"paper_id": 2758575,
|
||||
"record_id": 3815535,
|
||||
"submission_id": 532224,
|
||||
"decision_id": 571775,
|
||||
"edition_id": 104044
|
||||
}
|
||||
]
|
||||
},
|
||||
"history_protection": {
|
||||
"submissions": {
|
||||
"rows": 532202,
|
||||
"sha256": "9d41955c80fc91dae4f65d65e0f69198ce2d79508ceebfe943bdc19825983df8"
|
||||
},
|
||||
"decisions": {
|
||||
"rows": 571754,
|
||||
"sha256": "c8dc762d089d1656ee1308dacec9109605e1a4be7e5fcfa8f95013e77a0cb55c"
|
||||
},
|
||||
"decision_sources": {
|
||||
"rows": 571754,
|
||||
"sha256": "231f15b4c70435977d6ab784112ff5c748541eeb4d54c28bb861f8835f1eba10"
|
||||
},
|
||||
"submission_sources": {
|
||||
"rows": 581986,
|
||||
"sha256": "b9d8c39e945df340daf1c695a6359cfb549b8bd2f391cb139dabba2c16afa242"
|
||||
},
|
||||
"submission_identifiers": {
|
||||
"rows": 383956,
|
||||
"sha256": "0e8f514e186e08bd00a65f8e14fc7f63db6add76722271e0111e14f5d27b9dbe"
|
||||
}
|
||||
},
|
||||
"evidence": {
|
||||
"parent_acceptance_sha256": "8a08015bbd725b400de53477837e850ca92a50479d7f86f2cfed56d50bb22577",
|
||||
"accepted_report": "/data/paper-index/collected/ismir2026-import21-v1/postgresql-postcheck-v1.json",
|
||||
"accepted_report_sha256": "153ef13dbac9f4243c81bdd2a16d3ade74372d3e45ca5e84acc2b7258bca7bfc",
|
||||
"baseline": "/data/paper-index/collected/ismir2026-baseline21-v1/baseline.json",
|
||||
"baseline_sha256": "204b571b47e164d68930158c2578f609814969940628d421a2b64c3a6ce1b018",
|
||||
"baseline_archive_sha256": "ecac32e1d3890786a5670ff06f88fa7bd92155306cf8133f3164de0cb9f48e6b",
|
||||
"preparation_sha256": "6781edfaa67798d76329fb5c7948d83693607c96352a38eccce0067faa95deaa",
|
||||
"acceptance_archive": "/data/paper-index/validation/ismir2026-import21-acceptance-recovery-v1.tar.gz",
|
||||
"acceptance_archive_sha256": "55e1cc8301548ff0abc195add5f40d7082a06e4f087f6da5838ffd2f6b33339c",
|
||||
"fresh_packets_sha256": "79e0e8187a522478c443d036fdbf2f8d5e6a22c949d46fb0592a664ec1bc22fe",
|
||||
"local_acceptance_check_sha256": "04f4ec6969a71655d70f446347646f63efc56e5a5a8e974b931703482e5e7ceb",
|
||||
"census_summary_sha256": "ee4ec2e6d6eaed55ce3475e735bfba9bd94e512893cd3c3024035b48b744f475",
|
||||
"ci_run": 1257,
|
||||
"ci_tests": 556,
|
||||
"ci_proof_sha256": "5ad08aa907caec5a1ca63df5e7345db4802dab4e3a91293a0303bf481a903ff8"
|
||||
},
|
||||
"consumed_queue": "/data/paper-index/collected/ismir2026-prepared21-v2: NEVER REPLAY",
|
||||
"previous_consumed_queue": "/data/paper-index/collected/ismir2026-prepared10-v1: NEVER REPLAY",
|
||||
"query_example_rechecked": {
|
||||
"paper_id": 2757497,
|
||||
"publication_outcome": "accepted",
|
||||
"rejected_attempts": 1,
|
||||
"rejection_source_observations": 2,
|
||||
"distinct_ismir_accepted_attempts": 1,
|
||||
"original_ismir_signed_decision_verified": false,
|
||||
"ismir_decision_date": null
|
||||
},
|
||||
"new_batch_prior_negative_attempts": 0,
|
||||
"new_batch_prior_negative_observations": 0,
|
||||
"source_tag_links_cumulative": 267,
|
||||
"presence_census": {
|
||||
"snapshot_at": "2026-11-18 06:46:44.099675+00",
|
||||
"scope": "Fixed278 venue-year presence only, not paper/track/abstract/public-negative completeness or final deduplication. Legacy43 gaps deferred, not filled.",
|
||||
"target_venues": 43,
|
||||
"expected_slots": 278,
|
||||
"populated_slots": 274,
|
||||
"missing_slots": 4,
|
||||
"missing": [
|
||||
{
|
||||
"venue": "CIKM",
|
||||
"year": 2026
|
||||
},
|
||||
{
|
||||
"venue": "ECCV",
|
||||
"year": 2026
|
||||
},
|
||||
{
|
||||
"venue": "ICDM",
|
||||
"year": 2026
|
||||
},
|
||||
{
|
||||
"venue": "IROS",
|
||||
"year": 2026
|
||||
}
|
||||
],
|
||||
"delta_from_v21": {
|
||||
"newly_present": [],
|
||||
"previous_missing": 4,
|
||||
"current_missing": 4,
|
||||
"ISMIR_source_observations": 31
|
||||
},
|
||||
"snapshot_sha256": "5aabc91217ad787946ba50d6a670296b1e3152a54f5ecaa0e3a9b3a952b63fa7",
|
||||
"fixed_scope_sha256": "a835bef5b9c3a0f76069d840cecdb9a6bcf65d3a38796d70dbfdfd4b49242264",
|
||||
"accepted_import_sha256": "153ef13dbac9f4243c81bdd2a16d3ade74372d3e45ca5e84acc2b7258bca7bfc",
|
||||
"database_written": false
|
||||
},
|
||||
"next": [
|
||||
"Remaining109 observations require review;31 exact-source-abstract candidates are now consumed, not reusable queues.",
|
||||
"Future increments require a fresh partial31/140 parent and new preservation baseline.",
|
||||
"Keep raw authors/title/abstract/keyword versions; no numeric miniconf-to-OpenReview-ID assumption, global person aliases or inferred research types.",
|
||||
"Continue all43 venue/year/track formal-source and public-negative availability audits, then final whole-DB dedup after imports."
|
||||
]
|
||||
}
|
||||
@@ -0,0 +1,14 @@
|
||||
# 公开决定来源补查
|
||||
|
||||
核查日期为 2026 年 11 月 18 日。范围是六个会议的官方发布政策、投稿入口和统计口径。逐论文决定以明确的状态和可追溯来源为准,本次核查新增入库决定 0 条。
|
||||
|
||||
| 会议 | 已核验的来源范围 | 待补事项 |
|
||||
| --- | --- | --- |
|
||||
| ICML 2026 | 征稿和评审指南允许被拒投稿的作者选择公开原稿与评审材料,作者指南的终稿提交段落说明录用稿的发布流程。既有归档只覆盖其中可逐条核验的子集。 [征稿](https://icml.cc/Conferences/2026/CallForPapers) · [评审指南](https://icml.cc/Conferences/2026/ReviewerInstructions) · [作者指南](https://icml.cc/Conferences/2026/AuthorInstructions) | 查找可枚举的官方公开决定入口。当前匿名页面只显示加载界面,公开子集的总量未知。 [OpenReview](https://openreview.net/group?id=ICML.cc%2F2026%2FConference) |
|
||||
| AISTATS 2026 | 征稿指定 OpenReview 投稿,评审指南描述评审和决定流程。 [征稿](https://virtual.aistats.org/Conferences/2026/CallForPapers) · [评审指南](https://virtual.aistats.org/Conferences/2026/ReviewerGuidelines) | 匿名会议页面只显示加载界面,公开逐论文负面决定来源待核。 [OpenReview](https://openreview.net/group?id=aistats.org%2FAISTATS%2F2026%2FConference) |
|
||||
| RecSys 2026 | 官方征稿列出 EasyChair 投稿、不同贡献类型和 desk reject 政策。公开统计表按年度汇总长论文,本次看到的最新行是 2025。 [征稿](https://recsys.acm.org/recsys26/call/) · [提交统计](https://recsys.acm.org/statistics/) | 逐论文公开拒稿、撤稿和 desk reject 来源待核。统计分母按类别、年份和来源记录。 |
|
||||
| COLM 2026 | 主会政策明确规定拒稿和撤稿及其元数据不由会议公开。官方站另提供预论文集。 [征稿](https://colm.cc/Conferences/2026/CallForPapers) · [预论文集](https://proceedings.colm.cc/paper_files/paper/2026) | 个别作者主动披露的决定仍须保留来源,并核查论文与投稿身份。 |
|
||||
| MLSys 2026 | 研究与产业是不同投稿类别,征稿列出 HotCRP 和双盲评审政策。 [征稿](https://mlsys.org/Conferences/2026/CallForPapers) · [HotCRP](https://mlsys26.hotcrp.com/) | 两类逐论文公开负面决定来源待核。 |
|
||||
| UAI 2020、2021、2026 | 2020 和 2021 使用 CMT。2026 使用 OpenReview,作者可选择公开被拒稿件,默认私密。 [2020 评审说明](https://www.auai.org/uai2020/reviewer_instructions.php) · [2021 投稿说明](https://www.auai.org/uai2021/submission_instructions) · [2026 投稿说明](https://www.auai.org/uai2026/submission_instructions) | 这三届的公开逐论文负面决定尚未核验。2022–2025 的镜像范围和恢复线索见 [UAI 镜像核查](uai-public-negative-source-status-2026-11-17.md)。 |
|
||||
|
||||
继续补漏时,按会议、年份、类别和来源记录覆盖。正式论文集用于核验公开发表。汇总接收率用于核查统计口径。作者后来公开录用的版本需要先核对作品身份,再与原先的投稿及决定关联。
|
||||
@@ -0,0 +1,320 @@
|
||||
{
|
||||
"version": 1,
|
||||
"as_of": "2026-11-18",
|
||||
"scope": "930 cached ResearchArcade non-ICLR negative claims, NeurIPS 2021–2025 and EMNLP 2023; not a conference-completeness audit",
|
||||
"database_snapshot": {
|
||||
"database": "foundry_data",
|
||||
"schema": "research",
|
||||
"at": "2026-11-18 03:48:34.65659+00",
|
||||
"snapshot": "1048885:1048885:"
|
||||
},
|
||||
"database_written": false,
|
||||
"approved_for_import": 0,
|
||||
"claims": 930,
|
||||
"known_same_scope_attempt_claims": 925,
|
||||
"all_925_have_rejected_status": true,
|
||||
"without_same_scope_attempt_claims": 5,
|
||||
"by_venue_year": [
|
||||
{
|
||||
"venue": "EMNLP",
|
||||
"year": 2023,
|
||||
"claims": 10
|
||||
},
|
||||
{
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"claims": 136
|
||||
},
|
||||
{
|
||||
"venue": "NeurIPS",
|
||||
"year": 2022,
|
||||
"claims": 153
|
||||
},
|
||||
{
|
||||
"venue": "NeurIPS",
|
||||
"year": 2023,
|
||||
"claims": 176
|
||||
},
|
||||
{
|
||||
"venue": "NeurIPS",
|
||||
"year": 2024,
|
||||
"claims": 201
|
||||
},
|
||||
{
|
||||
"venue": "NeurIPS",
|
||||
"year": 2025,
|
||||
"claims": 254
|
||||
}
|
||||
],
|
||||
"evidence": {
|
||||
"diff_report_sha256": "d49361a8622f3b3f59e824690ceb8bf1c878be35c3e3758366d3b46a407673f5",
|
||||
"candidate_packets_sha256": "3b21df27e0f30e56b81086d16d3943c4563bbe382200e866b15de14510e07472",
|
||||
"dataset_revision": "c61bb99b0452429c32f85b1e392f3afdb65731c4",
|
||||
"dataset_parquet_sha256": "bb6e165ee152a95d91a7ffa16abc9d4a6effbdfcf59183b3bb819d09171689d1",
|
||||
"candidate_jsonl_sha256": "848e646596e8c747e0c636c2381a139978aff8f2c19f01ad49c98e7f4270c5e6",
|
||||
"collector_repository": "https://github.com/ulab-uiuc/research-arcade",
|
||||
"collector_revision": "ecf9c7c2490840b029bd2add4f02d04aca1d014e",
|
||||
"collector_openreview_crawler_sha256": "4539d951ca05b20785169638eab3884148aa6c354a34c5ecc04e15c84a5910fe",
|
||||
"collector_revision_proven_to_generate_dataset": false
|
||||
},
|
||||
"collector_semantics": {
|
||||
"legacy_2020_2022": "Reads Blind_Submission note.id; first reply with content.decision, no invitation/signature verification. Not a verified final-decision note.",
|
||||
"legacy_2019": "Reads first recommendation, which must not be promoted to a final decision. Outside active 2020+ backfill.",
|
||||
"modern_branch": "Reads Submission note.id and last venueid component; Conference replaced by venue literal. Withdrawn_Submission skipped.",
|
||||
"identity": "The exported field paper_openreview_id is note.id; forum equality or equivalence to another mirror ID is not independently proved.",
|
||||
"coverage": "Six-field table has no authors or decision signatures; no completeness inference for public withdrawn/rejected records."
|
||||
},
|
||||
"held_claims": [
|
||||
{
|
||||
"source_row": 33219,
|
||||
"note_id_claim": "FSf9h6t152",
|
||||
"venue": "NeurIPS",
|
||||
"year": 2025,
|
||||
"raw_decision": "Rejected_Submission",
|
||||
"title": "Hierarchical Restructuring of Graph Neural Networks for Robust Learning under Edge Perturbations",
|
||||
"same_scope_identifier_match": false,
|
||||
"exact_title_candidates": [],
|
||||
"approved_for_import": false,
|
||||
"disposition": "hold_no_exact_title_candidate_not_proof_of_novelty"
|
||||
},
|
||||
{
|
||||
"source_row": 44669,
|
||||
"note_id_claim": "Pk972F7J8Um",
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"raw_decision": "Reject",
|
||||
"title": "Differential Privacy of Dirichlet Posterior Sampling",
|
||||
"same_scope_identifier_match": false,
|
||||
"exact_title_candidates": [
|
||||
{
|
||||
"paper_id": 3527429,
|
||||
"existing_identifiers": [
|
||||
{
|
||||
"scheme": "openreview",
|
||||
"value": "fgBxcKcz9ea",
|
||||
"paper_id": 3527429
|
||||
},
|
||||
{
|
||||
"scheme": "papercopilot_openreview",
|
||||
"value": "nips/2021/fgBxcKcz9ea",
|
||||
"paper_id": 3527429
|
||||
}
|
||||
],
|
||||
"source_abstract_exact_equal": true,
|
||||
"existing_attempts": [
|
||||
{
|
||||
"id": 444666,
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"source_submission_id": "nips/2021/fgBxcKcz9ea",
|
||||
"statuses": [
|
||||
"rejected"
|
||||
]
|
||||
}
|
||||
],
|
||||
"attempts_truncated": false,
|
||||
"identity_approved": false,
|
||||
"id_alias_approved": false
|
||||
}
|
||||
],
|
||||
"approved_for_import": false,
|
||||
"disposition": "hold_distinct_note_ids_even_with_equal_title_and_abstract"
|
||||
},
|
||||
{
|
||||
"source_row": 44696,
|
||||
"note_id_claim": "ki4eJ1fSJNq",
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"raw_decision": "Reject",
|
||||
"title": "Fitting large mixture models using stochastic component selection",
|
||||
"same_scope_identifier_match": false,
|
||||
"exact_title_candidates": [
|
||||
{
|
||||
"paper_id": 1545058,
|
||||
"existing_identifiers": [
|
||||
{
|
||||
"scheme": "arxiv",
|
||||
"value": "2110.04776",
|
||||
"paper_id": 1545058
|
||||
},
|
||||
{
|
||||
"scheme": "openreview",
|
||||
"value": "ZS394D3djsg",
|
||||
"paper_id": 1545058
|
||||
},
|
||||
{
|
||||
"scheme": "papercopilot_openreview",
|
||||
"value": "nips/2021/ZS394D3djsg",
|
||||
"paper_id": 1545058
|
||||
}
|
||||
],
|
||||
"source_abstract_exact_equal": true,
|
||||
"existing_attempts": [
|
||||
{
|
||||
"id": 444652,
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"source_submission_id": "nips/2021/ZS394D3djsg",
|
||||
"statuses": [
|
||||
"rejected"
|
||||
]
|
||||
}
|
||||
],
|
||||
"attempts_truncated": false,
|
||||
"identity_approved": false,
|
||||
"id_alias_approved": false
|
||||
}
|
||||
],
|
||||
"approved_for_import": false,
|
||||
"disposition": "hold_distinct_note_ids_even_with_equal_title_and_abstract"
|
||||
},
|
||||
{
|
||||
"source_row": 45757,
|
||||
"note_id_claim": "bDP7I2fYwLj",
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"raw_decision": "Reject",
|
||||
"title": "WeaveNet for Approximating Assignment Problems",
|
||||
"same_scope_identifier_match": false,
|
||||
"exact_title_candidates": [
|
||||
{
|
||||
"paper_id": 3527390,
|
||||
"existing_identifiers": [
|
||||
{
|
||||
"scheme": "openreview",
|
||||
"value": "2rAiDBJgR_",
|
||||
"paper_id": 3527390
|
||||
},
|
||||
{
|
||||
"scheme": "papercopilot_openreview",
|
||||
"value": "nips/2021/2rAiDBJgR_",
|
||||
"paper_id": 3527390
|
||||
}
|
||||
],
|
||||
"source_abstract_exact_equal": true,
|
||||
"existing_attempts": [
|
||||
{
|
||||
"id": 444580,
|
||||
"venue": "NeurIPS",
|
||||
"year": 2021,
|
||||
"source_submission_id": "nips/2021/2rAiDBJgR_",
|
||||
"statuses": [
|
||||
"rejected"
|
||||
]
|
||||
}
|
||||
],
|
||||
"attempts_truncated": false,
|
||||
"identity_approved": false,
|
||||
"id_alias_approved": false
|
||||
}
|
||||
],
|
||||
"approved_for_import": false,
|
||||
"disposition": "hold_distinct_note_ids_even_with_equal_title_and_abstract"
|
||||
},
|
||||
{
|
||||
"source_row": 49327,
|
||||
"note_id_claim": "yB8cQIICqe",
|
||||
"venue": "EMNLP",
|
||||
"year": 2023,
|
||||
"raw_decision": "Desk_Rejected_Submission",
|
||||
"title": "EZ-STANCE: A Large Dataset for Zero-Shot Stance Detection",
|
||||
"same_scope_identifier_match": false,
|
||||
"exact_title_candidates": [],
|
||||
"approved_for_import": false,
|
||||
"disposition": "hold_possible_later_publication_with_changed_title_and_dataset_size_no_identity_or_author_transfer",
|
||||
"later_publication_retrieval_lead": {
|
||||
"url": "https://aclanthology.org/2024.acl-long.838/",
|
||||
"http_status": 200,
|
||||
"bytes": 40557,
|
||||
"response_sha256": "b50dd55c48e31e4c10e97927ce01679369735834aa2c0f8d9f41f8e8e02fc776",
|
||||
"checked_at": "2026-11-18T04:07:03.699767+00:00",
|
||||
"anthology_id": "2024.acl-long.838",
|
||||
"doi": "10.18653/v1/2024.acl-long.838",
|
||||
"title": "EZ-STANCE: A Large Dataset for English Zero-Shot Stance Detection",
|
||||
"venue": "ACL",
|
||||
"year": 2024,
|
||||
"publication_verified": true,
|
||||
"link_to_anonymous_rejection_verified": false,
|
||||
"authors_transferred_to_anonymous_claim": false,
|
||||
"dataset_pairs_published_version": 47316,
|
||||
"dataset_pairs_rejection_mirror_version": 30606,
|
||||
"database_written": false,
|
||||
"current_database_evidence": {
|
||||
"snapshot": {
|
||||
"at": "2026-11-18 04:09:23.720956+00",
|
||||
"snapshot": "1051887:1051887:"
|
||||
},
|
||||
"paper_id": 2131805,
|
||||
"verified_identifiers": [
|
||||
{
|
||||
"scheme": "acl",
|
||||
"value": "2024.acl-long.838",
|
||||
"paper_id": 2131805
|
||||
},
|
||||
{
|
||||
"scheme": "acl_anthology",
|
||||
"value": "2024.acl-long.838",
|
||||
"paper_id": 2131805
|
||||
},
|
||||
{
|
||||
"scheme": "doi",
|
||||
"value": "10.18653/v1/2024.acl-long.838",
|
||||
"paper_id": 2131805
|
||||
}
|
||||
],
|
||||
"existing_attempts": [
|
||||
{
|
||||
"venue": "ACL",
|
||||
"year": 2024,
|
||||
"track": "main",
|
||||
"statuses": [
|
||||
"accepted"
|
||||
]
|
||||
}
|
||||
],
|
||||
"packet_file_sha256": "de5dfa658177adef87134cf7723afc086bd7395405a31166ff9597b03b3c8af3",
|
||||
"original_rejection_link_verified": false
|
||||
}
|
||||
}
|
||||
}
|
||||
],
|
||||
"access_probes": [
|
||||
{
|
||||
"local_evidence": "researcharcade-neurips2021-original-note-check-v1/report.json",
|
||||
"sha256": "746d752535b516abba43b0e8e6ad901bf39c07da532188dfa97ff911869084ae",
|
||||
"receipt_summary": {},
|
||||
"results": [
|
||||
{
|
||||
"requested_note_id": "Pk972F7J8Um",
|
||||
"status": 403,
|
||||
"url": "https://api.openreview.net/notes?id=Pk972F7J8Um&select=id%2Cforum%2Cinvitation%2Creaders%2Csignatures%2Ccontent.title%2Ccontent.abstract%2Ccontent.venue%2Ccontent.venueid",
|
||||
"sha256": "b0113384cebc9d724284062f776b45044d22450219112342721d8f9683571c08",
|
||||
"checked_at": "2026-11-18T04:01:58.643552+00:00"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"local_evidence": "emnlp2023-primary-public-metadata-probe-v1/notes-probe.json",
|
||||
"sha256": "47799dc4520e113d6a2e7c9ea2f362577bb21e3ad583ee02149a36c21cf6ea9d",
|
||||
"receipt_summary": {
|
||||
"url": "https://api.openreview.net/notes?invitation=EMNLP%2F2023%2FConference%2F-%2FSubmission&limit=1&select=id%2Cnumber%2Cforum%2Cinvitation%2Creaders%2Csignatures%2Ccdate%2Ctcdate%2Cmdate%2Ctmdate%2Ccontent.title%2Ccontent.abstract%2Ccontent.authors%2Ccontent.venue%2Ccontent.venueid%2Ccontent.keywords",
|
||||
"http_status": 403,
|
||||
"response_sha256": "798e964735eb1eff1e4954d5c9e79aad2afd40c3f88dcb82f8ef5bbc55cc0a61",
|
||||
"response_bytes": 546,
|
||||
"at": "2026-11-18T03:38:50.112624+00:00"
|
||||
}
|
||||
},
|
||||
{
|
||||
"local_evidence": "emnlp2023-primary-public-metadata-probe-v1/group-access.json",
|
||||
"sha256": "8fcb9ccd5615a737c17ef0eeedc9121e85a84ebd63667e83df0bda9ffb83d792",
|
||||
"receipt_summary": {
|
||||
"url": "https://api.openreview.net/groups?id=EMNLP%2F2023%2FConference",
|
||||
"http_status": 200,
|
||||
"response_sha256": "7c0cc75e2e8be1c18dffc96a9ceabea34f0f7810139c0a990a6bb117d66f6456",
|
||||
"response_bytes": 13143,
|
||||
"at": "2026-11-18T03:38:49.929085+00:00"
|
||||
}
|
||||
}
|
||||
],
|
||||
"access_limit": "NeurIPS 2021 first original note request and EMNLP 2023 source-declared Submission query returned 403. Requests stopped. No alternate API version, endpoint, authentication, or challenge bypass. 403 does not prove zero public records."
|
||||
}
|
||||
@@ -0,0 +1,52 @@
|
||||
# 公开负面来源核查:2026-11-18
|
||||
|
||||
本次检查 ResearchArcade 缓存中 NeurIPS 2021–2025、EMNLP 2023 的 930 条负面声明,不包括 ICLR 或 2020 年以前的补漏。原始候选文件未重新下载,PostgreSQL 查询为只读。逐条待办、来源哈希与数据库快照见[机器可读记录](public-negative-source-audit-2026-11-18.json)。没有新增论文、决定、ID 别名或合并。
|
||||
|
||||
## 当前库比对
|
||||
|
||||
925 条声明通过同会议、同年份的 OpenReview 标识匹配到投稿,且每条都有 rejected 决定。这里只证明对应拒稿已记录,未核验这些论文的所有后续录用历史,也未证明会议公开拒稿全集完整。
|
||||
|
||||
另 5 条没有同届标识匹配:
|
||||
|
||||
|来源 note.id 声明|会议|标题或识别信息|处置|
|
||||
|---|---|---|---|
|
||||
|FSf9h6t152|NeurIPS 2025|Hierarchical Restructuring of Graph Neural Networks for Robust Learning under Edge Perturbations|无精确标题候选;不是已证明的新论文|
|
||||
|Pk972F7J8Um|NeurIPS 2021|Differential Privacy of Dirichlet Posterior Sampling|候选 paper 3527429,已有同届 rejected,原 ID fgBxcKcz9ea|
|
||||
|ki4eJ1fSJNq|NeurIPS 2021|Fitting large mixture models using stochastic component selection|候选 paper 1545058,已有同届 rejected,原 ID ZS394D3djsg;另保留 arXiv 2110.04776|
|
||||
|bDP7I2fYwLj|NeurIPS 2021|WeaveNet for Approximating Assignment Problems|候选 paper 3527390,已有同届 rejected,原 ID 2rAiDBJgR_|
|
||||
|yB8cQIICqe|EMNLP 2023|EZ-STANCE: A Large Dataset for Zero-Shot Stance Detection|镜像称 Desk_Rejected_Submission;原始公开范围及身份未核验|
|
||||
|
||||
三个标题候选的镜像摘要都与已有 PaperCopilot 来源摘要逐字相等,但来源 ID 不同。相等文本不能独立证明两个 note.id 指向同一个 forum,也不能据此创建第二次投稿或批准 ID 别名。现有实体包没有截断投稿列表,均只有已记录的 NeurIPS 2021 拒稿;这不代表从未在其他地方录用。
|
||||
|
||||
## 采集器证据与使用限制
|
||||
|
||||
作者仓库 [research-arcade](https://github.com/ulab-uiuc/research-arcade) 固定版本 `ecf9c7c2490840b029bd2add4f02d04aca1d014e` 的 `openreview_crawler.py` 显示:
|
||||
|
||||
- 2020–2022 的旧接口路径读取 Blind_Submission 的 `note.id`,从 replies 中取第一个带 `decision` 字段的值,未验证决定 invitation 或签名。
|
||||
- 2019 路径取第一个 `recommendation`,不能将其升级为正式决定。该年度属于保留线索的范围,本轮不补导。
|
||||
- 新接口路径读取 Submission 的 `note.id`,用 `venueid` 尾段作为标签;遇 Conference 再取 `venue`,并跳过 Withdrawn_Submission。
|
||||
|
||||
三个源码文件均核对 Git blob SHA-1 和完整文件 SHA-256;没有执行第三方采集器。源码版本早于数据集版本,但未证明该提交实际生成了此数据集。数据集仅有六个元数据字段,缺少作者和决定签名,因此既不能证明完整 withdrawn 覆盖,也不能将导出的 ID 无条件认作已验证 forum。
|
||||
|
||||
## 原始来源访问结果
|
||||
|
||||
EMNLP 2023 官方分组元数据返回 200,其声明的 Submission invitation 与既有配置一致;对该 invitation 的匿名元数据查询返回 403。NeurIPS 2021 对第一条候选 note `Pk972F7J8Um` 的匿名元数据查询同样返回 403,余下五个 ID 未发请求。均已停止对应来源查询,没有换接口版本、登录或绕过访问限制。403 不能说明没有公开投稿;未保留完整错误正文,也不将具体拒绝原因猜测为已知。
|
||||
|
||||
[EMNLP 2023 官方政策](https://2023.emnlp.org/blog/EMNLP-2023-openreview/)说明公开录用稿件与选择公开的拒稿,现有 9 条 rejected 或本镜像的 1 条 desk-rejected 均不能证明完整公开集合。
|
||||
|
||||
## EZ-STANCE 后续发表线索
|
||||
|
||||
独立的 [ACL Anthology 官方条目](https://aclanthology.org/2024.acl-long.838/)确认相近标题的 EZ-STANCE 在 ACL 2024 Long Papers 发表,DOI `10.18653/v1/2024.acl-long.838`。正式标题多了 English,数据集由拒稿镜像中的 30,606 对变为 47,316 对。已保存官方 HTML 的完整哈希,登记为版本关系核查线索;尚未证明与匿名 EMNLP 2023 声明属于同一投稿或同一科学工作,没有把已发表作者转填到匿名记录,也没有写入关联或第二份论文。后续必须核验版本关系并保留各自的投稿决定证据。
|
||||
|
||||
当前 PostgreSQL 的正式标题变体查询命中已有 paper `2131805`,其 ACL/DOI 标识与官方条目一致,已有 `ACL / 2024 / main / accepted` 投稿历史。该实体无需重复新建。匿名 EMNLP 声明与该实体的关系仍未批准,不能把这条线索表述为已经确认的“先拒后录”。快照、标识及完整实体包哈希已记入机器记录。
|
||||
|
||||
## 下次从哪里继续
|
||||
|
||||
1. 保留五条候选及旧 ID,等待可验证的原始公开元数据或独立可靠来源;不要反复重放已拒绝的原始接口请求。
|
||||
2. 对三对 ID 分歧先取得 note/forum 关系证据,再独立审核科学工作身份与逐次投稿关系;保留已有拒稿及以后发现的录用。
|
||||
3. 两条没有精确标题候选的记录仍需标题变体、作者及摘要交叉核查;匿名元数据不能用于推断作者身份。
|
||||
4. 继续其余会议与年度的公开负面可用性核查,不将这 930 条当作全会、全年份覆盖,也不将 accepted-only 镜像算作拒稿补全。
|
||||
|
||||
另修正交接记录:TMLR 是额外的期刊政策线索,当前 43 个会议的 registry 中没有 TMLR;没有因此更改会议分母。2000–2019 已发现线索仍保留。
|
||||
|
||||
ICML、AISTATS、RecSys、COLM、MLSys 和 UAI 的官方政策、投稿入口和后续补查范围另见[公开决定来源补查](public-decision-source-boundaries-2026-11-18.md)。
|
||||
@@ -0,0 +1,51 @@
|
||||
# UAI 公开负面决定来源:2026-11-17 核查
|
||||
|
||||
**公开拒稿缺口仍未补齐,总体仍为 13/17。** 本页保存新来源的实际范围与恢复线索,避免下次把录用子集误当作全部投稿。主动补漏范围仍从2020年开始;本次镜像只含2022–2025,不能替代其他年份或其他会议的核查。
|
||||
|
||||
## 政策与访问边界
|
||||
|
||||
已有逐年核查记录保存在 [覆盖清单](inventory-2026-11-08.json) 的 UAI `policy_checks`:2023–2026 官方说明允许决定后公开录用及作者选择公开的拒稿,默认私密拒稿不属于公开可采集范围。政策并不证明任何镜像完整。
|
||||
|
||||
- [2023 投稿说明](https://www.auai.org/uai2023/submission_instructions)
|
||||
- [2024 投稿说明](https://auai.org/uai2024/submission_instructions)
|
||||
- [2025 投稿说明](https://www.auai.org/uai2025/submission_instructions)
|
||||
- [2026 投稿说明](https://www.auai.org/uai2026/submission_instructions)
|
||||
|
||||
先前匿名 notes 请求遇到 ChallengeRequiredError/403,仍作为访问限制保留。本轮没有认证、重试或绕过这些入口;读取的是整理者公开发布的固定版本镜像。复制的 `readers`、`signatures` 和邀请字段仍只是镜像内的来源断言,不是对当前原始接口可见性或签名真实性的验证。
|
||||
|
||||
## 实际镜像范围
|
||||
|
||||
来源:[latkes/ai-research-index-iclr-openreview](https://huggingface.co/datasets/latkes/ai-research-index-iclr-openreview/tree/4c9e09931a7814d7500630ffd064b74f1c01bcf3),固定提交 `4c9e09931a7814d7500630ffd064b74f1c01bcf3`。目录中存在 `papers/uai-YYYY/papers.jsonl` 与四个 `raw/uaiYYYY.jsonl`,不能仅凭数据集说明宣称拒稿覆盖。
|
||||
|
||||
| 年份 | 整理版及原始论坛数 | 原始回复对象 | 实际回复角色 | 标准结构化决定字段 |
|
||||
| --- | ---: | ---: | --- | --- |
|
||||
| 2022 | 230 | 1059 | Official_Review 829,Decision 230 | Accept (Poster) 194,Accept (Oral) 36 |
|
||||
| 2023 | 243 | 0 | 没有保存回复 | 没有保存决定记录 |
|
||||
| 2024 | 200 | 1994 | Official_Review 905,Rebuttal 889,Meta_Review 200 | 没有单独 Decision 记录或结构化 decision 字段 |
|
||||
| 2025 | 229 | 0 | 没有保存回复 | 没有保存决定记录 |
|
||||
|
||||
整理版902条均被整理者标为 `accepted=true`;原始根记录的 `venue` 为相应会议名称或 oral/poster/spotlight 文本,不能替代逐次决定证据,更不能算作902条已认证的原始录用决定。这不是四届全部投稿或全部录用的完整性证明。2024年还有 `Edit` 邀请,属于同一批回复上的另一个邀请,不能再累加成1994个新回复。
|
||||
|
||||
为排除字段漏读,对含回复的2022、2024快照另做了全部3053个回复的字段名、包装结构、邀请角色和短决定类字段核查;重新读取的正文哈希与首次捕获完全一致。2022的230条明确字段是录用标签。2024仅保存评审、作者回复和元评审,没有可据此新增的结构化拒稿/撤稿决定。没有从元评审文本、评分、推荐或缺少录用推断负面决定,也没有保存评审正文、作者回复正文或论文全文。
|
||||
|
||||
**这只描述上述固定镜像子集,不证明UAI没有公开拒稿。** 本批没有生成负面决定导入队列,也未写数据库。2020、2021、2026及其他可用来源仍需分别核查;某文件未出现在这个镜像目录中不代表全网不存在。
|
||||
|
||||
## 产物与恢复
|
||||
|
||||
服务器目录均相对 `/data/paper-index/collected/`:
|
||||
|
||||
- `uai2022-2025-mirror-metadata-v1`:902条整理版元数据投影、字段盘点与标签计数。
|
||||
- `uai2022-2025-raw-decisions-v1`:902个原始论坛的元数据、明确决定字段及来源正文/逐行哈希。
|
||||
- `uai2022-2024-reply-schema-v2`:3053个回复的元数据、字段名与短标签;评审正文不保存。角色核查包含直接记录和 `note` 包装,未发生静默遗漏。
|
||||
|
||||
源响应正文仅在内存中读取、校验,保存元数据投影与固定来源URL/字节哈希;不是保留了含评审文本的完整原始快照。复核原始源字节时应使用固定提交,并区分投影文件哈希与源正文哈希。恢复包已在本机独立核验全部902个论坛、3053个回复的元数据结构、逐文件哈希及来源关联。
|
||||
|
||||
| 恢复证据 | SHA256 |
|
||||
| --- | --- |
|
||||
| 整理版元数据恢复包 | 55889fb2b9005260cef1001b2cf36e203f3b7e1deeec0291f058b52eeae880e7 |
|
||||
| 原始元数据/回复结构恢复包 | caade7df899dd3e65572d06e2bf2326fad692cc5e1c1e4ab002fdc6d7823743f |
|
||||
| 原始结构独立核验报告 | 5f0ef2948d0db68fe0cd89eb4c8b57a1088b94b53bed98f180d2b96629746e72 |
|
||||
|
||||
服务器恢复包分别为 `/data/paper-index/validation/uai2022-2025-mirror-metadata-recovery-v1.tar.gz` 与 `uai2022-2025-raw-reply-recovery-v1.tar.gz`。本地同名备份位于 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/`。先检查实际单元、完整manifest和哈希,再恢复;不要因进度文件旧状态或日志安静重跑。
|
||||
|
||||
后续继续寻找能明确覆盖作者公开拒稿选择的来源,保留每次投稿及决定原文标签、原始ID与镜像关系,核对跨来源重叠和后续录用。UAI缺口不关闭,全部43个目标的逐年公开负面决定核查也不缩减;最终全库去重仍为独立必做阶段。
|
||||
@@ -0,0 +1,16 @@
|
||||
# ECCV 2026 正式出版章节采集
|
||||
|
||||
`collect eccv2026-publications` 读取已归档的 Springer LNCS 17001–17078 目录及所选章节的公开 HTML,输出可复核的 Paper Atlas 记录。目录覆盖 2,835 个章节 DOI。会议节目单与出版方卷册简介均为 2,834 条,逐条差异仍待核。Workshop 论文集另行处理。
|
||||
|
||||
```sh
|
||||
paper-atlas collect eccv2026-publications \
|
||||
--toc-snapshots /data/paper-index/collected/eccv2026-springer-toc-v1 \
|
||||
--chapter-snapshots /data/paper-index/validation/eccv-verified-chapter-pages \
|
||||
--out /data/paper-index/collected/eccv-publisher-metadata
|
||||
```
|
||||
|
||||
目录归档含 `receipts.jsonl`、`volume-checkpoint.json`、`summary.json` 及 156 页 HTML。摘要在章节详情页读取,章节归档沿用 `springer-selected` 的 `receipts.json` 格式。两类归档都按响应大小和 SHA-256 校验。目录卷号、分页、章节 DOI 归属、标题及 2026 年出处需要一致。已有输出目录需另查回执或换新空目录后运行。
|
||||
|
||||
输出含规范化 JSONL、manifest、采集回执以及所选章节的内容寻址存证。每条保留出版 DOI、卷号、页码、按顺序的完整作者及公开摘要。目录证据用 URL 和哈希连接到输入归档,保存归档以便复核。主会出版观察依据 Springer 正式论文集,原始投稿号、决定日期及签名仍未知。工具仅处理已归档的公开 HTML。
|
||||
|
||||
生成的文件待做论文身份复核后方可导入。准备生产批次时核实每个 DOI 与现有 arXiv 标识的绑定,保持两者的来源证据。对已存在的同届节目投稿需先核对投稿身份及轨道,再附加出版来源,避免因直接导入创建重复投稿。具有 ICLR 等既往决定的论文须保留原记录。新的增量导入 manifest 需以当前实际来源覆盖为父版本,留存待核差集和独立验收回执。
|
||||
@@ -0,0 +1,40 @@
|
||||
# ICASSP 2026 注册元数据采集
|
||||
|
||||
这是长期维护项目中的可重复运行采集器,按需离线处理冻结的公开元数据;不会启动数据库导入、联网或下载论文 PDF。命令仅输出待身份审查的来源版本。
|
||||
|
||||
## 输入与使用
|
||||
|
||||
2026-11-17 在数据服务器采集 Crossref 公开 REST API。先从正式节目表抽查的两篇标题确定注册会议名称和 DOI 前缀,再依据 IEEE DOI 前缀、proceedings-article 类型及注册的活动开始日期 `2026-05-03` 分页获取;日期过滤并不意味着所有结果都属于 ICASSP。
|
||||
|
||||
需要服务器 `/data/paper-index/collected/icassp2026-registry-capture-v1` 下的 `manifest.json` 与 `page-001.json` 至 `page-010.json`。将其父目录传给:
|
||||
|
||||
```sh
|
||||
paper-atlas collect icassp2026-registry \
|
||||
--snapshots /data/paper-index/collected \
|
||||
--out /data/paper-index/collected/icassp2026-normalized-<new-run>
|
||||
```
|
||||
|
||||
输出目录必须不存在。复跑前先检查原任务和产物,不能覆盖成功或失败输出。采集器验证固定清单 SHA256、全部页面原始字节、HTTP 回执、时区、页面数量、总数及唯一 DOI,再生成 JSONL、导入格式的部分覆盖清单、全部原始来源证据和采集报告。导入格式不代表身份审查通过;本命令批准导入数始终为零。
|
||||
|
||||
## 数据范围
|
||||
|
||||
- 10 页共有 4,645 个唯一 DOI;56 个属于 ICPEA,明确排除。
|
||||
- ICASSP 的 4,589 个注册条目中,7 个是封面、目录、作者索引、委员会和审稿人名单;依据已审查的准确 DOI、标题和页码排除,不以“没有作者”泛化排除论文。
|
||||
- 剩余 **4,582 条论文**全部缺少注册摘要,输出 `null`。保留完整注册字段、作者顺序、机构、原始标题、日期及原始来源哈希。
|
||||
- 23 个作者条目只提供单个姓名字段,原样保留并标记;不猜测名字拆分、拼音边界或人物别名。
|
||||
- `accepted` 表示已登记在该会议论文集中的发表观察,并非原始签名录用决定。DOI 用 `publication_doi`,不复制先前投稿系统的 ID。
|
||||
- track 为 `proceedings_unclassified`,不从技术主题推断 main、special session、workshop 或研究类型;没有自造标签。
|
||||
|
||||
正式节目表有 4,589 个编号和标题,与注册条目不是已证明的一一对应关系。独立标题规范化比较得到 4,415 个来源候选和 174 个未匹配节目标题;标题匹配不是作品身份批准,未匹配不等于拒稿、撤稿或未发表。节目表作者与注册作者的版本也不应自动拼接。
|
||||
|
||||
注册活动日期为 5 月 3–8 日,会议网站为 5 月 4–8 日,两者原始值均保留。Crossref 记录中的链接只是原始字段,不会访问 PDF。IEEE Open Preview 入口读取遇到 HTTP 418,未绕过访问限制。公开拒稿元数据是否可得仍为 unknown,本批不是拒稿覆盖证明,也不是第 14 项整体完成证明。
|
||||
|
||||
## 证据与恢复
|
||||
|
||||
注册来源清单 SHA256:`1f8ffa040b1f682a4b65f476f4115cb95629056d8de5c324c8417d823846ae93`。
|
||||
|
||||
原始证据恢复包在服务器 `/data/paper-index/validation/icassp2026-registry-capture-v1-recovery.tar.gz`,本地备份位于 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/` 同名文件,SHA256 为 `cdc506e95800fb8d37ed87f85740e96f56150f2d96daa971de21a5a7a9ab8388`。安全解包并核验后使用上述命令;测试夹具也冻结了全部 10 页及独立论文清单,供远程 CI 验证。
|
||||
|
||||
本批还需生产规范化验收、数据库当前 DOI/作品关联及历次投稿审查。所有导入结束后仍须执行全库去重、覆盖、摘要和历史完整性审计。编译、打包及测试均交给用户指定的 `ces-control:host` runner。
|
||||
|
||||
参考:[Crossref 筛选器文档](https://www.crossref.org/documentation/retrieve-metadata/rest-api/rest-api-filters/)、[正式录用标题表](https://cmsworkshops.com/ICASSP2026/papers/accepted_papers.php)、[会议日程](https://2026.ieeeicassp.org/technical-program/)、[发表和展示类型说明](https://2026.ieeeicassp.org/publishing-and-paper-presentation-options/)。
|
||||
@@ -44,3 +44,13 @@ paper-atlas collect icdm2026-program \
|
||||
原始来源恢复包为 `/data/paper-index/validation/remaining2026-official-leads-v1-recovery.tar.gz`,SHA-256 为 `0be8c84bb5c591e5afaaa5f95dd5334bf13a9be1671cb942e46e998f502495bb`。独立盘点为 `/data/paper-index/validation/icdm2026-independent-program-inventory-v1.json`,SHA-256 为 `e13eaea0a158dcaafdf5dfb4af7948e4793b0e98041ec87d42b52513acfdc5f6`;本机工作记录目录也保留同名备份。
|
||||
|
||||
下次继续先验证源码版本、原始证据和实际输出,再寻找可公开取得的摘要与正式出版记录,核查逐篇版本、稳定标识及旧投稿历史,准备新的只读预检和可审阅入库队列。完成日程规范化不意味着 ICDM2026 整届出版覆盖、公开负面决定普查或最终全库去重完成。
|
||||
|
||||
## 2026-11-18 官方展示来源核查
|
||||
|
||||
口头日程的 324 个节目编号全部出现在官方海报排期。海报表有 343 个不同编号,其中长演讲组安排 177 项海报,短演讲组安排 166 项。19 个只见于海报表的编号作为独立来源线索登记。两份排期的覆盖范围各自保留,正式出版目录仍待核验。[官方详细日程](https://icdm2026.neu.edu.cn/DetailedSessions/list.htm) · [官方海报表](https://www3.cs.stonybrook.edu/~icdm2026/PosterwDemo/list.htm)
|
||||
|
||||
共同编号 DM1319 的标题在两个来源中实质不同。详细日程描述基于 FFE-FRCN 的数据平台,海报表描述 FFE-FRCN 的少样本检测方法。两份标题及来源位置均已留存,该编号的跨来源作品关联待核验。
|
||||
|
||||
生产只读快照确认 ICDM 2026 仍无入库届次。节目、作者和独立录用证据支持继续处理 DM1125、DM790、DM1157 三项既有作品关联。DM1125 和 DM1157 各有一次 ICLR 拒稿投稿及两条来源观察,DM790 有一次 ICLR 撤回投稿。三项均尚未导入,节目原始摘要字段缺失。另有三个候选存在作者名称顺序或名单差异,继续暂缓作品关联。[FALCON 作者记录](https://arxiv.org/abs/2608.29635) · [PPGFlowECG 作者记录](https://arxiv.org/abs/2509.19774) · [DeepSynC 大学录用说明](https://dm.cs.univie.ac.at/news/story/deepsync-paper-accepted-at-icdm-2026)
|
||||
|
||||
本轮仍未核实公开的 ICDM 2026 拒稿或撤回决定导出。[逐条差集、标题冲突、快照和后续入口](coverage/icdm2026-presentation-scope-2026-11-18.json)
|
||||
|
||||
@@ -0,0 +1,83 @@
|
||||
# ISMIR 2026 官方节目与摘要采集
|
||||
|
||||
这是长期维护项目中的离线采集器,按需处理冻结的官方元数据,不联网、不执行网页脚本、不下载论文 PDF、不连接数据库。输出仍须经过作品身份与投稿历史审查。
|
||||
|
||||
## 使用
|
||||
|
||||
```sh
|
||||
paper-atlas collect ismir2026-program \
|
||||
--snapshots /data/paper-index/collected \
|
||||
--out /data/paper-index/collected/ismir2026-normalized-<new-run>
|
||||
```
|
||||
|
||||
输入父目录需要三个冻结目录:
|
||||
|
||||
- `remaining2026-listings-research-v2`:清单、`ismir-accepted.html`、`ismir-call.html`。
|
||||
- `ismir2026-script-research-v1`:清单及 `papers.js`,用于证明官方程序明确加载的数据文件和 poster 链接模板;只读取脚本原文。
|
||||
- `ismir2026-json-research-v1`:清单及 `papers.json`,提供 140 项完整元数据和公开摘要。
|
||||
|
||||
固定清单和全部输入原始字节均校验 SHA256、HTTP 回执、官方 URL 及带时区的采集时间。已有输出目录不能覆盖;先检查原任务是否已结束,再选择新目录。失败保留 `failed.json`,只有全批成功才发布 `collection-report.json`。
|
||||
|
||||
## 来源与版本
|
||||
|
||||
[正式录用名单](https://ismir2026.ismir.net/accepted-papers)和[官方虚拟平台](https://ismir2026program.ismir.net/papers.html)各有 140 项,经来源内标题对应核查一致;其中 2 处标题有直引号/弯引号文字差异,3 处作者名单有差异,两个来源的原始写法分别保留并标记,不静默修正或拼接作者。
|
||||
|
||||
作者差异涉及节目 ID `60`、`126`、`433`,包含大小写和姓名完整度问题。正式名单的作者机构可能含括号和分号,采集器只在括号外拆分作者;作者原始字符串也保留。来源内名单对应不代表数据库作品身份已经获准关联。
|
||||
|
||||
- 原始 JSON 摘要全文保留,包括数学符号、换行和 Unicode,不作扁平化提取。
|
||||
- 数字 `id`、`forum` 是 miniconf 节目定位符,不视为原始 OpenReview 投稿 ID;不推断 DOI 或 arXiv ID,投稿标识列表保持空。
|
||||
- `accepted` 来自正式录用名单与节目平台,不是原始签名决定。`track=program_unclassified`,不从主题、session 或远程展示方式推断研究方法、workshop 或论文类型。
|
||||
- `ismir_program:keyword` 保存来源非空关键词,规范空白并去除同条记录内完全重复值;原始关键词列表(含空项)保留在元数据。
|
||||
- `ismir_program:presentation` 保存来源展示方式。Remote/In-person 是展示方式,不是研究类型。
|
||||
- poster URL 严格依据冻结的官方 JS 模板生成;本命令不请求 poster 页面。
|
||||
|
||||
## 公开评审与拒稿边界
|
||||
|
||||
原始来源中有 **101 组明确选择公开的评审材料**;其余 39 项 `publish_reviews=No`,评审正文为空。采集器验证此偏好和正文的一致性,原文保存在来源元数据,不把评审意见或 meta-review 中的推荐转换成额外的最终决定。
|
||||
|
||||
这些记录来自录用节目,不能据此当作 101 条公开拒稿记录。公开拒稿导出是否可得仍记为 unknown,需要单独核查。[征稿与评审流程](https://ismir2026.ismir.net/authors/call-for-papers)中的评审规则也不等于具体投稿决定。
|
||||
|
||||
输出包含一份 140 条 JSONL、部分覆盖清单、七个原始来源文件及报告,批准导入数为零。采集成功不等于会议论文/track/公开负面决定覆盖完整,也不替代最后的全库去重审计。
|
||||
|
||||
## 恢复与验证
|
||||
|
||||
`papers.json` SHA256:`864d384cbc928554504e1b39849137bb380143e247a48f2d71563a698cb8e62d`。
|
||||
|
||||
原始 JSON 和 JS 恢复包:服务器 `/data/paper-index/validation/ismir2026-json-research-recovery-v1.tar.gz`,本地 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/` 同名文件;SHA256 为 `bba7b236cfe4d8b259d6c1230b3fe09a3e9071f5203884c92b016ca0718fb095`。
|
||||
|
||||
正式名单与政策恢复包:同目录 `remaining2026-listings-research-v2-recovery.tar.gz`,SHA256 为 `f5ef7c2e73d46712146e4d6711ad5ad562e16b8b55c0cd75d1041cd0b81142cb`。安全恢复三个源目录、核验文件后可重跑上述命令。
|
||||
|
||||
完整来源和独立 140 项清单在测试夹具中冻结。测试、构建、打包、安装及实际隔离目录 CLI 验证只在用户指定 `ces-control:host` runner 执行。生产规范化后还需独立核验全部记录,并用 PostgreSQL 只读预检检查候选作品和既往投稿;不自动合并来源差异或人物身份。
|
||||
|
||||
## 2026-11-18:首批 10 条验收(历史阶段)
|
||||
|
||||
本段保留首批历史状态;当前累计结果和增量父状态见下方最新验收段。
|
||||
|
||||
首批 10 项已逐篇核查研究内容,并通过当前数据库实体、作者、摘要及完整历史的只读复核,随后使用已通过远程 CI 的包内 `import records` 导入。新增论文实体 **0**,新增来源记录、投稿、录用观察各 **10**,来源标签证据 **94** 条;规范标题、作者、摘要及旧历史保留。原始来源摘要另存,不覆盖规范摘要。覆盖为 `partial10/140`,还有 **130** 项待核验。数值节目 ID 仍只作为节目定位符,未冒充 OpenReview 投稿 ID,未新增全局人名别名或推断研究类型。
|
||||
|
||||
[逐条验收与恢复记录](coverage/ismir2026-import10-2026-11-18.json)包含生产行 ID、保护基线、归档及证据哈希。服务器基线保护此前 6,384 篇与本批 10 篇的并集 **6,394** 篇,全库旧投稿、决定及来源关系哈希未变;独立新快照的 10 个完整实体包、来源字段、标签和标识也逐项核验。Foundry `/readyz` 返回200,17,987,751,936字节的 SQLite 备份保持只读。该验收不是最终全库科学工作去重。
|
||||
|
||||
例如 paper `2757497`(Learning Music Style for Piano Arrangement Through Cross-Modal Bootstrapping)保留一次 ICLR2026 拒稿投稿和两条拒稿来源观察,并新增独立 ISMIR2026 录用投稿。实际执行以下查询,返回 `publication_outcome=accepted`,同时保留原拒稿历史:
|
||||
|
||||
```sh
|
||||
paper-atlas query --pg-service foundry-research --pg-schema research \
|
||||
--paper 2757497 --outcome accepted
|
||||
```
|
||||
|
||||
ISMIR 的 accepted 来自正式名单和节目,尚无原始签名决定或具体决定日期;不能把采集时间当作决定时间。决定的规范时间使用包内既有毫秒格式,来源回执的原始微秒时间仍保留于元数据。
|
||||
|
||||
**禁止重放** `/data/paper-index/collected/ismir2026-prepared10-v1` 的文件。导入单元 `paper-atlas-ismir2026-import10-v1.service`、invocation `538ce9c9324a4f59b5bb49d3094d5265` 已成功结束;该阶段建立了 `partial10/140` 父状态;后续累计结果见下方最新段,不得沿用已消费的 `0/140` 父状态。保护基线恢复包为 `/data/paper-index/validation/ismir2026-baseline10-recovery-v1.tar.gz`,验收恢复包为 `/data/paper-index/validation/ismir2026-import10-acceptance-recovery-v1.tar.gz`;本机 paper-index 工作目录有同名副本及独立检查记录。
|
||||
|
||||
## 2026-11-18:新增 21 条,累计 31/140
|
||||
|
||||
第二批21项已通过逐篇研究内容审查、新鲜只读父状态检查、保护基线及独立入库验收。新增论文实体、来源和edition均为0;新增来源记录、投稿、录用观察各21条,新增173条来源标签证据。累计31条录用来源观察、267条标签证据,覆盖仍为 `partial31/140`,**109项待核验**。三十一项摘要都保留在原始来源记录中,既有规范资料没有被覆盖。
|
||||
|
||||
本批明确保留不同研究工作,例如UOT-IR音乐压缩与SymphonyGen交响生成,以及AI混合音轨占比回归与BAMM真实广播检测;不因作者或主题相近合并。源关键词和展示方式继续使用各自namespace,没有推断全局人名别名、原始OpenReview投稿ID或研究类型。
|
||||
|
||||
[第二批逐条验收及恢复记录](coverage/ismir2026-import21-2026-11-18.json)包含21组生产行ID、源码提交、全部证据哈希和新计数。保护并核验了6,415篇作品的规范资料、旧来源与标签,以及所有旧全局投稿/决定/来源关系指纹;新快照另导出21个完整实体包逐字段核验。首批10项保留;paper2757497的实际查询仍显示accepted及原ICLR一次拒稿下的两条来源观察。
|
||||
|
||||
本批使用固定源码 `3754545003e1f96b88c6533cc3a69dc557f07305`,已在远程 `ces-control` CI1257通过556项测试,含真实PostgreSQL只读增量门禁回归。该修复解决预检对Mapping行对象切片的错误,未弱化覆盖校验;旧的只读准备v1失败时没有写库,修复后v2重新检查通过。默认运行时未因本批操作切换。
|
||||
|
||||
**禁止重放** `/data/paper-index/collected/ismir2026-prepared21-v2`。服务 `paper-atlas-ismir2026-import21-v1.service`、invocation `78077b901a274426bb161ef8d69963ad` 已成功结束。新批次必须从当前 `partial31/140` 准备,不能再用 `10/140` 或 `0/140`。保护基线恢复包为 `/data/paper-index/validation/ismir2026-baseline21-recovery-v1.tar.gz`,验收恢复包为 `/data/paper-index/validation/ismir2026-import21-acceptance-recovery-v1.tar.gz`;本机paper-index目录有同名副本与独立检查结果。
|
||||
|
||||
新的固定范围盘点仍为274/278个年度有数据;本批提高ISMIR来源覆盖,没有填补其他空年度,也没有关闭公开拒稿或最终全库去重任务。Foundry就绪检查200,SQLite备份继续只读。
|
||||
@@ -0,0 +1,73 @@
|
||||
# MM 2026 出版条目采集与恢复
|
||||
|
||||
`collect mm2026-publications` 是包内可复用的冻结快照规范化命令,读取完整注册目录和官方节目,不连接数据库、不请求网络、不下载PDF。总体仍为13/17,来源规范化、测试通过或节目匹配都不能替代身份批准、生产入库或最终去重。
|
||||
|
||||
## 使用
|
||||
|
||||
在已配置的服务器环境中,`--registry-snapshots` 是包含 `mm2026-container-registry-v2/` 的父目录;`--program-snapshots` 是包含 `mm2026-program-and-public-probes-v1/` 的父目录。父论文集正文使用独立固定快照,不能从不存在的文章ISBN字段猜测:
|
||||
|
||||
```sh
|
||||
paper-atlas collect mm2026-publications \
|
||||
--registry-snapshots /data/paper-index/collected \
|
||||
--program-snapshots /data/paper-index/collected \
|
||||
--parent-snapshot /data/paper-index/collected/mm2026-publication-metadata-v1/parent.json \
|
||||
--out /data/paper-index/collected/mm2026-publications-native-v3
|
||||
```
|
||||
|
||||
输出目录必须不存在。失败输出保留 `failed.json`,完整 `collection-report.json` 只在全部校验成功后原子生成;不要覆盖输出或因旧进度文件、日志安静重跑。先观察实际单元/报告,再选择新的恢复目录。
|
||||
|
||||
## 注册目录与节目分别保留
|
||||
|
||||
源目录基于 Crossref [官方精确过滤器](https://www.crossref.org/documentation/retrieve-metadata/rest-api/rest-api-filters/) 的 `container-title` 和出版者member320,实际捕获8页、1918个不同DOI。每页请求、时间、原始正文、字节哈希、总量与唯一DOI均核验。父DOI为 `10.1145/3767308`,题名为 *Proceedings of the 34th ACM International Conference on Multimedia*,明确事件为MM2026;文章均注册为 `proceedings-article`。
|
||||
|
||||
ISBN查询和父Unixref XML只返回父项,不能代表逐篇目录。宽泛 `query.container-title` 的检索总量也不能当会议篇数。当前1918是这个明确论文集过滤范围内的注册条目数,不是已经去重的科研作品数、所有MM workshop或全部28个来源track的出版全集。
|
||||
|
||||
每个出版条目都输出,稳定外部标识为真实DOI。身份标识仅为 `doi`,投稿侧标识仅为 `publication_doi`;不推断原始OpenReview/forum/投稿号或arXiv标识。来源名 `mm_registry_program` 明确表示派生的注册/节目材料,不能把其中所有字段都说成Crossref原文。
|
||||
|
||||
原始注册响应及父记录复制到 `evidence/registry/`,完整节目及四个既有403回执复制到 `evidence/program/`,不重新请求受限入口。每项元数据保留注册正文的文件/哈希、数组位置和规范JSON对象哈希;常用书目字段另作投影,完整引用、资助等原始字段仍在原始响应中,未丢弃也不在每条JSONL重复膨胀。
|
||||
|
||||
## 分轨、摘要与版本差异
|
||||
|
||||
按原始题名的 Unicode/字母数字规范化与完整作者多重集合比对,1726 项有唯一节目对应。原始题名没有候选时,再比较经过受限排版投影的题名:只移除平衡、无属性的 `i`、`scp`、`sup`、`sub` 标签,保留上下标内容,HTML 实体只解码一次。未知或不平衡标签、属性、注释及声明不参与这种投影;不使用模糊题名或人物别名。
|
||||
|
||||
26 项新增对应均保留完整作者多重集合一致、唯一节目 ID 以及逐项复核名单,复制节目原文的摘要、track 和展示上下文。原始注册题名、作者和节目字段继续逐字保留,比较文本单独存入 `program_title_formatting_comparison`;这只证明节目材料对应,不批准科学作品合并。原始题名已有作者分歧时,不通过排版回退绕过。其余 122 项原始作者分歧、1 项排版题名下的作者分歧、43 项题名差异,共 166 项仍在 `proceedings_unclassified`。
|
||||
|
||||
| 输出分区 | 注册条目 |
|
||||
| --- | ---: |
|
||||
| main | 1555 |
|
||||
| grand_challenge | 77 |
|
||||
| interactive_art | 17 |
|
||||
| datasets | 41 |
|
||||
| reproducibility | 1 |
|
||||
| bni | 22 |
|
||||
| open_source | 11 |
|
||||
| doctoral_symposium | 2 |
|
||||
| demo_and_video | 23 |
|
||||
| industry_talk | 3 |
|
||||
| proceedings_unclassified | 166 |
|
||||
|
||||
注册摘要 1918 项仍全缺失。1743 项有明确对应的节目摘要:1717 项使用 `abstract_source=official_program_exact_metadata_correspondence`;新增 26 项使用 `official_program_reviewed_title_formatting_correspondence`,两者都逐字复制节目原文。报告分别列出 `abstracts_from_literal_program_correspondence` 和 `abstracts_from_reviewed_title_formatting`;保留的汇总字段 `abstracts_from_exact_program_correspondence` 为 1743,表示两类非模糊对应之和。
|
||||
|
||||
其余 175 项保持摘要缺失,包括 9 项节目本身没有摘要及 166 个版本待核条目。注册原始摘要仍标为缺失,不能把节目内容说成出版者摘要。
|
||||
|
||||
8 个作者仅有单一姓名字段,保留登记值并标记,不猜测 given/family;37 个含排版标记或实体的题名仍保留字面原文和原始标记。排版比较保留上下标字符,不改写原始字段。
|
||||
|
||||
完整2143个节目ID和4326个展示位置仍在匹配证据或391个未匹配节目中,28个原始track均保留。391个节目与166个出版版本待核来自两种观察口径,不能直接相加当成不同论文。Workshop、艺术、行业等贡献不能因为不在这个论文集范围就被当成拒稿、撤稿或不值得保留。
|
||||
|
||||
注册出版支持来源侧 `status=accepted` 的发表观察,`raw_decision` 明确不是原始签署决定;出版日期仅放在注册元数据中,不伪造决定日期。`mm_registry_program:crossref:type` 与 `mm_registry_program:mm_program:track` 分别保存注册书目类型和节目 track;共同的派生来源前缀满足写入器的来源校验,两个标签命名空间仍相互独立,原始平台字段与证据继续保留在各自元数据中,`research_type_classified=false`:书目类型不是研究类型,行业、艺术、demo条目不能默认变成主轨研究论文。
|
||||
|
||||
## 只读数据库预检与导入边界
|
||||
|
||||
每个分区都有带count/SHA256的manifest,coverage始终为partial。规范化完成后,对实际报告中的全部输出文件运行现有 `audit source-identities --files ... --pg-service foundry-research --pg-schema research --out <新目录>`,不能靠已有66个arXiv/DOI候选代表全目录。预检使用只读一致性快照,并保存来源/标识/历次投稿/逐次决定的证据包;缺失或截断必须单列。
|
||||
|
||||
`approved_for_import=0`,每项 `scientific_work_identity_approved=false`。DOI绑定可能有误,精确题名/作者匹配不证明科学作品身份,没有匹配也不证明新作品。进入增量导入前仍须审查已有绑定、科学版本与候选历史,做新鲜保护检查、完整基线/备份,再使用原生写入者。不得覆盖旧拒稿、后续录用、标签、原始元数据或人物姓名版本;既有已消费队列不可重放。
|
||||
|
||||
## 恢复
|
||||
|
||||
完整注册目录及独立节目比对恢复包:`/data/paper-index/validation/mm2026-container-registry-recovery-v2.tar.gz`,SHA256 `ddb05a8134f6437b3fb40250e45ed6350d55431d953812abda3300871b6dd0ac`。其中含注册源目录、父记录 `validation/mm2026-parent.json`、逐项比对和请求方法证据。安全解包后可把该父记录路径传给 `--parent-snapshot`。
|
||||
|
||||
节目原始来源恢复包见 [MM来源说明](mm2026-source-status.md)。本地独立核验备份位于 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/`。固定回执/正文哈希变化、总量变化、未知注册上下文或节目对应改变都须重新核查,不能放宽为无条件接受。最终全库去重、近似题名作者、标识冲突、摘要、约束、覆盖与查询审计仍是独立必做阶段。
|
||||
|
||||
实际导入回归在隔离的 SQLite 和 PostgreSQL 测试库中运行全部1918条、全部11个输出分区,核验原始字段、1918个书目类型标签、1752个节目标签、publication_doi 身份、partial覆盖,以及已有作品的规范信息和旧拒稿决定不被覆盖。这些测试不会写入生产 research schema;测试通过仍不替代逐条生产身份审阅。
|
||||
|
||||
旧的 `native-v1`、`native-v2` 来源目录及已消费的导入队列不可覆盖或重放。新版本写入新的输出目录,先核对已导入记录的源字段和分轨是否变化;新增来源关联仍须新鲜身份预检、保护基线和独立验收。程序输出的分区预期数量已改变,未来增量 manifest 必须明确保留旧覆盖父值并声明新预期值,不能重置已导入计数。
|
||||
@@ -0,0 +1,139 @@
|
||||
# ACM MM 2026 官方节目来源状态
|
||||
|
||||
2026-11-17 已冻结官网正式日程、技术日程、征稿/评审政策、两个日程脚本和脚本明确引用的 `program-data.json`。本页记录可恢复的来源证据,不宣称这些节目贡献全部都是已发表科研论文;总体仍为 13/17,第 14 步未完成。
|
||||
|
||||
## 已取得的数据
|
||||
|
||||
[官方技术日程](https://2026.acmmm.org/site/technical-programme.html)明确加载 `technical-programme.js`,该脚本读取 [program-data.json](https://2026.acmmm.org/site/program-data.json)。原始 JSON 为 10,808,209 字节,生成时间为 `2026-11-14T18:36:26.860677+00:00`,保存了 5 天、20 个房间、157 个日程场次。
|
||||
|
||||
递归盘点得到 **4,326 次论文样式的节目观察,涉及 2,143 个不同节目 ID**。额外 2,183 次是直接列表和子场次/多次展示中重复的观察。相同 ID 的标题、作者、摘要、track 和机构字段一致;所有来源位置均保留,不能把重复展示累加成新论文。
|
||||
|
||||
独立盘点的逐 ID 来源 track 数量如下。它们来自 `paper.track`;157 个外层日程条目的 `entry.track` 均为空,不能用外层空字段覆盖真实来源分类。展示形式和学术 track 也不是同一概念。
|
||||
|
||||
| 来源 track 原文 | 不同节目 ID 数 |
|
||||
| --- | ---: |
|
||||
| MRAC | 35 |
|
||||
| SUMAC | 9 |
|
||||
| NeuroMM | 10 |
|
||||
| IMPACT | 8 |
|
||||
| RichMediaGAI | 10 |
|
||||
| MMSports | 16 |
|
||||
| MUWS | 5 |
|
||||
| EReL-MIR | 7 |
|
||||
| GreenMM | 10 |
|
||||
| DFF | 12 |
|
||||
| Main Track | 1703 |
|
||||
| Grand Challenge | 97 |
|
||||
| Doctoral Symposium | 4 |
|
||||
| Datasets | 50 |
|
||||
| Interactive Art | 23 |
|
||||
| Reproducibility | 1 |
|
||||
| Demo and Video | 27 |
|
||||
| Open Source | 14 |
|
||||
| Industry Talk | 7 |
|
||||
| BNI | 25 |
|
||||
| QoEVMA | 4 |
|
||||
| HUMA | 14 |
|
||||
| WIPAI | 8 |
|
||||
| AMI | 11 |
|
||||
| Mulsemedia | 9 |
|
||||
| UAVM | 11 |
|
||||
| MCHM | 7 |
|
||||
| JAV-CG | 6 |
|
||||
|
||||
共有 1,927 个节目 ID 带摘要,216 个缺少摘要。Main Track 的 1,703 个 ID 中,有两项同时缺少作者与摘要:
|
||||
|
||||
- `mfp4971`:ASDG: MLLM-based Audio Semantics for Dance Generation。
|
||||
- `mfp7801`:Dynamic Late Fusion Multiple Kernel k-means Clustering based on Adaptive Sample Selection and Alignment。
|
||||
|
||||
两项仍有科学标题,原文没有明确撤稿或拒稿标签;因此只能标记字段缺失、暂缓关联,不能推断负面决定。非主轨的 workshop、grand challenge、行业报告、艺术、博士生等项目还需逐轨核对存档与贡献类型,不能把全部 2,143 个节目 ID 都计作已发表科研论文。原始公开联系邮箱不进入独立盘点、人物别名或论文标识。
|
||||
|
||||
## 公开负面决定的访问边界
|
||||
|
||||
[官方 CFP](https://2026.acmmm.org/site/cfp-guidelines.html)链接至 [OpenReview 群组](https://openreview.net/group?id=acmmm.org/ACMMM/2026/Conference)。匿名群组元数据可读,但根据群组给出的原始 submission invitation、rejected/withdrawn/desk-rejected venue ID,进行的四次 `limit=1` 匿名 notes 核查全部返回 **HTTP 403**。
|
||||
|
||||
保存了准确查询、时间、状态、响应原文和哈希;没有认证、重试或绕过权限。403 是已观察接口的访问限制,不代表零条拒稿、从未拒稿,或排除了其他可能公开的来源。评审公开规则也不能替代具体最终决定记录。当前不从评分、评审推荐、节目缺席或空字段制造 rejected/withdrawn 事件。
|
||||
|
||||
## 产物与恢复
|
||||
|
||||
原始目录(均相对 `/data/paper-index/collected/`):
|
||||
|
||||
- `mm2026-program-policy-research-v1`:六页官方日程和分轨/评审政策。
|
||||
- `mm2026-navigation-research-v1`:官网明确加载的导航原文。
|
||||
- `mm2026-program-data-discovery-v1`:两份脚本及公开群组元数据。
|
||||
- `mm2026-program-and-public-probes-v1`:完整节目 JSON 和四个匿名访问限制回执。
|
||||
|
||||
独立盘点:`/data/paper-index/validation/mm2026-independent-program-inventory-v1.json`,包含每个节目 ID 的所有来源位置、track、标题/作者、摘要可用性及核心版本数量。全部响应原文、回执及盘点已在本机备份,并独立核验了原始字节哈希。
|
||||
|
||||
| 证据 | SHA256 |
|
||||
| --- | --- |
|
||||
| program-data.json | 08e2976ecba2d0d535000b0935f41c85076ec674e518a1df2c5843d9bde4d083 |
|
||||
| 独立逐 ID 盘点 | 375dd975fdff023f2ff591d47066ec616c2bfc3b717b6016ab162d74d587f3e7 |
|
||||
| 完整来源恢复包 | 8247149ed7437a36c13ca2b6128431e2b0f3f8a388a5ddefb62427b0c1ce6f8d |
|
||||
|
||||
服务器恢复包:`/data/paper-index/validation/mm2026-source-discovery-and-inventory-recovery-v1.tar.gz`;本地备份:`/Users/zzyang/Documents/Codex/2026-11-06/paper-index/` 同名文件。先安全解包并核验回执,再做规范化,不能重新抓取 403 入口作为恢复方式。
|
||||
|
||||
## 可复用采集命令
|
||||
|
||||
在已配置的服务器环境中运行,`--snapshots` 是包含 `mm2026-program-and-public-probes-v1/` 的父目录;恢复时先安全解包上述已核验恢复包。命令只读快照,不请求网络、不连接数据库,不重新访问403入口,也不下载PDF:
|
||||
|
||||
```sh
|
||||
paper-atlas collect mm2026-program \
|
||||
--snapshots /data/paper-index/collected \
|
||||
--out /data/paper-index/collected/mm2026-native-v1
|
||||
```
|
||||
|
||||
输出目录必须不存在;已存在、失败或部分输出均不能覆盖重跑,应先查看报告,另选新的目录。`collection-report.json` 仅在全量验证成功后生成;失败保留 `failed.json`,不冒充完成。
|
||||
|
||||
采集器核验固定回执、正文哈希、URL、抓取时间、完整逐轨数量和字段缺失盘点。28个逐轨 JSONL 保留原始标题、作者、机构、逐字摘要以及同一节目ID的全部展示位置、时间、poster/oral号和展示kind;157个场次及子场次/非论文活动另存 `sessions.jsonl`。外层空track不覆盖 `paper.track`,不同核心版本、未知字段或未处理嵌套需要重新审查,不能静默丢弃或取第一条。公开联系邮箱数组只留在原始证据中,不进入规范记录。`evidence/` 保留原始节目JSON、完整回执及四个既有403响应。
|
||||
|
||||
这些文件的格式为 `mm_program_observation_v1_not_ingestion_record`,所有项 `approved_for_import=false`。节目ID仅为 `mm_program:program_id` 来源定位符,track标签使用 `mm_program:track`;不建立科学作品、原始投稿ID或最终决定,不推断 DOI/arXiv/OpenReview 标识、论文类型、拒稿或撤稿。行业报告、艺术、workshop及其他来源贡献均保留,尚需逐轨核对贡献类型与正式出版依据。
|
||||
|
||||
本批尚未形成生产导入队列。后续做贡献类型、作品/出版身份及既往投稿审查,再生成经批准的增量队列。已消费的其他会议导入队列不可重放;最终全库去重和完整审计仍为独立必做阶段。
|
||||
|
||||
## 直接审计节目身份与既往历史
|
||||
|
||||
全量节目观察可以直接进入只读候选审计,无需伪造成录用论文队列:
|
||||
|
||||
```sh
|
||||
PGSERVICEFILE=/etc/paper-atlas/pg_service.conf \
|
||||
paper-atlas audit source-identities \
|
||||
--mm-program-report /data/paper-index/collected/mm2026-native-v1/collection-report.json \
|
||||
--pg-service foundry-research --pg-schema research \
|
||||
--out /data/paper-index/collected/mm2026-preflight-v1
|
||||
```
|
||||
|
||||
`--mm-program-report` 与常规 `--files` 互斥。审计先核对冻结原文、全部28轨的逐文件哈希/数量,以及每个节目ID的完整字段和展示上下文;缺轨、重复ID、篡改后的标识/决定或上下文均在数据库连接前拒绝。报告中记录的绝对文件路径仅作来源证据,实际读取报告同目录中的对应文件。
|
||||
|
||||
候选审计中的 `status=unknown` 仅为内部查询占位符,不代表原始提交的最终状态。节目ID以 `mm_program` 来源键查询,不转成全局 DOI、arXiv 或 OpenReview 标识;保留原始 `listed_in_official_program` 观察与全部展示位置。所有贡献类型、出版身份、最终决定核验标志均为false;缺作者的两项也保留并标记。
|
||||
|
||||
查询使用 PostgreSQL 只读一致性快照,候选作品的原始来源、标识、历次投稿和逐次决定压缩保存并记录哈希;截断会显式标记。标题/作者候选匹配不是作品身份批准,没有匹配也不证明是新论文。既往拒稿和后续录用不能互相覆盖,评分或评审推荐不能充当最终决定。输出目录必须不存在;遇到已有输出或静默运行,先检查具体单元和报告,不重启重放。
|
||||
|
||||
## 已完成的规范化、只读审计与出版核对
|
||||
|
||||
采集器提交 `c8db611` 和直接节目审计提交 `09a2d4e` 已通过PR #65、#66由用户合并。授权远程runner的535/543项测试、wheel构建安装和安装后全量命令检查通过;没有本机测试或构建。服务器规范化和只读候选报告也已独立验收,并在本机备份、复核全部2143项与814个候选证据包。
|
||||
|
||||
| 只读候选分类 | 来源节目项数 |
|
||||
| --- | ---: |
|
||||
| 标题与完整作者多重集合一致,仍需作品核查 | 660 |
|
||||
| 仅部分作者重叠 | 153 |
|
||||
| 无精确标题候选,不证明是新作品 | 1329 |
|
||||
| 有标题候选但无作者一致性 | 1 |
|
||||
|
||||
全部4326个展示上下文一致,814个证据包均无截断。24篇候选作品涉及24次已有负面投稿、37条负面来源观察;37不能直接算作37次拒稿。所有候选均尚未批准身份或入库,旧负面决定必须在之后关联时保留。
|
||||
|
||||
对已有作品的原始arXiv记录、标题/完整作者多重集合和MM2026语境进一步筛查,得到67个 Main Track 出版DOI核对候选。独立查询其明确给出的DOI及父论文集,共68次单次请求,67次HTTP200、1次404,没有认证或重试。父注册记录为[第34届ACM Multimedia论文集](https://doi.org/10.1145/3767308),ISBN `9798400722134`,出版者ACM,出版年2026。
|
||||
|
||||
其中66项的原始arXiv引用DOI与注册DOI一致,按既有审计规则规范化的标题、完整作者多重集合、出版年、出版者、明确会议名和论文集标题也一致。注册文章未提供ISBN,不能伪称逐篇ISBN已核验;归属依据是文章明确给出的会议和论文集字段,以及独立的原始arXiv语境和DOI。仍需逐项作品/版本审查及新鲜数据库保护检查,**不是66项已入库或最终身份批准**。
|
||||
|
||||
`mfp7067` 对应既有候选作品2749947,原始arXiv提供的 `10.1145/3767308.3836169` 返回404,暂缓。404不是撤稿/拒稿证据,也不据此删除或修正既有标识;需独立核对后再处理。
|
||||
|
||||
完整会议出版名单仍需继续核对。实际ISBN查询只返回父论文集一个条目,父DOI的公开Unixref XML也只含该父DOI、0个 `conference_paper` 节点;都不能代表逐篇名单。按会议名检索得到的 `total-results` 是出版者全年范围内的宽泛检索量,不能当作MM篇数;必须逐项核对明确会议/论文集字段、节目与分轨归属,并保留未匹配项。全部28个来源track继续保留,不能把本批Main Track候选当成会议完整覆盖。
|
||||
|
||||
| 已独立验证的恢复证据 | SHA256 |
|
||||
| --- | --- |
|
||||
| 全轨规范化恢复包 | b119081618b8ce98037dc139be159512f9fdb86a0ccfb127c8b8a1c5722f3d87 |
|
||||
| 只读候选/814证据包恢复包 | 42d0d0404bba04e64e638344b050ad4e7287c3e4c75a777b7069ecc7d7a6ed60 |
|
||||
| 67项DOI及父论文集原始注册回执恢复包 | dd7a0ff76f12070f403cc7352ac54269e454758d807896a086c58c1c1ac7b798 |
|
||||
|
||||
服务器恢复包位于 `/data/paper-index/validation/`,分别名为 `mm2026-native-recovery-v1.tar.gz`、`mm2026-preflight-recovery-v1.tar.gz`、`mm2026-publication-metadata-recovery-v1.tar.gz`;本地同名备份在 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/`。来源目录为 `collected/mm2026-native-v1`、`mm2026-preflight-v1`、`mm2026-publication-metadata-v1`。先核对完整manifest及实际单元结果,不能用旧进度文件的running字样重启。其他会议已消费的导入队列仍禁止重放;本轮均为来源核查、只读候选审计,没有生产写入。
|
||||
@@ -51,3 +51,33 @@ paper-atlas collect recsys2026-program \
|
||||
服务器恢复包 `/data/paper-index/validation/remaining2026-listings-research-v2-recovery.tar.gz`,本地备份 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/remaining2026-listings-research-v2-recovery.tar.gz`,SHA256 为 `f5ef7c2e73d46712146e4d6711ad5ad562e16b8b55c0cd75d1041cd0b81142cb`。安全恢复原始目录、核对哈希后重跑上述命令。
|
||||
|
||||
完整来源与独立 263 项基准已保存在测试夹具,所有测试、构建和安装验证交给指定的 `ces-control:host` runner。生产导入前还需独立核验规范化结果,并检查当前数据库已有 DOI 绑定及历次投稿。最终全库去重审计仍是独立必做阶段。
|
||||
|
||||
## 2026-11-17 PostgreSQL 补录验收
|
||||
|
||||
采集器仍只生成待审查文件;本节记录随后单独完成的生产补录。47 条来源记录已关联到 47 篇既有作品,依据是已绑定的同一发表 DOI、规范化标题、完整作者集合,以及原始 arXiv 来源明确引用的同一 DOI。逐条当前资料和历史包已与只读预检核对;没有新增论文实体、人物别名、arXiv 标识或原始 OpenReview ID。
|
||||
|
||||
| track | 已导入观察 / 来源应有观察 |
|
||||
| --- | ---: |
|
||||
| demo | 1 / 18 |
|
||||
| doctoral_symposium | 2 / 10 |
|
||||
| industry | 18 / 83 |
|
||||
| long | 13 / 65 |
|
||||
| past_present_future | 2 / 13 |
|
||||
| reproducibility | 1 / 4 |
|
||||
| research_practice_notes | 4 / 34 |
|
||||
| resource | 2 / 7 |
|
||||
| short | 4 / 29 |
|
||||
|
||||
九个 track 全部保持 **partial**。总计 47/263 个来源观察已导入,216 个待审查,其中包括三组重复 DOI 的六项观察。来源的 263 项观察对应 260 个不同 DOI,不能把观察数量当成不同科学论文数量。
|
||||
|
||||
本批新增一个来源、九个 edition、47 条来源记录、47 次投稿/47 条录用观察、47 个 `publication_doi` 投稿标识及47条来源类型标签证据关联。源摘要分别保存,47 篇既有作品的规范摘要保持不变;原始程序字段、来源链接、时间和类型标签也逐条核验。Research/PastPresentFuture/Doctoral Symposium 仍属于来源 namespace,不强制映射成统一研究方法类别。
|
||||
|
||||
独立验收确认:5,812 篇受保护旧作品及其旧资料、来源、标签、标识不变,全部旧投稿/决定原文与关联历史不变;foundry-core `/readyz` 为 200,受保护 SQLite 备份大小/只读权限不变。论文实体仍为 3,567,888;不能用 `max(id)` 代替行数。
|
||||
|
||||
- 完整验收:`/data/paper-index/collected/recsys2026-import47-v1/postgresql-postcheck-v1.json`,SHA256 `7c7e2dcd06bdbc839666740a989c1f187fa47290800a9055cf4540be64336cf4`。
|
||||
- 保护基线:`/data/paper-index/collected/recsys2026-import47-baseline-v1/baseline.json`,SHA256 `aa86d51ff18774a505b1c59510ad867c6221c152334dc4983d31c5290cb8ac0f`。
|
||||
- 导入前完整队列/基线恢复包:`/data/paper-index/validation/recsys2026-reviewed47-recovery-v1.tar.gz`,SHA256 `ae054a6447e8c2864fd8e365867cd3308dee974f375fafff5653fff2a51044e1`;本地同名备份位于 `/Users/zzyang/Documents/Codex/2026-11-06/paper-index/`,已核验全部 47 个原始身份包、九份输入和基线。
|
||||
|
||||
执行代码固定在提交 `8c147f5e6edda900e2eed19372ab031f661b31c5` 的独立验证目录,85 个包文件校验通过;24 个写入关键文件与此前已验收版本一致。授权 `ces-control:host` runner 的 CI1230/job1327通过525项测试及安装后实际RecSys263项/260DOI验证。没有改变服务器默认运行代码目录或共享数据库配置。
|
||||
|
||||
**九份输入均已消费,禁止重放。** 恢复时先检查实际 PostgreSQL、执行回执和终态验收;不能因日志安静、观测超时或重新启动 Goal 就复跑。后续216条须另建审查与增量队列,并重新取当前基线。本批通过不等于整届 RecSys、公开拒稿覆盖或最终全库去重已完成。
|
||||
|
||||
@@ -0,0 +1,52 @@
|
||||
# ReviewArena 镜像的索引字段投影
|
||||
|
||||
ReviewArena 是公开评审镜像,固定版本为 `c2978add17c2099219eaddbc2599974d69d4d09b`。上游数据卡描述七类会议、2020–2026 年的部分来源;这不是逐会议、逐年度的公开决定全集。文件清单有 15 个 Parquet 分片,覆盖 COLM、CoRL、EMNLP、ICLR、ICML、NeurIPS、TMLR。这里不包括 UAI,不能用它填补 UAI 缺口。
|
||||
|
||||
- [固定版本上游数据卡](https://huggingface.co/datasets/anonymousNeurIPS2026submission4281/reviewarena/blob/c2978add17c2099219eaddbc2599974d69d4d09b/README.md)
|
||||
- [固定版本文件元数据 API](https://huggingface.co/api/datasets/anonymousNeurIPS2026submission4281/reviewarena/tree/c2978add17c2099219eaddbc2599974d69d4d09b/data?recursive=false&expand=false)
|
||||
- [Arrow ParquetFile 投影接口](https://arrow.apache.org/docs/python/generated/pyarrow.parquet.ParquetFile.html)
|
||||
|
||||
## 运行
|
||||
|
||||
仅在已授权远程 runner 安装 `paper-atlas[mirrors]`,使用固定 PyArrow 26.0.1 二进制 wheel。本机不执行安装、构建或项目测试。常规 CI 包含投影、安全边界和镜像决定语义测试;真实网络采集由独立的 `ReviewArena metadata projection` workflow 手动触发,指定已通过 CI 的代码 ref 和 split,例如 `emnlp`。该 workflow 仍使用 `ces-control` / `aliyun-ces-control`,在服务器固定数据目录生成输出和恢复归档,不更改默认 runtime 或数据库。
|
||||
|
||||
包内命令如下,`--out` 必须是既有父目录内的新目录:
|
||||
|
||||
```sh
|
||||
paper-atlas collect reviewarena-metadata \
|
||||
--split emnlp \
|
||||
--out /data/paper-index/collected/reviewarena-emnlp-reviewed-version
|
||||
```
|
||||
|
||||
输出目录已存在时拒绝执行。遇到禁止访问、来源变化或读取失败,保留失败报告,不自动重试或切换到整文件下载。不能把原始访问拒绝解释为没有拒稿。
|
||||
|
||||
已保存的完整元数据投影可以离线重新解释决定字面值。提供原 `collection-report.json` 及其已独立验证的精确文件 SHA-256,程序核对全部元数据文件哈希、逐行来源身份和原始字段,生成新目录;不会再次访问上游 Parquet:
|
||||
|
||||
```sh
|
||||
paper-atlas collect reviewarena-metadata --split emnlp \
|
||||
--from-projection /data/paper-index/collected/<既有投影>/collection-report.json \
|
||||
--projection-sha256 <已验证的原报告SHA256> \
|
||||
--out /data/paper-index/collected/<新解释版本>
|
||||
```
|
||||
|
||||
手动 workflow 对应输入为 `projection_name` 与 `projection_sha256`。离线报告的 `metadata_bytes_read` 为本次请求字节数 `0`,`source_projection_bytes` 保留原始列块投影字节数;原报告哈希建立恢复链,原始记录与来源定位不变。
|
||||
|
||||
## 实际读取和保存范围
|
||||
|
||||
仅投影以下字段:`forum_id`、`conference`、`year`、`track`、`venue_id`、`paper_number`、`title`、`abstract`、`authors`、`keywords`、`venue`、`decision`。
|
||||
|
||||
程序先读取文件魔数、尾部长度和有界 Parquet 页脚,在内存中解析 schema 和列块偏移,然后只请求这些列的字典/数据页。Arrow 可能请求较大的页脚窗口,未获准区域由本地稀疏文件合成零字节,不会为该窗口补下载相邻正文页。逐个选中列块还要检查不与未选列块重叠。单文件上限为 64 MiB 所需字节、512 个范围请求和 100,000 行;响应必须是准确的 HTTP 206 / Content-Range,ETag 变化时停止。该边界适用于每个分片,不能据此启动其他分片的未授权全文件下载。
|
||||
|
||||
不读取 `markdown`、`reviews_json`、`decision_comment`、`author_rebuttal` 等正文或评审内容的数据页;不保存原 Parquet、页脚原文或未选字段统计值。保存的 JSONL 只有上述索引字段、来源定位和审计标记。报告记录每个请求范围及其 SHA-256、选中列名、页脚哈希、投影数量和保存文件哈希,便于独立复核。
|
||||
|
||||
上游 LFS SHA-256 与大小是固定版本 API 提供的对象声明。由于不下载整个 Parquet,本流程**不声称验证了完整 Parquet 内容哈希**。范围对象大小/ETag 和各段哈希用于检查本次读取的一致性。
|
||||
|
||||
## 决定和身份边界
|
||||
|
||||
`mirror_decision_category` 只解释明确的镜像字面值;未知、缺失或未审核的决定保持 `unknown`,不会根据 venue、是否录用、摘要缺失或目录缺席推断拒稿。类别数量是镜像行数,不能解释为独立作品数量或原始官方决定数量。
|
||||
|
||||
真实 EMNLP 2023 固定分片包含 2,009 行:原始决定为 `Accept-Main` 994 行、`Accept-Findings` 1,015 行,原始 `track` 却全部为 `main`。两种已核对字面值属于镜像录用声明;`mirror_decision_track_claim` 分开保留其明确的 main/findings 声明,原 `track` 和所有 `raw_fields` 不被重写。`Accept (Poster)` 等展示形式不能自行推出主会归属。此分片没有明确负面决定,不能用于补齐拒稿缺口,更不能由此断言 EMNLP 原始平台没有公开拒稿。首次实际投影仅请求 1,720,498 字节;原文件 90,794,691 字节,未下载完整文件。
|
||||
|
||||
上游 `forum_id`、`venue_id`、投稿号、作者和决定均作为原始声明保留。`forum_id` 不直接成为全局 OpenReview 标识;稳定行定位由数据集、固定 commit、文件和行号组成。输出 `identifiers` 为空,原始论坛身份和决定签名标记为未验证,保留匿名和缺失作者。源数据中的字符串命名不能被当成真实人员别名。
|
||||
|
||||
输出故意不是可直接 `import records` 的规范化队列,没有授予导入或合并权限。下一步必须检查原始来源、跨来源重合和作品身份,再决定哪些来源观察能进入生产。即使明确镜像拒稿,也要保留该投稿历史并继续查找后续其他会议的录用;不能把“目前没有关联录用”表述为从未录用。本采集不代替最终全库去重、决定冲突、来源覆盖和查询验收。
|
||||
@@ -0,0 +1,57 @@
|
||||
# 出版来源版本与覆盖修正
|
||||
|
||||
`paper-atlas maintain source-revisions` 用于同一出版清单的新版本:稳定 DOI 全集、标题、作者和决定不变,但官方分轨、摘要或对应来源证据已经过重新核对。它不导入新论文、不合并实体、不推断拒稿,不把出版分类当成研究类型。
|
||||
|
||||
例如,同一清单的“未分类”由 192 条减少到 166 条,而主会等分区增加。直接放松增量导入的计数保护会隐藏清单丢失。本命令先核验完整的新旧清单及数据库快照,再单独修正版本;普通 `import records` 的单调计数规则保持原样。
|
||||
|
||||
## 准备完整证据
|
||||
|
||||
在工作区外保存 inventory JSON,包含以下字段:
|
||||
|
||||
| 字段 | 内容 |
|
||||
| --- | --- |
|
||||
| `format` | `1` |
|
||||
| `source`, `venue`, `year`, `event_key`, `list_url` | 明确的已有来源、会议缩写、年份、事件键和 HTTPS 出版清单地址 |
|
||||
| `previous_rows`, `rows` | 完整旧版、新版规范化来源记录数组,每版最多 10,000 条;每条包含稳定 DOI、来源字段、作者、摘要、track、决定、标签、原始来源 metadata |
|
||||
| `evidence` | 一个或多个 `{ "url": "https://…", "sha256": "…" }`,绑定已独立核验的原始证据文件 |
|
||||
|
||||
两版必须有完全相同的 DOI 集合;仅允许 `abstract`、`track`、`tags`、`metadata` 的有据变化。必须保留 collector 原始字段,不能将工作身份核查产生的额外锚点混入原始来源版本。只有明确 `accepted` 的出版观察可进入本流程。证据 URL/hash 是审核输入,本命令不联网确认它们;执行者必须另行验证文件哈希、完整性及对应关系。
|
||||
|
||||
数据库可以部分导入,甚至各分区的预期数暂时来自不同版本。现有覆盖必须为 `partial`、无错误,导入数与实际来源记录一致;预期数必须恰好属于旧版或新版的同一分区。已有目标 edition 和对应覆盖记录必须存在,目标中可以有其他来源及既有投稿。本版本不会新建 edition,也不会处理共享来源投稿、多次既有 scope curation 的再迁移,或成员集合变化;遇到这些情况会拒绝并要求单独核查。
|
||||
|
||||
## 先生成只读计划
|
||||
|
||||
```sh
|
||||
paper-atlas maintain source-revisions \
|
||||
--pg-service foundry-research --pg-schema research \
|
||||
--inventory /data/paper-index/validation/reviewed-source-version.json \
|
||||
--out /data/paper-index/validation/source-revision-plan.json
|
||||
```
|
||||
|
||||
计划包含明确数据库身份、两个完整来源版本、当前版本父节点、覆盖表原值和修正值、逐条变更,以及受影响会议 scope 内所有论文的规范记录、稳定标识、全部跨会议投稿和决定、来源关系、标签与已存在的 annotation。拒稿后录用的历史也在保护范围内,不能只校验当前会议的录用观察。
|
||||
|
||||
计划和原始证据应保存到服务器与独立恢复位置,核对完整差异及计划文件 SHA-256。执行前重新核对生产保护基线与单写入安排。输出文件必须是不存在的新路径,禁止覆盖旧报告。
|
||||
|
||||
## 按文件哈希应用并验收
|
||||
|
||||
```sh
|
||||
paper-atlas maintain source-revisions \
|
||||
--pg-service foundry-research --pg-schema research \
|
||||
--plan /data/paper-index/validation/source-revision-plan.json \
|
||||
--plan-sha256 <已审核的计划文件SHA256> \
|
||||
--out /data/paper-index/validation/source-revision-result.json
|
||||
```
|
||||
|
||||
应用时重新生成计划并要求完全一致;PostgreSQL 使用已有 writer 锁及相关表写锁,SQLite 使用立即事务。任何保护数据、当前版本或目标标识冲突都使操作失败。操作保留投稿和决定 ID、原始决定证据、规范论文文本、全部稳定标识、来源关系和现有标签;只修改计划列出的投稿/投稿标识 edition,以及来源记录的摘要与 metadata 当前投影。旧投影完整保存在版本历史中,来源原始获取时间不会伪造为修正时间。
|
||||
|
||||
覆盖只调整明确分区的预期数和来源观察数,保留 `partial`、cursor、成功时间和错误字段原值。尚未导入的分区不产生虚构投稿或来源观察。覆盖数不能被解读为科学作品数量,或所有 track、摘要、负面决定已完整。
|
||||
|
||||
同一事务显式增加可共享的可选表 `source_scope_revisions`,保持原 21 张业务表的身份。每个版本存储新来源全集、父版本、完整计划及修正前数据;应用接口只追加版本,不更新或删除旧版本。没有新建 PostgreSQL 实例。该表纳入 SQLite→PostgreSQL 的可选表迁移、全量内容哈希、约束与计数验收。`version_sha256` 是规范化计划内容哈希;CLI 的 `--plan-sha256` 是计划文件字节哈希,二者用途不同。
|
||||
|
||||
提交前在同一事务比较全部保护数据与明确预期的变更;失败则全部回滚,包括新增表和版本记录。相同计划的重放还会重新验收当前状态;如后续数据已变化则拒绝,不会重复迁移或恢复旧状态。完整计划与 before-image 用于审查和恢复准备,本命令没有自动回滚历史版本的入口;不得手工删除版本来绕开刷新保护。
|
||||
|
||||
## 后续刷新及限制
|
||||
|
||||
`import records` 检查当前版本的逐条完整来源字段;`Store.record` 也检查原始文本、作者、来源 metadata 和基础标识,防止直接调用绕回旧版摘要。来源字段必须与当前版本相符,可以额外携带已经核验的工作身份/curation 证据,不能替换来源自身字段。相同来源、会议事件和清单地址下未登记的成员必须另行核验新版本。
|
||||
|
||||
版本表通过包内接口追加,未声明数据库管理员的任意 SQL 为安全接口。新来源或新版本在导入前仍需要独立工作身份核查;姓名、标题或无候选都不证明作品身份。实现测试和版本修正验收也不替代所有导入结束后的全库去重、标识冲突、近似标题/作者、摘要、来源覆盖与查询审计。
|
||||
@@ -74,6 +74,74 @@ def load_inputs(files):
|
||||
return records, inputs
|
||||
|
||||
|
||||
def load_mm_program_inputs(report_path):
|
||||
"""Read every MM observation without turning a program into import records.
|
||||
|
||||
Reconcile the complete output against its frozen raw source before any DB
|
||||
connection. The synthetic unknown status is an audit placeholder only.
|
||||
"""
|
||||
from paper_atlas.collectors import mm2026
|
||||
|
||||
report_path = Path(report_path).absolute()
|
||||
raw_report = mm2026.safe_read(report_path, 256*1024)
|
||||
report = json.loads(raw_report)
|
||||
if (report.get('source') != 'mm_program' or report.get('records') != 2143
|
||||
or report.get('program_observations') != 4326 or len(report.get('files', [])) != 28
|
||||
or report.get('source_body_sha256') != mm2026.BODY_SHA
|
||||
or report.get('source_manifest_sha256') != mm2026.MANIFEST_SHA
|
||||
or report.get('database_written') is not False or report.get('approved_for_import') != 0):
|
||||
raise ValueError('Unreviewed MM observation report')
|
||||
root = report_path.parent
|
||||
raw_source = mm2026.safe_read(root/'evidence/program-data.json', 12*1024**2)
|
||||
raw_receipts = mm2026.safe_read(root/'evidence/manifest.json', 128*1024)
|
||||
if digest(raw_source) != mm2026.BODY_SHA or digest(raw_receipts) != mm2026.MANIFEST_SHA:
|
||||
raise ValueError('MM raw evidence changed')
|
||||
expected_rows, _ = mm2026.parse_program(raw_source)
|
||||
expected = {r['paper_id']:r for r in expected_rows}
|
||||
records, inputs, seen, tracks = [], [], set(), set()
|
||||
for item in report['files']:
|
||||
track = item.get('track_literal')
|
||||
if (track not in mm2026.TRACK_COUNTS or track in tracks
|
||||
or item.get('format') != 'mm_program_observation_v1_not_ingestion_record'
|
||||
or item.get('approved_for_import') != 0):
|
||||
raise ValueError('Duplicate or unreviewed MM track format')
|
||||
tracks.add(track)
|
||||
# Stored absolute paths are provenance, not instructions to read elsewhere.
|
||||
path = root/Path(item['file']).name
|
||||
if path.suffix != '.jsonl':
|
||||
raise ValueError('Invalid MM observation filename')
|
||||
raw = mm2026.safe_read(path, 16*1024**2)
|
||||
lines = raw.splitlines(keepends=True)
|
||||
if digest(raw) != item.get('file_sha256') or len(lines) != item.get('count') or len(lines) != mm2026.TRACK_COUNTS[track]:
|
||||
raise ValueError('MM observation file count/hash mismatch')
|
||||
for number,line in enumerate(lines, 1):
|
||||
row = json.loads(line)
|
||||
pid = row.get('paper_id')
|
||||
if not isinstance(pid,str) or pid in seen or row != expected.get(pid) or row['track'] != track:
|
||||
raise ValueError('MM observation differs from frozen source or repeats an ID')
|
||||
seen.add(pid)
|
||||
records.append({'source':'mm_program','external_id':'2026/'+pid,
|
||||
'title':row['title'],'authors':row['authors'],'venue':'MM','year':2026,'track':track,
|
||||
'status':'unknown','identifiers':{},'file':str(path),'line':number,
|
||||
'raw_line_sha256':digest(line),'source_record_hits':[],
|
||||
'identifier_hits':[],'title_candidate_ids':[],
|
||||
'input_format':'mm_program_observation_v1_not_ingestion_record',
|
||||
'program_id':pid,'program_contexts':row['contexts'],
|
||||
'source_observed_status':row['observed_status'],
|
||||
'missing_source_fields':row['missing_fields'],
|
||||
'publication_identity_verified':False,'final_decision_verified':False,
|
||||
'contribution_type_verified':False})
|
||||
inputs.append({'file':str(path),'file_sha256':digest(raw),'count':len(lines),
|
||||
'manifest_sha256':digest(raw_report),'manifest':{
|
||||
'source':'mm_program','venue':'MM','year':2026,'track':track,'status':'unknown',
|
||||
'format':'mm_program_observation_v1_not_ingestion_record',
|
||||
'scope':'Read-only lookup input only; unknown is not a source decision.',
|
||||
'source_report':str(report_path),'source_body_sha256':digest(raw_source)}})
|
||||
if seen != expected.keys() or tracks != mm2026.TRACK_COUNTS.keys():
|
||||
raise ValueError('Incomplete MM observation input; all28tracks required')
|
||||
return records, inputs
|
||||
|
||||
|
||||
def _lookup(connection, values, columns, lateral):
|
||||
"""Bound scalar VALUES allow the existing text codec to encode each key."""
|
||||
width = len(columns)
|
||||
@@ -187,9 +255,11 @@ def _atomic_json(path, value):
|
||||
temporary.replace(path)
|
||||
|
||||
|
||||
def audit(files, out, *, pg_service, pg_schema):
|
||||
def audit(files, out, *, pg_service, pg_schema, mm_program_report=None):
|
||||
# Reject input corruption before opening a database connection or reserving output.
|
||||
records, inputs = load_inputs(files)
|
||||
if bool(files) == bool(mm_program_report):
|
||||
raise ValueError('Choose source files or one MM program report')
|
||||
records, inputs = load_mm_program_inputs(mm_program_report) if mm_program_report else load_inputs(files)
|
||||
out = Path(out)
|
||||
out.mkdir(parents=True, exist_ok=False)
|
||||
(out / 'audit.lock').write_text(timestamp() + '\n')
|
||||
@@ -229,6 +299,7 @@ def audit(files, out, *, pg_service, pg_schema):
|
||||
'Name agreement does not prove person or scientific-work identity.',
|
||||
'Bound identifiers can be incorrect; review titles, authors and original evidence.',
|
||||
'Source/history limits are explicit in compressed entity packets.',
|
||||
'Program observations and unknown audit status do not establish contribution type, publication or a final decision.',
|
||||
'Not a final whole-database deduplication or coverage audit.']}
|
||||
_atomic_json(out / 'postgresql-preflight.json', report)
|
||||
return report
|
||||
@@ -243,10 +314,14 @@ def audit(files, out, *, pg_service, pg_schema):
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog, description=__doc__)
|
||||
parser.add_argument('--files', nargs='+', required=True, type=Path)
|
||||
inputs = parser.add_mutually_exclusive_group(required=True)
|
||||
inputs.add_argument('--files', nargs='+', type=Path)
|
||||
inputs.add_argument('--mm-program-report', type=Path,
|
||||
help='Read all frozen MM program observations directly; never generate an import queue')
|
||||
parser.add_argument('--pg-service', required=True)
|
||||
parser.add_argument('--pg-schema', required=True)
|
||||
parser.add_argument('--out', required=True, type=Path)
|
||||
args = parser.parse_args(argv)
|
||||
report = audit(args.files, args.out, pg_service=args.pg_service, pg_schema=args.pg_schema)
|
||||
report = audit(args.files, args.out, pg_service=args.pg_service, pg_schema=args.pg_schema,
|
||||
mm_program_report=args.mm_program_report)
|
||||
print(json.dumps({k: report[k] for k in ('status', 'records', 'kind_counts', 'approved_for_import')}, ensure_ascii=False))
|
||||
|
||||
@@ -20,16 +20,23 @@ COMMANDS = {
|
||||
('collect','public-openreview'): ('paper_atlas.collectors.public_openreview', {}, '核验其他会议的公开拒稿及撤稿标签'),
|
||||
('collect','icml-public-archive'): ('paper_atlas.collectors.icml_public_archive', {}, 'ICML 2026 公开拒稿原始投稿快照'),
|
||||
('collect','peerread'): ('paper_atlas.collectors.peerread', {}, 'PeerRead 历史决定补充'),
|
||||
('collect','reviewarena-metadata'): ('paper_atlas.collectors.reviewarena', {}, '仅投影固定版本评审镜像的索引字段,保留决定声明而不下载正文或评审页'),
|
||||
('collect','colm'): ('paper_atlas.collectors.proceedings', {'source':'colm'}, 'COLM 官方论文集'),
|
||||
('collect','neurips'): ('paper_atlas.collectors.proceedings', {'source':'neurips'}, 'NeurIPS 官方论文集'),
|
||||
('collect','mlsys'): ('paper_atlas.collectors.proceedings', {'source':'mlsys'}, 'MLSys 官方历年标题与摘要'),
|
||||
('collect','ismir2026-program'): ('paper_atlas.collectors.ismir2026', {}, '核验 ISMIR 2026 公开节目与录用名单并保留原始摘要、来源差异和自愿公开评审'),
|
||||
('collect','icaps2026-registry'): ('paper_atlas.collectors.icaps_registry2026', {}, '核验 ICAPS 2026 Crossref 注册元数据并保留独立版本'),
|
||||
('collect','eccv2026-program'): ('paper_atlas.collectors.eccv2026', {}, '核验ECCV2026暂定节目并断点采集公开元数据'),
|
||||
('collect','eccv2026-publications'): ('paper_atlas.collectors.eccv2026_publications', {}, '核验ECCV2026 Springer 正式出版章节'),
|
||||
('collect','icassp2026-registry'): ('paper_atlas.collectors.icassp2026', {}, '核验 ICASSP 2026 冻结注册论文元数据'),
|
||||
('collect','icdm2026-program'): ('paper_atlas.collectors.icdm2026', {}, '核验 ICDM2026 冻结日程及明确的研究/应用轨道'),
|
||||
('collect','cikm2026-sources'): ('paper_atlas.collectors.cikm2026', {}, '核对 CIKM 2026 节目、附属议程与 Crossref 出版登记'),
|
||||
('collect','sdm2026-proceedings'): ('paper_atlas.collectors.sdm2026', {}, '核验 SDM 2026 注册论文并单独保留公开日程版本'),
|
||||
('collect','icaps2026-program'): ('paper_atlas.collectors.icaps2026', {}, '核验 ICAPS 2026 官方录用名单并分开既有论文展示'),
|
||||
('collect','rss2026-proceedings'): ('paper_atlas.collectors.rss2026', {}, '核验 RSS 2026 官方论文集与独立录用表'),
|
||||
('collect','recsys2026-program'): ('paper_atlas.collectors.recsys2026', {}, '核验 RecSys 2026 公开录用贡献并保留 DOI 来源版本'),
|
||||
('collect','mm2026-program'): ('paper_atlas.collectors.mm2026', {}, '恢复 MM 2026 全部来源 track 与重复展示上下文,待贡献和身份核查'),
|
||||
('collect','mm2026-publications'): ('paper_atlas.collectors.mm_publications2026', {}, '核验 MM 2026 全部注册出版条目并分别保留节目分轨、摘要与版本差异'),
|
||||
('collect','cvf'): ('paper_atlas.collectors.vision', {'source':'cvf'}, 'CVPR / ICCV 官方开放目录'),
|
||||
('collect','ecva'): ('paper_atlas.collectors.vision', {'source':'ecva'}, 'ECCV 官方开放目录'),
|
||||
('collect','aamas-proceedings'): ('paper_atlas.collectors.aamas', {}, '按官方目录区分 AAMAS 发表形式和摘要'),
|
||||
@@ -42,6 +49,7 @@ COMMANDS = {
|
||||
('import','arxiv'): ('paper_atlas.ingestion.arxiv', {}, '导入 arXiv 元数据 ZIP'),
|
||||
('import','records'): ('paper_atlas.ingestion.records', {}, '导入核验过的会议文件'),
|
||||
('maintain','edition-tracks'): ('paper_atlas.ingestion.edition_tracks', {}, '按逐条来源证据修正既有投稿轨道并保护刷新'),
|
||||
('maintain','source-revisions'): ('paper_atlas.ingestion.source_revisions', {}, '按完整来源版本核验并修正出版分轨与覆盖数,保留历史并阻止旧版刷新'),
|
||||
('maintain','joint-editions'): ('paper_atlas.ingestion.joint_editions', {}, '按已核验计划修正既有联合会议届次归属'),
|
||||
('maintain','event-identity'): ('paper_atlas.ingestion.event_identity', {}, '按核验计划增加同年会议事件身份'),
|
||||
('maintain','merge'): ('paper_atlas.ingestion.maintenance', {}, '协调单写入合并与恢复'),
|
||||
|
||||
@@ -0,0 +1,164 @@
|
||||
"""Collect CIKM 2026 programme and Crossref metadata from frozen snapshots."""
|
||||
|
||||
import argparse
|
||||
from collections import Counter
|
||||
from hashlib import sha256
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
|
||||
from paper_atlas.io.html import Document, clean
|
||||
|
||||
PROGRAM = 'https://cikm2026.diag.uniroma1.it/cikm-program/program.html'
|
||||
VOLUME = '10.1145/3799682'
|
||||
CONTAINER = 'Proceedings of the 35th ACM International Conference on Information and Knowledge Management'
|
||||
TRACKS = {'FULL': 596, 'SHORT': 234, 'APPLIED': 86, 'RESOURCE': 46, 'DEMO': 46}
|
||||
SUPPLEMENTAL = {'KEYNOTE', 'TUTORIAL', 'INDUSTRY', 'POSTER', 'PHD', 'PHD · REMOTE', 'WORKSHOP', 'ANALYTICUP'}
|
||||
PAGE_COUNTS = (250, 250, 250, 250, 81)
|
||||
HASHES = {
|
||||
'program.html': '3eaa668b6aaf09a671d8a3e19a2889b9ab21afaa7ded013794afa94bf35281d6',
|
||||
'CIKM2026_program.html': 'e3630bd1247df713b8b618f1d3cd9709e7020b2524bb677f82a10f83badd1732',
|
||||
'crossref-parent.json': 'f3370418d86ccfcfaeb972073bc5fcd1c1c737caed96570fc83f99db0f7d204a',
|
||||
'crossref-page-1.json': '16bed29f8ee82d668b124c369b632c5c1de0753bdbec7eb72ecb22ffdda5c710',
|
||||
'crossref-page-2.json': '54424ca31d99462394c9fde5fecfd7dd16337351d21ede3169f19af54d4f49c5',
|
||||
'crossref-page-3.json': '2dc31d89f41966614ed66a11d1ebaa4b2a6904dcf92945b7185e72e7e65382dd',
|
||||
'crossref-page-4.json': 'fbbec2b52650dc99a0251c743f035a6fd95be263b04c55c12134cf8473bdde76',
|
||||
'crossref-page-5.json': '06151c5f770c1e37e8ea9dabe5210be9c969a14fa3bf9925a2d7730ced47d6e5',
|
||||
'crossref-pages.json': '8be1f8f05139f7a1231b73ad6430f28d2f625809f1530cd9de654f8b3c68c1ae',
|
||||
}
|
||||
|
||||
|
||||
def fields(node, css_class):
|
||||
return [child for child in node.walk() if css_class in child.attrs.get('class', '').split()]
|
||||
|
||||
|
||||
def parse_program(raw, *, require_doi):
|
||||
document = Document(raw.decode('utf-8'))
|
||||
papers = {}
|
||||
for card in fields(document.root, 'paper'):
|
||||
pid = card.attrs.get('data-pid')
|
||||
if not pid:
|
||||
continue
|
||||
titles = fields(card, 'ptitle')
|
||||
labels = fields(card, 'badge')
|
||||
abstracts = fields(card, 'abstxt')
|
||||
if pid in papers or len(titles) != 1 or len(labels) != 1 or len(abstracts) > 1:
|
||||
raise ValueError(f'Invalid or repeated programme card {pid}')
|
||||
title = clean(titles[0].text())
|
||||
label = clean(labels[0].text())
|
||||
authors = [clean(author.text()) for author in fields(card, 'au')]
|
||||
abstract = clean(abstracts[0].text()) if abstracts else None
|
||||
doi_url = titles[0].attrs.get('href')
|
||||
doi = doi_url.removeprefix('https://doi.org/') if doi_url else None
|
||||
if label not in TRACKS or not title or not authors or not all(authors):
|
||||
raise ValueError(f'Incomplete programme metadata {pid}')
|
||||
if require_doi and (not doi or not re.fullmatch(re.escape(VOLUME) + r'\.\d+', doi) or not abstract):
|
||||
raise ValueError(f'Missing programme DOI or abstract {pid}')
|
||||
papers[pid] = {'programme_id': pid, 'title': title, 'author_display_names': authors,
|
||||
'abstract': abstract, 'programme_label': label, 'programme_doi_link': doi_url,
|
||||
'doi': doi if require_doi else None}
|
||||
return papers
|
||||
|
||||
|
||||
def parse_supplemental(raw):
|
||||
document = Document(raw.decode('utf-8'))
|
||||
events = {}
|
||||
for session in fields(document.root, 'sessblock'):
|
||||
section = session.attrs.get('id', '')
|
||||
if not section.startswith(('pre-', 'kn-')):
|
||||
continue
|
||||
for card in fields(session, 'paper'):
|
||||
labels = fields(card, 'badge')
|
||||
if len(labels) != 1:
|
||||
raise ValueError(f'Ambiguous supplemental label in {section}')
|
||||
label = clean(labels[0].text())
|
||||
if label not in SUPPLEMENTAL:
|
||||
continue
|
||||
titles = fields(card, 'ptitle')
|
||||
abstracts = fields(card, 'abstxt')
|
||||
locator = card.attrs.get('id') or section
|
||||
if locator in events or len(titles) != 1 or len(abstracts) > 1:
|
||||
raise ValueError(f'Ambiguous supplemental entry {locator}')
|
||||
events[locator] = {'programme_locator': locator, 'programme_section': section,
|
||||
'programme_label': label, 'title': clean(titles[0].text()),
|
||||
'author_display_names': [clean(a.text()) for a in fields(card, 'au')],
|
||||
'abstract': clean(abstracts[0].text()) if abstracts else None}
|
||||
return events
|
||||
|
||||
|
||||
def parse_registry(snapshots):
|
||||
volume = json.loads(snapshots['crossref-parent.json'])['message']
|
||||
if (volume['DOI'], volume['title'], volume['ISBN']) != (VOLUME, [CONTAINER], ['9798400725395']):
|
||||
raise ValueError('Unexpected Crossref volume')
|
||||
records = {}
|
||||
for number, count in enumerate(PAGE_COUNTS, 1):
|
||||
name = f'crossref-page-{number}.json'
|
||||
message = json.loads(snapshots[name])['message']
|
||||
if message['total-results'] != 1081 or len(message['items']) != count:
|
||||
raise ValueError(f'Unexpected Crossref pagination in {name}')
|
||||
for item in message['items']:
|
||||
doi = item['DOI'].lower()
|
||||
if (doi in records or not re.fullmatch(re.escape(VOLUME) + r'\.\d+', doi)
|
||||
or item.get('container-title') != [CONTAINER] or item.get('type') != 'proceedings-article'
|
||||
or item.get('published', {}).get('date-parts') != [[2026, 11, 6]]):
|
||||
raise ValueError(f'Unexpected Crossref record {doi}')
|
||||
records[doi] = {'registry': item, 'source_file': name}
|
||||
return records
|
||||
|
||||
|
||||
def collect(snapshots, output):
|
||||
snapshots = Path(snapshots)
|
||||
output = Path(output)
|
||||
if output.exists() or output.is_symlink():
|
||||
raise FileExistsError(output)
|
||||
bodies = {}
|
||||
for name, expected in HASHES.items():
|
||||
path = snapshots / name
|
||||
if path.is_symlink() or not path.is_file() or path.stat().st_size > 8 * 1024**2:
|
||||
raise ValueError(f'Invalid snapshot file {name}')
|
||||
raw = path.read_bytes()
|
||||
if sha256(raw).hexdigest() != expected:
|
||||
raise ValueError(f'Snapshot hash mismatch for {name}')
|
||||
bodies[name] = raw
|
||||
|
||||
programme = parse_program(bodies['program.html'], require_doi=True)
|
||||
alternate = parse_program(bodies['CIKM2026_program.html'], require_doi=False)
|
||||
supplemental = parse_supplemental(bodies['program.html'])
|
||||
registry = parse_registry(bodies)
|
||||
programme_dois = {r['doi'] for r in programme.values()}
|
||||
tracks = Counter(r['programme_label'] for r in programme.values())
|
||||
if (tracks != TRACKS or len(programme_dois) != len(programme) or programme_dois - registry.keys()
|
||||
or len(alternate) != 1005 or alternate.keys() - programme.keys()
|
||||
or len(supplemental) != 73 or len(registry) - len(programme_dois) != 73):
|
||||
raise ValueError('Unexpected programme or registry coverage')
|
||||
groups = {
|
||||
'programme': programme,
|
||||
'alternate-programme': alternate,
|
||||
'supplemental-programme': supplemental,
|
||||
'crossref': registry,
|
||||
}
|
||||
output.mkdir(parents=True, exist_ok=False)
|
||||
files = {}
|
||||
for name, rows in groups.items():
|
||||
filename = name + '.jsonl'
|
||||
payload = ''.join(json.dumps({'source_key': key, **row}, ensure_ascii=False) + '\n'
|
||||
for key, row in sorted(rows.items())).encode()
|
||||
(output / filename).write_bytes(payload)
|
||||
files[filename] = {'records': len(rows), 'sha256': sha256(payload).hexdigest()}
|
||||
report = {'status': 'collected_pending_identity_review', 'files': files, 'input_sha256': HASHES,
|
||||
'programme_by_label': dict(tracks), 'programme_dois_in_registry': len(programme_dois),
|
||||
'registry_outside_doi_programme': sorted(registry.keys() - programme_dois),
|
||||
'programme_ids_missing_from_alternate': sorted(programme.keys() - alternate.keys()),
|
||||
'database_written': False, 'public_negative_decisions_verified': False}
|
||||
partial = output / 'collection-report.json.part'
|
||||
partial.write_text(json.dumps(report, ensure_ascii=False, indent=2) + '\n', encoding='utf-8')
|
||||
partial.replace(output / 'collection-report.json')
|
||||
return report
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog, description=__doc__)
|
||||
parser.add_argument('--snapshots', type=Path, required=True)
|
||||
parser.add_argument('--out', type=Path, required=True)
|
||||
options = parser.parse_args(argv)
|
||||
print(json.dumps(collect(options.snapshots, options.out), ensure_ascii=False))
|
||||
@@ -0,0 +1,184 @@
|
||||
"""Normalize frozen ECCV 2026 Springer proceedings and preserve publisher evidence."""
|
||||
import argparse
|
||||
import fcntl
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
import tempfile
|
||||
|
||||
from paper_atlas.io.html import Document, clean
|
||||
from paper_atlas.io.springer import parse_chapter, read_page, snapshot_index
|
||||
|
||||
SOURCE = 'eccv2026_springer'
|
||||
ORIGIN = 'https://link.springer.com'
|
||||
BOOK_TITLE = 'Computer Vision – ECCV 2026'
|
||||
LIST_URL = ORIGIN + '/conference/eccv'
|
||||
BOOK = re.compile(r'10\.1007/978-3-032-\d{5}-[\dXx]')
|
||||
CHAPTER = re.compile(BOOK.pattern + r'_[1-9]\d*')
|
||||
VOLUMES = set(range(17001, 17079))
|
||||
PUBLICATIONS = 2835
|
||||
|
||||
|
||||
def fingerprint(raw):
|
||||
return hashlib.sha256(raw).hexdigest()
|
||||
|
||||
|
||||
def toc_page(html, book_doi, volume):
|
||||
tree = Document(html).root
|
||||
text = clean(tree.text())
|
||||
if BOOK_TITLE not in text or 'Malmö' not in text or f'(LNCS, volume {volume})' not in text:
|
||||
raise ValueError('Unexpected Springer ECCV volume scope')
|
||||
entries = {}
|
||||
for node in tree.walk():
|
||||
if node.tag != 'li' or node.attrs.get('data-test') != 'chapter':
|
||||
continue
|
||||
headings = [n for n in node.walk() if n.tag == 'h3']
|
||||
if len(headings) != 1:
|
||||
raise ValueError('Ambiguous Springer contents heading')
|
||||
heading, = headings
|
||||
title = clean(heading.text())
|
||||
links = [n.attrs.get('href', '') for n in heading.walk() if n.tag == 'a']
|
||||
if title in ('Front Matter', 'Back Matter') and not links:
|
||||
continue
|
||||
if len(links) != 1 or not title:
|
||||
raise ValueError('Authored contents entry lacks one chapter link')
|
||||
link, = links
|
||||
doi = link.removeprefix('/chapter/')
|
||||
if (not link.startswith('/chapter/') or not CHAPTER.fullmatch(doi)
|
||||
or not doi.startswith(book_doi + '_') or doi in entries):
|
||||
raise ValueError('Unexpected or duplicate Springer chapter DOI')
|
||||
entries[doi] = title
|
||||
if not entries:
|
||||
raise ValueError('Empty Springer contents page')
|
||||
return entries
|
||||
|
||||
|
||||
def publisher_inventory(root):
|
||||
root = Path(root).resolve()
|
||||
receipt_bytes = (root / 'receipts.jsonl').read_bytes()
|
||||
inventory_bytes = (root / 'volume-checkpoint.json').read_bytes()
|
||||
summary = json.loads((root / 'summary.json').read_bytes())
|
||||
volumes = json.loads(inventory_bytes)
|
||||
receipt_rows = [json.loads(line) for line in receipt_bytes.splitlines()]
|
||||
receipts = {row['file']: row for row in receipt_rows}
|
||||
if (summary['raw_receipts_sha256'] != fingerprint(receipt_bytes)
|
||||
or summary['volume_checkpoint_sha256'] != fingerprint(inventory_bytes)
|
||||
or set(volumes) != {str(n) for n in VOLUMES}
|
||||
or len(receipts) != len(receipt_rows)):
|
||||
raise ValueError('Incomplete or inconsistent Springer contents snapshot')
|
||||
entries, used = {}, set()
|
||||
for number, volume in sorted(volumes.items()):
|
||||
book_url = volume['book']
|
||||
book_doi = book_url.removeprefix(ORIGIN + '/book/')
|
||||
if not BOOK.fullmatch(book_doi) or book_url != ORIGIN + '/book/' + book_doi:
|
||||
raise ValueError('Unexpected Springer volume URL')
|
||||
if set(volume['files']) != {'1', '2'}:
|
||||
raise ValueError('Missing reviewed Springer contents page')
|
||||
volume_dois = set()
|
||||
for index, file in volume['files'].items():
|
||||
receipt = receipts[file]
|
||||
path = root / file
|
||||
page_url = book_url + ('?page=2' if index == '2' else '')
|
||||
if (file in used or Path(file).name != file or not path.resolve().is_relative_to(root)
|
||||
or receipt['url'] != page_url or receipt['status'] != 200
|
||||
or not 0 < receipt['bytes'] <= 3_000_000):
|
||||
raise ValueError('Unsafe, reused or mismatched Springer contents receipt')
|
||||
raw = path.read_bytes()
|
||||
if len(raw) != receipt['bytes'] or fingerprint(raw) != receipt['sha256']:
|
||||
raise ValueError('Springer contents bytes disagree with receipt')
|
||||
used.add(file)
|
||||
page_entries = toc_page(raw.decode(), book_doi, int(number))
|
||||
if entries.keys() & page_entries.keys():
|
||||
raise ValueError('Springer chapter repeats across contents pages')
|
||||
volume_dois.update(page_entries)
|
||||
entries.update({doi: {'title': title, 'lncs_volume': int(number), 'book_doi': book_doi,
|
||||
'evidence_url': page_url, 'evidence_sha256': receipt['sha256']}
|
||||
for doi, title in page_entries.items()})
|
||||
if volume_dois != set(volume['chapter_dois']) or len(volume_dois) != len(volume['chapter_dois']):
|
||||
raise ValueError('Springer volume chapter set differs from checkpoint')
|
||||
if used != receipts.keys() or len(entries) != PUBLICATIONS or summary['chapter_link_count'] != PUBLICATIONS:
|
||||
raise ValueError('Springer contents chapter total differs from the reviewed archive')
|
||||
return entries
|
||||
|
||||
|
||||
def publisher_row(chapter, toc, evidence):
|
||||
doi = chapter['doi']
|
||||
if clean(chapter['title']) != toc['title']:
|
||||
raise ValueError('Chapter title differs from Springer contents, review the version')
|
||||
url = ORIGIN + '/chapter/' + doi
|
||||
return {
|
||||
'source': SOURCE, 'external_id': doi, 'url': url,
|
||||
'venue': 'ECCV', 'year': 2026, 'track': 'main',
|
||||
'title': chapter['title'], 'authors': chapter['authors'], 'abstract': chapter['abstract'],
|
||||
'status': 'accepted', 'identifiers': {'doi': doi},
|
||||
'submission_identifiers': [{'scheme': 'publication_doi', 'value': doi}],
|
||||
'evidence_url': url, 'observed_at': evidence['observed_at'],
|
||||
'raw_decision': 'Chapter published in the refereed ECCV 2026 conference proceedings.',
|
||||
'tags': [{'namespace': SOURCE + ':publication_format', 'value': 'proceedings_chapter'}],
|
||||
'metadata': {
|
||||
'book_title': BOOK_TITLE, 'book_doi': toc['book_doi'], 'lncs_volume': toc['lncs_volume'],
|
||||
'authors_citation_raw': chapter['authors_citation_raw'], 'first_page': chapter['first_page'],
|
||||
'last_page': chapter['last_page'], 'publisher_chapter_evidence': evidence,
|
||||
'publisher_contents_evidence': {'url': toc['evidence_url'], 'sha256': toc['evidence_sha256']},
|
||||
'scope': 'Refereed conference proceedings, LNCS 17001–17078.',
|
||||
'original_decision_signature_verified': False,
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
def collect(out, *, toc_snapshots, chapter_snapshots):
|
||||
contents = publisher_inventory(toc_snapshots)
|
||||
chapters = snapshot_index(chapter_snapshots)
|
||||
urls = {ORIGIN + '/chapter/' + doi: doi for doi in contents}
|
||||
if not chapters or not chapters.keys() <= urls.keys():
|
||||
raise ValueError('No chapter snapshots or chapter outside reviewed ECCV proceedings')
|
||||
rows, raw = [], {}
|
||||
for url in sorted(chapters):
|
||||
doi = urls[url]
|
||||
body, evidence = read_page(chapters, url)
|
||||
chapter = parse_chapter(body.decode(), doi, book_title=BOOK_TITLE, publication_year=2026)
|
||||
rows.append(publisher_row(chapter, contents[doi], evidence))
|
||||
raw[evidence['file']] = body
|
||||
out = Path(out)
|
||||
out.mkdir(parents=True, exist_ok=True)
|
||||
lock_path = out / 'eccv2026-publications.collect.lock'
|
||||
filename = 'eccv2026-springer-main.jsonl'
|
||||
with lock_path.open('a') as lock:
|
||||
fcntl.flock(lock, fcntl.LOCK_EX | fcntl.LOCK_NB)
|
||||
if any(p != lock_path for p in out.iterdir()):
|
||||
raise FileExistsError('Existing output must be inspected, not overwritten')
|
||||
with tempfile.TemporaryDirectory(prefix='.eccv2026-publications-', dir=out) as temporary:
|
||||
stage = Path(temporary)
|
||||
(stage / 'evidence').mkdir()
|
||||
for path, body in raw.items():
|
||||
(stage / path).write_bytes(body)
|
||||
data = ''.join(json.dumps(row, ensure_ascii=False) + '\n' for row in rows).encode()
|
||||
(stage / filename).write_bytes(data)
|
||||
manifest = {
|
||||
'source': SOURCE, 'venue': 'ECCV', 'year': 2026, 'track': 'main', 'parser_version': 1,
|
||||
'file': str(out / filename), 'file_sha256': fingerprint(data), 'count': len(rows),
|
||||
'list_url': LIST_URL, 'source_expected_count': PUBLICATIONS, 'status': 'partial',
|
||||
'missing_abstract_count': sum(not r['abstract'] for r in rows),
|
||||
}
|
||||
report = {
|
||||
'status': 'collected_pending_identity_review', 'records': len(rows), 'files': [manifest],
|
||||
'toc_chapters': len(contents), 'chapter_details_missing': len(contents) - len(rows),
|
||||
'database_written': False,
|
||||
}
|
||||
metadata = Path(filename).with_suffix('.manifest.json')
|
||||
(stage / metadata).write_text(json.dumps(manifest, indent=2) + '\n')
|
||||
(stage / 'collection-report.json').write_text(json.dumps(report, indent=2) + '\n')
|
||||
for item in ('evidence', filename, metadata, 'collection-report.json'):
|
||||
(stage / item).replace(out / item)
|
||||
return report
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog)
|
||||
parser.add_argument('--toc-snapshots', required=True)
|
||||
parser.add_argument('--chapter-snapshots', required=True)
|
||||
parser.add_argument('--out', required=True)
|
||||
args = parser.parse_args(argv)
|
||||
print(json.dumps(collect(args.out, toc_snapshots=args.toc_snapshots,
|
||||
chapter_snapshots=args.chapter_snapshots), ensure_ascii=False))
|
||||
@@ -0,0 +1,172 @@
|
||||
"""Normalize the frozen ICASSP 2026 Crossref census without network or database writes."""
|
||||
import argparse
|
||||
from datetime import datetime
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
|
||||
from paper_atlas.io.html import clean
|
||||
|
||||
SOURCE = 'crossref_icassp_registry'
|
||||
TRACK = 'proceedings_unclassified'
|
||||
SNAPSHOT = 'icassp2026-registry-capture-v1'
|
||||
MANIFEST_SHA = '1f8ffa040b1f682a4b65f476f4115cb95629056d8de5c324c8417d823846ae93'
|
||||
CONTAINER = 'ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)'
|
||||
OTHER_CONTAINER = '2026 IEEE 6th International Conference in Power Engineering Applications (ICPEA)'
|
||||
PREFIX = '10.1109/icassp55912.2026.'
|
||||
NONPAPERS = {
|
||||
PREFIX+'11463724': ('ICASSP 2026 Reviewers', 'xii-lxxxvi'),
|
||||
PREFIX+'11464828': ('Author Index', 'A-1-A-152'),
|
||||
PREFIX+'11464776': ('Title Page', 'i-ii'),
|
||||
PREFIX+'11462944': ('ICASSP 2026 Committees', 'vi-xii'),
|
||||
PREFIX+'11464852': ('ICASSP 2026 Organizing Committee', 'iii-iii'),
|
||||
PREFIX+'11460797': ('ICASSP 2026 Cover Page', 'c1-c1'),
|
||||
PREFIX+'11463523': ('Table of Contents', 'xlix-cdlxii'),
|
||||
}
|
||||
|
||||
|
||||
def digest(raw):
|
||||
return hashlib.sha256(raw).hexdigest()
|
||||
|
||||
|
||||
def load_snapshot(root):
|
||||
root = Path(root); directory = root/SNAPSHOT; path = directory/'manifest.json'
|
||||
if root.is_symlink() or directory.is_symlink() or path.is_symlink() or path.stat().st_size > 128*1024:
|
||||
raise ValueError('Unsafe snapshot manifest')
|
||||
raw = path.read_bytes()
|
||||
if digest(raw) != MANIFEST_SHA:
|
||||
raise ValueError('Reviewed manifest hash changed')
|
||||
manifest = json.loads(raw)
|
||||
if (manifest['status'] != 'registry_census_completed_not_import_approval'
|
||||
or manifest['records_captured'] != 4645 or manifest['unique_dois'] != 4645
|
||||
or manifest['database_written'] is not False or manifest['approved_for_import'] != 0
|
||||
or len(manifest['pages']) != 10):
|
||||
raise ValueError('Incomplete frozen census receipt')
|
||||
files = {'manifest.json': raw}; rows = []
|
||||
for index, receipt in enumerate(manifest['pages'], 1):
|
||||
name = receipt['file']; path = directory/name
|
||||
if (name != f'page-{index:03}.json' or path.is_symlink()
|
||||
or path.stat().st_size > 4*1024**2):
|
||||
raise ValueError('Unsafe snapshot page')
|
||||
body = path.read_bytes()
|
||||
if (receipt['http_status'] != 200 or receipt['url'] != receipt['final_url']
|
||||
or not receipt['url'].startswith('https://api.crossref.org/works?')
|
||||
or datetime.fromisoformat(receipt['at']).utcoffset() is None
|
||||
or receipt['bytes'] != len(body) or receipt['sha256'] != digest(body)):
|
||||
raise ValueError('Snapshot receipt/body disagree')
|
||||
data = json.loads(body); message = data['message']
|
||||
expected = 500 if index < 10 else 145
|
||||
if data['status'] != 'ok' or message['total-results'] != 4645 or len(message['items']) != expected:
|
||||
raise ValueError('Incomplete registry page')
|
||||
files[name] = body
|
||||
for original in message['items']:
|
||||
rows.append((original, receipt))
|
||||
return manifest, files, rows
|
||||
|
||||
|
||||
def registry_papers(rows):
|
||||
seen = set(); papers = []; nonpapers = []; other = []
|
||||
for original, receipt in rows:
|
||||
doi = original.get('DOI')
|
||||
if not isinstance(doi, str) or doi in seen:
|
||||
raise ValueError('Missing or repeated registry DOI')
|
||||
seen.add(doi)
|
||||
if original.get('container-title') == [OTHER_CONTAINER]:
|
||||
if doi.startswith(PREFIX):
|
||||
raise ValueError('Conference DOI/container conflict')
|
||||
other.append(doi)
|
||||
continue
|
||||
if (original.get('container-title') != [CONTAINER]
|
||||
or original.get('event', {}).get('name') != CONTAINER
|
||||
or not re.fullmatch(re.escape(PREFIX)+r'[0-9]+', doi)
|
||||
or original.get('publisher') != 'IEEE' or original.get('type') != 'proceedings-article'
|
||||
or original.get('URL') != 'https://doi.org/'+doi
|
||||
or original.get('published') != {'date-parts': [[2026, 5, 3]]}
|
||||
or original.get('event', {}).get('start') != {'date-parts': [[2026, 5, 3]]}
|
||||
or original.get('event', {}).get('end') != {'date-parts': [[2026, 5, 8]]}):
|
||||
raise ValueError('Wrong registered conference, DOI, publisher or dates')
|
||||
if doi in NONPAPERS:
|
||||
title, page = NONPAPERS[doi]
|
||||
if original.get('title') != [title] or original.get('page') != page or original.get('author'):
|
||||
raise ValueError('Changed reviewed nonpaper evidence')
|
||||
nonpapers.append({'doi': doi, 'title': title, 'page': page})
|
||||
continue
|
||||
titles = original.get('title')
|
||||
if not isinstance(titles, list) or len(titles) != 1 or not isinstance(titles[0], str) or not clean(titles[0]):
|
||||
raise ValueError('Missing registry title')
|
||||
authors = original.get('author')
|
||||
if not isinstance(authors, list) or not authors or original.get('abstract'):
|
||||
raise ValueError('Missing authors or changed abstract availability')
|
||||
names = []; flags = []
|
||||
for position, author in enumerate(authors):
|
||||
if (not isinstance(author, dict) or not isinstance(author.get('family'), str)
|
||||
or not clean(author['family']) or ('given' in author and not isinstance(author['given'], str))
|
||||
or author.get('sequence') != ('first' if position == 0 else 'additional')):
|
||||
raise ValueError('Incomplete registry author or sequence')
|
||||
name = clean(' '.join(x for x in (author.get('given'), author['family']) if x))
|
||||
if not name or '@' in name or name.casefold() == 'anonymous':
|
||||
raise ValueError('Unreviewed author identity')
|
||||
if not author.get('given'):
|
||||
flags.append({'kind': 'registry_single_name_field_not_split', 'author_position': position+1})
|
||||
names.append(name)
|
||||
papers.append({'original': original, 'receipt': receipt, 'title': clean(titles[0]), 'authors': names, 'review_flags': flags})
|
||||
if len(seen) != 4645 or len(papers) != 4582 or len(other) != 56 or {r['doi'] for r in nonpapers} != set(NONPAPERS):
|
||||
raise ValueError('Incomplete reviewed registry partition')
|
||||
return sorted(papers, key=lambda p: p['original']['DOI']), nonpapers, other
|
||||
|
||||
|
||||
def collect(out, *, snapshots):
|
||||
out = Path(out)
|
||||
if out.exists():
|
||||
raise ValueError('Output exists; inspect previous operation before retrying')
|
||||
out.mkdir(parents=True)
|
||||
try:
|
||||
manifest, files, items = load_snapshot(snapshots)
|
||||
papers, nonpapers, other = registry_papers(items)
|
||||
evidence_dir = out/'evidence'; evidence_dir.mkdir()
|
||||
for name, body in files.items():
|
||||
(evidence_dir/name).write_bytes(body)
|
||||
rows = []
|
||||
for paper in papers:
|
||||
original = paper['original']; receipt = paper['receipt']; doi = original['DOI']
|
||||
rows.append({'source': SOURCE, 'external_id': doi, 'url': original['URL'],
|
||||
'title': paper['title'], 'authors': paper['authors'], 'abstract': None,
|
||||
'identifiers': {'doi': doi}, 'submission_identifiers': [{'scheme': 'publication_doi', 'value': doi}],
|
||||
'venue': 'ICASSP', 'year': 2026, 'track': TRACK, 'status': 'accepted',
|
||||
'raw_decision': 'Publication registered in Crossref ICASSP 2026 proceedings; not an original signed decision.',
|
||||
'evidence_url': receipt['url'], 'observed_at': receipt['at'], 'tags': [],
|
||||
'metadata': {'registry_fields': original, 'review_flags': paper['review_flags'],
|
||||
'source_evidence': {'file': 'evidence/'+receipt['file'], 'sha256': receipt['sha256'], 'url': receipt['url']},
|
||||
'source_manifest': {'file': 'evidence/manifest.json', 'sha256': MANIFEST_SHA},
|
||||
'original_decision_signature_verified': False,
|
||||
'abstract_availability': 'Absent in frozen registry deposit; no abstract inferred or fused from program versions.',
|
||||
'author_transform': 'Whitespace normalization of given + family in registered order; single fields retained, not guessed or split; all original author fields retained.',
|
||||
'publication_dates': {k: original[k] for k in ('published', 'published-print', 'published-online') if k in original},
|
||||
'date_scope': 'Registered event starts May 3; conference website says May 4–8. Source values preserved without substitution.',
|
||||
'scope': 'Registered publication version only. Requires work/DOI/history review before import. No automatic program match, aliases, track/type or public-negative inference.'}})
|
||||
path = out/'icassp-crossref-registry-2026-unclassified.jsonl'
|
||||
raw = ''.join(json.dumps(r, ensure_ascii=False)+'\n' for r in rows).encode(); path.write_bytes(raw)
|
||||
file_manifest = {'source': SOURCE, 'parser_version': 1, 'venue': 'ICASSP', 'year': 2026, 'track': TRACK,
|
||||
'count': len(rows), 'file': str(path), 'file_sha256': digest(raw), 'list_url': manifest['pages'][0]['url'],
|
||||
'status': 'partial', 'missing_abstract_count': len(rows)}
|
||||
path.with_suffix('.manifest.json').write_text(json.dumps(file_manifest, indent=2)+'\n')
|
||||
report = {'status': 'collected_pending_identity_review', 'records': len(rows), 'files': [file_manifest],
|
||||
'filtered_registry_records': 4645, 'excluded_nonpaper_items': nonpapers, 'excluded_other_conference_dois': other,
|
||||
'missing_abstract_count': len(rows), 'single_name_author_fields': sum(len(p['review_flags']) for p in papers),
|
||||
'database_written': False, 'network_requests': 0, 'pdfs_downloaded': False, 'approved_for_import': 0,
|
||||
'public_rejected_availability': 'unknown', 'source_manifest_sha256': MANIFEST_SHA,
|
||||
'scope': 'Registry census, not completeness of program, abstracts, tracks, negative decisions or work identities. Official program has4589 titles;174 unmatched normalized title candidates remain separate source discrepancies, not rejection or withdrawal evidence.'}
|
||||
part = out/'collection-report.json.part'; part.write_text(json.dumps(report, ensure_ascii=False, indent=2)+'\n'); part.replace(out/'collection-report.json')
|
||||
return report
|
||||
except Exception as error:
|
||||
(out/'failed.json').write_text(json.dumps({'status': 'failed_not_complete', 'error_type': type(error).__name__, 'database_written': False})+'\n')
|
||||
raise
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog, description=__doc__)
|
||||
parser.add_argument('--snapshots', required=True)
|
||||
parser.add_argument('--out', required=True)
|
||||
args = parser.parse_args(argv)
|
||||
print(json.dumps(collect(args.out, snapshots=args.snapshots), ensure_ascii=False))
|
||||
@@ -0,0 +1,208 @@
|
||||
"""Frozen ISMIR 2026 accepted program, abstracts and opt-in review provenance."""
|
||||
import argparse
|
||||
from collections import Counter
|
||||
from datetime import datetime
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
import unicodedata
|
||||
|
||||
from paper_atlas.io.html import Document, Node, clean
|
||||
|
||||
SOURCE = 'ismir_program'
|
||||
TRACK = 'program_unclassified'
|
||||
PAPERS = 'https://ismir2026program.ismir.net/papers.html'
|
||||
ACCEPTED = 'https://ismir2026.ismir.net/accepted-papers'
|
||||
SNAPSHOTS = {
|
||||
'remaining2026-listings-research-v2': ('fea7b710587636db06794ee195cd083634c41e9054a217859d9dd0ccdd1afab9',
|
||||
'research_completed_not_coverage_or_import_approval', {
|
||||
'ismir-accepted.html': ACCEPTED,
|
||||
'ismir-call.html': 'https://ismir2026.ismir.net/authors/call-for-papers'}),
|
||||
'ismir2026-script-research-v1': ('5885b4bcfc2fc65374e5dc7a0d74f4eb86a0a36b8689f9fe8edbd10d49a08794',
|
||||
'public_script_research_only', {'papers.js': 'https://ismir2026program.ismir.net/static/js/papers.js'}),
|
||||
'ismir2026-json-research-v1': ('8be3e23277c199720264d940917118572655d18c0f9ac6db1b9cd30a6561b34d',
|
||||
'public_metadata_json_research_only', {'papers.json': 'https://ismir2026program.ismir.net/papers.json'}),
|
||||
}
|
||||
REVIEW_KEYS = {'meta_review','publish_reviews','review1','review2','review3','review4'}
|
||||
|
||||
|
||||
def digest(raw):
|
||||
return hashlib.sha256(raw).hexdigest()
|
||||
|
||||
|
||||
def one(items):
|
||||
if len(items) != 1:
|
||||
raise ValueError('Expected exactly one reviewed source field')
|
||||
return items[0]
|
||||
|
||||
|
||||
def title_key(text):
|
||||
return ''.join(c for c in unicodedata.normalize('NFKD', text).casefold() if c.isalnum())
|
||||
|
||||
|
||||
def load_snapshots(root):
|
||||
root = Path(root)
|
||||
if root.is_symlink():
|
||||
raise ValueError('Unsafe snapshot root')
|
||||
result = {}
|
||||
for name, (expected, status, urls) in SNAPSHOTS.items():
|
||||
directory = root/name; path = directory/'manifest.json'
|
||||
if directory.is_symlink() or path.is_symlink() or path.stat().st_size > 128*1024:
|
||||
raise ValueError('Unsafe snapshot manifest')
|
||||
raw = path.read_bytes()
|
||||
if digest(raw) != expected:
|
||||
raise ValueError('Reviewed manifest hash changed')
|
||||
manifest = json.loads(raw)
|
||||
if manifest['status'] != status or manifest['database_written'] is not False:
|
||||
raise ValueError('Unreviewed snapshot scope')
|
||||
files = {'manifest.json': raw}; receipts = {}
|
||||
for filename, url in urls.items():
|
||||
receipt = one([r for r in manifest['entries'] if r.get('file') == filename])
|
||||
path = directory/filename
|
||||
if path.is_symlink() or path.stat().st_size > 8*1024**2:
|
||||
raise ValueError('Unsafe snapshot source file')
|
||||
body = path.read_bytes()
|
||||
if (receipt['url'] != url or receipt['final_url'] != url or receipt['http_status'] != 200
|
||||
or receipt['bytes'] != len(body) or receipt['sha256'] != digest(body)
|
||||
or datetime.fromisoformat(receipt['at']).utcoffset() is None):
|
||||
raise ValueError('Source receipt/body disagree')
|
||||
files[filename] = body; receipts[filename] = receipt
|
||||
result[name] = (files, receipts)
|
||||
return result
|
||||
|
||||
|
||||
def author_names(literal):
|
||||
# Institution names can contain semicolons; split only outside parentheses.
|
||||
parts = []; depth = 0; start = 0
|
||||
for index, char in enumerate(literal):
|
||||
if char == '(': depth += 1
|
||||
elif char == ')':
|
||||
depth -= 1
|
||||
if depth < 0: raise ValueError('Unbalanced author affiliation')
|
||||
elif char == ';' and depth == 0:
|
||||
parts.append(literal[start:index]); start = index+1
|
||||
if depth:
|
||||
raise ValueError('Unbalanced author affiliation')
|
||||
parts.append(literal[start:])
|
||||
names = [clean(p.split('(', 1)[0]) for p in parts]
|
||||
if any(not name for name in names):
|
||||
raise ValueError('Missing accepted-list author')
|
||||
return names
|
||||
|
||||
|
||||
def accepted_listing(raw):
|
||||
doc = Document(raw.decode('utf-8')); tables = []
|
||||
for table in doc.root.walk():
|
||||
if table.tag != 'table': continue
|
||||
rows = []
|
||||
for row in table.walk():
|
||||
if row.tag == 'tr':
|
||||
rows.append([clean(n.text()) for n in row.children if isinstance(n, Node) and n.tag in ('th','td')])
|
||||
if rows and rows[0] == ['Paper Title', 'Authors']:
|
||||
tables.append(rows)
|
||||
rows = one(tables)[1:]; result = {}
|
||||
if len(rows) != 140:
|
||||
raise ValueError('Incomplete official accepted-list census')
|
||||
for row in rows:
|
||||
if len(row) != 2 or not row[0] or not row[1] or title_key(row[0]) in result:
|
||||
raise ValueError('Missing or ambiguous accepted-list title')
|
||||
result[title_key(row[0])] = {'title': row[0], 'authors_literal': row[1], 'author_names': author_names(row[1])}
|
||||
return result
|
||||
|
||||
|
||||
def program_observations(rows, listing):
|
||||
if not isinstance(rows, list) or len(rows) != 140:
|
||||
raise ValueError('Incomplete public program census')
|
||||
observations = []; seen = set(); used = set()
|
||||
for original in rows:
|
||||
if not isinstance(original, dict) or not isinstance(original.get('content'), dict):
|
||||
raise ValueError('Invalid public program record')
|
||||
pid = original.get('id'); content = original.get('content', {})
|
||||
if not isinstance(pid, str) or not re.fullmatch(r'[1-9][0-9]*', pid) or pid in seen or original.get('forum') != pid:
|
||||
raise ValueError('Missing/repeated numeric program locator, not an OpenReview ID')
|
||||
seen.add(pid); title = content.get('title'); authors = content.get('authors'); abstract = content.get('abstract')
|
||||
if (not isinstance(title, str) or not clean(title) or not isinstance(abstract, str) or not abstract.strip()
|
||||
or not isinstance(authors, list) or not authors or any(not isinstance(a, str) or not clean(a) for a in authors)):
|
||||
raise ValueError('Missing program title, author sequence or abstract')
|
||||
key = title_key(title)
|
||||
if key not in listing or key in used:
|
||||
raise ValueError('Program/accepted-list title membership mismatch')
|
||||
used.add(key); accepted = listing[key]
|
||||
keywords = content.get('keywords'); presentation = content.get('paper_presentation')
|
||||
session = content.get('session'); reviews = original.get('reviews')
|
||||
if (not isinstance(keywords, list) or any(not isinstance(k, str) for k in keywords)
|
||||
or presentation not in ('Remote','In-person','Remote during oral, In-person during nominee')
|
||||
or not isinstance(session, list) or len(session) != 1 or not isinstance(session[0], str) or session[0] != original.get('session')
|
||||
or not re.fullmatch(r'(?:[1-9]|1[0-4])', session[0])):
|
||||
raise ValueError('Unreviewed source keywords, presentation or session')
|
||||
if (not isinstance(reviews, dict) or set(reviews) != REVIEW_KEYS
|
||||
or any(not isinstance(v, str) for v in reviews.values())
|
||||
or reviews['publish_reviews'] not in ('Yes','No')):
|
||||
raise ValueError('Unreviewed opt-in review structure')
|
||||
review_text = [reviews[k] for k in REVIEW_KEYS-{'publish_reviews'}]
|
||||
if (reviews['publish_reviews'] == 'No' and any(review_text)) or (reviews['publish_reviews'] == 'Yes' and not any(review_text)):
|
||||
raise ValueError('Review publication preference/body disagree')
|
||||
flags = []
|
||||
if title != accepted['title']: flags.append('program_and_accepted_title_literal_versions_differ')
|
||||
if authors != accepted['author_names']: flags.append('program_and_accepted_author_sequence_versions_differ')
|
||||
observations.append({'original': original, 'accepted_listing': accepted, 'review_flags': flags})
|
||||
if (used != listing.keys() or sum('program_and_accepted_title_literal_versions_differ' in o['review_flags'] for o in observations) != 2
|
||||
or {o['original']['id'] for o in observations if 'program_and_accepted_author_sequence_versions_differ' in o['review_flags']} != {'60','126','433'}
|
||||
or Counter(o['original']['reviews']['publish_reviews'] for o in observations) != {'Yes':101,'No':39}):
|
||||
raise ValueError('Unreviewed source-version or opt-in census')
|
||||
return observations
|
||||
|
||||
|
||||
def collect(out, *, snapshots):
|
||||
out = Path(out)
|
||||
if out.exists(): raise ValueError('Output exists; inspect previous operation before retrying')
|
||||
out.mkdir(parents=True)
|
||||
try:
|
||||
sources = load_snapshots(snapshots)
|
||||
official, _ = sources['remaining2026-listings-research-v2']; script, _ = sources['ismir2026-script-research-v1']
|
||||
program, receipts = sources['ismir2026-json-research-v1']; receipt = receipts['papers.json']
|
||||
if b"d3.json('papers.json')" not in script['papers.js'] or b'href="poster_${openreview.id}.html"' not in script['papers.js']:
|
||||
raise ValueError('Unreviewed public program data path or poster link template')
|
||||
observations = program_observations(json.loads(program['papers.json']), accepted_listing(official['ismir-accepted.html']))
|
||||
evidence = out/'evidence'; evidence.mkdir()
|
||||
for name, (files, _) in sources.items():
|
||||
directory = evidence/name; directory.mkdir()
|
||||
for filename, body in files.items(): (directory/filename).write_bytes(body)
|
||||
rows = []
|
||||
for observation in observations:
|
||||
original = observation['original']; content = original['content']; pid = original['id']
|
||||
tags = [{'namespace':SOURCE+':presentation','value':content['paper_presentation']}]
|
||||
tags += [{'namespace':SOURCE+':keyword','value':value} for value in dict.fromkeys(clean(k) for k in content['keywords'] if clean(k))]
|
||||
rows.append({'source':SOURCE,'external_id':'2026/'+pid,'url':'https://ismir2026program.ismir.net/poster_'+pid+'.html',
|
||||
'title':content['title'],'authors':content['authors'],'abstract':content['abstract'],
|
||||
'identifiers':{},'submission_identifiers':[],'venue':'ISMIR','year':2026,'track':TRACK,'status':'accepted',
|
||||
'raw_decision':'Present in official ISMIR2026 accepted list and public virtual program; not an original signed decision.',
|
||||
'evidence_url':ACCEPTED,'observed_at':receipt['at'],'tags':tags,
|
||||
'metadata':{'program_entry':original,'accepted_listing':observation['accepted_listing'],'review_flags':observation['review_flags'],
|
||||
'program_receipt':receipt,'source_evidence':{'file':'evidence/ismir2026-json-research-v1/papers.json','sha256':receipt['sha256'],'url':receipt['url']},
|
||||
'original_decision_signature_verified':False,'paper_type':None,
|
||||
'id_scope':'Numeric miniconf id/forum fields are program locators, not original OpenReview submission IDs; no inferred DOI or arxiv ID.',
|
||||
'track_scope':'Unclassified regular accepted-list program; no workshop or research-method class inferred from topic/session/remote presentation.',
|
||||
'review_scope':'101 source-opt-in review sets retained as raw public text;39No contain no review text. Reviewer recommendations/meta-review prose are not separately inferred final negative decisions.',
|
||||
'abstract_transform':'Original JSON abstract string retained verbatim, including math, newlines and Unicode; no markup flattening or PDF extraction.',
|
||||
'program_listing_scope':'Source versions remain separate.2title literal differences and3author-list differences flagged, not silently fused or resolved.',
|
||||
'scope':'Pending work/history review. Poster URL follows frozen official JS template; poster pages not downloaded. No work merge, aliases, negative decision or final coverage/dedup claim.'}})
|
||||
path = out/'ismir-program-2026-unclassified.jsonl';raw=''.join(json.dumps(r,ensure_ascii=False)+'\n' for r in rows).encode();path.write_bytes(raw)
|
||||
manifest={'source':SOURCE,'parser_version':1,'venue':'ISMIR','year':2026,'track':TRACK,'count':140,'file':str(path),
|
||||
'file_sha256':digest(raw),'list_url':ACCEPTED,'status':'partial','missing_abstract_count':0,'source_expected_count':140}
|
||||
path.with_suffix('.manifest.json').write_text(json.dumps(manifest,indent=2)+'\n')
|
||||
report={'status':'collected_pending_identity_review','records':140,'files':[manifest],'missing_abstract_count':0,
|
||||
'literal_title_version_differences':2,'author_sequence_version_differences':3,'review_optin_yes':101,'review_optin_no':39,
|
||||
'database_written':False,'network_requests':0,'pdfs_downloaded':False,'approved_for_import':0,
|
||||
'public_rejected_availability':'unknown_accepted_program_reviews_are_not_a_rejected_export',
|
||||
'scope':'140accepted observations;101opt-in review sets, not101rejected records. Source keyword/presentation tags are namespace-scoped; no original OpenReview ID assumption or source-version fusion.'}
|
||||
part=out/'collection-report.json.part';part.write_text(json.dumps(report,ensure_ascii=False,indent=2)+'\n');part.replace(out/'collection-report.json');return report
|
||||
except Exception as error:
|
||||
(out/'failed.json').write_text(json.dumps({'status':'failed_not_complete','error_type':type(error).__name__,'database_written':False})+'\n')
|
||||
raise
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser=argparse.ArgumentParser(prog=prog,description=__doc__);parser.add_argument('--snapshots',required=True);parser.add_argument('--out',required=True)
|
||||
args=parser.parse_args(argv);print(json.dumps(collect(args.out,snapshots=args.snapshots),ensure_ascii=False))
|
||||
@@ -0,0 +1,202 @@
|
||||
"""Restore frozen ACM MM 2026 program observations, not publication decisions."""
|
||||
import argparse
|
||||
from collections import Counter
|
||||
from datetime import datetime
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
|
||||
SOURCE = 'mm_program'
|
||||
SNAPSHOT = 'mm2026-program-and-public-probes-v1'
|
||||
URL = 'https://2026.acmmm.org/site/program-data.json'
|
||||
BODY_SHA = '08e2976ecba2d0d535000b0935f41c85076ec674e518a1df2c5843d9bde4d083'
|
||||
MANIFEST_SHA = 'e43ac6cdfc86bac61305553a044dd6148800b5aa41a5d1bf06c486e90d0ed7b8'
|
||||
TRACK_COUNTS = {'MRAC':35,'SUMAC':9,'NeuroMM':10,'IMPACT':8,'RichMediaGAI':10,
|
||||
'MMSports':16,'MUWS':5,'EReL-MIR':7,'GreenMM':10,'DFF':12,'Main Track':1703,
|
||||
'Grand Challenge':97,'Doctoral Symposium':4,'Datasets':50,'Interactive Art':23,
|
||||
'Reproducibility':1,'Demo and Video':27,'Open Source':14,'Industry Talk':7,
|
||||
'BNI':25,'QoEVMA':4,'HUMA':14,'WIPAI':8,'AMI':11,'Mulsemedia':9,
|
||||
'UAVM':11,'MCHM':7,'JAV-CG':6}
|
||||
CORE = ('paper_id','title','authors','affiliations','abstract','track')
|
||||
PRESENTATION = ('kind','start_time','end_time','poster_id','oral_id')
|
||||
|
||||
|
||||
def digest(raw):
|
||||
return hashlib.sha256(raw).hexdigest()
|
||||
|
||||
|
||||
def safe_read(path, limit):
|
||||
if any(p.is_symlink() for p in (path, *path.parents)) or not path.is_file() or path.stat().st_size > limit:
|
||||
raise ValueError('Unsafe snapshot path or size')
|
||||
return path.read_bytes()
|
||||
|
||||
|
||||
def load_snapshot(root):
|
||||
directory = Path(root)/SNAPSHOT
|
||||
raw = safe_read(directory/'manifest.json', 128*1024)
|
||||
if digest(raw) != MANIFEST_SHA:
|
||||
raise ValueError('Reviewed manifest hash changed')
|
||||
manifest = json.loads(raw)
|
||||
if manifest.get('database_written') is not False or len(manifest['entries']) != 5:
|
||||
raise ValueError('Unreviewed manifest scope')
|
||||
evidence = {'manifest.json':raw}
|
||||
for receipt in manifest['entries']:
|
||||
name = receipt['file']
|
||||
if Path(name).name != name or name in evidence:
|
||||
raise ValueError('Unsafe or duplicate receipt name')
|
||||
body = safe_read(directory/name, 12*1024**2)
|
||||
if (receipt['bytes'] != len(body) or receipt['sha256'] != digest(body)
|
||||
or receipt['url'] != receipt['final_url']
|
||||
or not receipt['content_type'].startswith('application/json')
|
||||
or datetime.fromisoformat(receipt['at']).utcoffset() is None):
|
||||
raise ValueError('Snapshot receipt/body disagree')
|
||||
if name == 'program-data.json':
|
||||
if receipt['url'] != URL or receipt['http_status'] != 200 or digest(body) != BODY_SHA:
|
||||
raise ValueError('Wrong program source')
|
||||
elif receipt['http_status'] != 403:
|
||||
raise ValueError('Unreviewed public-access result')
|
||||
evidence[name] = body
|
||||
if 'program-data.json' not in evidence:
|
||||
raise ValueError('Missing program source')
|
||||
return evidence
|
||||
|
||||
|
||||
def without_emails(value):
|
||||
"""Keep public session structure; contact arrays remain in raw evidence only."""
|
||||
if isinstance(value, dict):
|
||||
return {k:without_emails(v) for k,v in value.items() if k != 'emails'}
|
||||
if isinstance(value, list):
|
||||
return [without_emails(v) for v in value]
|
||||
return value
|
||||
|
||||
|
||||
def program_locations(node, pointer=''):
|
||||
"""Independent structural guard against silently skipping a new nesting form."""
|
||||
if isinstance(node, dict):
|
||||
if 'paper_id' in node:
|
||||
yield pointer
|
||||
for key,value in node.items():
|
||||
escaped = key.replace('~','~0').replace('/','~1')
|
||||
yield from program_locations(value, pointer+'/'+escaped)
|
||||
elif isinstance(node, list):
|
||||
for i,value in enumerate(node):
|
||||
yield from program_locations(value, pointer+'/'+str(i))
|
||||
|
||||
|
||||
def parse_program(raw):
|
||||
data = json.loads(raw)
|
||||
if (data.get('generated_at') != '2026-11-14T18:36:26.860677+00:00'
|
||||
or data.get('days') != [1,2,3,4,5] or len(data.get('rooms',[])) != 20
|
||||
or len(data.get('entries',[])) != 157):
|
||||
raise ValueError('Unreviewed program edition or census')
|
||||
records = {}; sessions = []; seen = []
|
||||
|
||||
def observe(paper, pointer, session_pointer, subsession_pointer=None):
|
||||
if not set(CORE) <= paper.keys() or set(paper)-set(CORE)-set(PRESENTATION)-{'emails'}:
|
||||
raise ValueError('Unreviewed program fields')
|
||||
for key in ('paper_id','title','abstract','track'):
|
||||
if not isinstance(paper[key],str):
|
||||
raise ValueError('Wrong program field type')
|
||||
for key in ('authors','affiliations'):
|
||||
if not isinstance(paper[key],list) or any(not isinstance(v,str) for v in paper[key]):
|
||||
raise ValueError('Wrong author or affiliation list')
|
||||
if not paper['paper_id'].strip() or not paper['title'].strip() or paper['track'] not in TRACK_COUNTS:
|
||||
raise ValueError('Missing identity/title or unreviewed source track')
|
||||
core = {k:paper[k] for k in CORE}
|
||||
pid = paper['paper_id']; seen.append(pointer)
|
||||
if pid not in records:
|
||||
missing = [k for k in ('authors','abstract') if not paper[k]]
|
||||
records[pid] = {**core,'source':SOURCE,'year':2026,'venue':'MM',
|
||||
'source_locator':{'namespace':'mm_program:program_id','value':pid},
|
||||
'observed_status':'listed_in_official_program',
|
||||
'publication_identity_verified':False,'final_decision_verified':False,
|
||||
'approved_for_import':False,'missing_fields':missing,
|
||||
'tags':[{'namespace':'mm_program:track','value':paper['track']}],
|
||||
'contexts':[]}
|
||||
elif {k:records[pid][k] for k in CORE} != core:
|
||||
raise ValueError('Conflicting core versions for one program ID; review required')
|
||||
records[pid]['contexts'].append({'source_pointer':pointer,'session_pointer':session_pointer,
|
||||
'subsession_pointer':subsession_pointer,
|
||||
'presentation':{k:paper[k] for k in PRESENTATION if k in paper}})
|
||||
|
||||
for i,entry in enumerate(data['entries']):
|
||||
pointer = '/entries/'+str(i)
|
||||
if entry['track'] != '':
|
||||
raise ValueError('Unreviewed outer track; do not override paper.track')
|
||||
session = without_emails({k:v for k,v in entry.items() if k not in ('papers','subsessions')})
|
||||
session.update(source_pointer=pointer,subsessions=[])
|
||||
for j,paper in enumerate(entry.get('papers') or []):
|
||||
observe(paper,pointer+'/papers/'+str(j),pointer)
|
||||
for j,sub in enumerate(entry.get('subsessions') or []):
|
||||
sp = pointer+'/subsessions/'+str(j)
|
||||
saved = without_emails({k:v for k,v in sub.items() if k != 'items'})
|
||||
saved.update(source_pointer=sp,nonpaper_items=[])
|
||||
for k,item in enumerate(sub.get('items') or []):
|
||||
ip = sp+'/items/'+str(k)
|
||||
if 'paper_id' in item:
|
||||
observe(item,ip,pointer,sp)
|
||||
else:
|
||||
saved['nonpaper_items'].append({'source_pointer':ip,'fields':without_emails(item)})
|
||||
session['subsessions'].append(saved)
|
||||
sessions.append(session)
|
||||
if set(seen) != set(program_locations(data)) or len(set(seen)) != len(seen):
|
||||
raise ValueError('Unhandled or duplicated program location')
|
||||
items = list(records.values())
|
||||
if (len(seen) != 4326 or len(items) != 2143
|
||||
or Counter(r['track'] for r in items) != TRACK_COUNTS
|
||||
or sum(not r['abstract'] for r in items) != 216
|
||||
or {r['paper_id'] for r in items if not r['authors']} != {'mfp4971','mfp7801'}):
|
||||
raise ValueError('Program count or missing-field census changed')
|
||||
return items,sessions
|
||||
|
||||
|
||||
def collect(out, *, snapshots):
|
||||
out = Path(out)
|
||||
if out.exists() or out.is_symlink():
|
||||
raise ValueError('Output exists; use a fresh directory')
|
||||
if any(p.is_symlink() for p in out.parents):
|
||||
raise ValueError('Unsafe output parent')
|
||||
out.mkdir(parents=True)
|
||||
try:
|
||||
evidence = load_snapshot(snapshots)
|
||||
rows,sessions = parse_program(evidence['program-data.json'])
|
||||
(out/'evidence').mkdir()
|
||||
for name,body in evidence.items():
|
||||
(out/'evidence'/name).write_bytes(body)
|
||||
outputs = []
|
||||
for track,count in TRACK_COUNTS.items():
|
||||
slug = re.sub('[^a-z0-9]+','_',track.lower()).strip('_')
|
||||
path = out/('mm-program-2026-'+slug+'.jsonl')
|
||||
raw = ''.join(json.dumps(r,ensure_ascii=False)+'\n' for r in rows if r['track']==track).encode()
|
||||
path.write_bytes(raw)
|
||||
outputs.append({'file':str(path),'file_sha256':digest(raw),'count':count,
|
||||
'track_literal':track,'format':'mm_program_observation_v1_not_ingestion_record',
|
||||
'status':'pending_contribution_and_identity_review','approved_for_import':0})
|
||||
session_raw = ''.join(json.dumps(r,ensure_ascii=False)+'\n' for r in sessions).encode()
|
||||
(out/'sessions.jsonl').write_bytes(session_raw)
|
||||
report = {'status':'collected_pending_contribution_and_identity_review','source':SOURCE,
|
||||
'records':2143,'program_observations':4326,'extra_presentation_observations':2183,
|
||||
'track_counts':TRACK_COUNTS,'files':outputs,'session_count':157,
|
||||
'sessions_file_sha256':digest(session_raw),'missing_abstract_count':216,
|
||||
'missing_author_program_ids':['mfp4971','mfp7801'],
|
||||
'source_body_sha256':BODY_SHA,'source_manifest_sha256':MANIFEST_SHA,
|
||||
'database_written':False,'pdfs_downloaded':False,'network_requests':0,'approved_for_import':0,
|
||||
'public_negative_access':{'anonymous_notes_probes':4,'http_statuses':[403]*4,
|
||||
'negative_decisions_inferred':0,'scope':'Observed endpoints only; not proof of zero public negative decisions.'},
|
||||
'scope':'Program IDs are source locators, not scientific-work or original submission IDs. All28source tracks retained; presentation times/kinds/locators and session context kept separately. No publication status, paper type, DOI, arxiv/OpenReview ID, rejection or withdrawal inferred. Blank fields are not negative decisions. Contribution/publication/work/history review and final whole-database audit remain.'}
|
||||
part = out/'collection-report.json.part'
|
||||
part.write_text(json.dumps(report,ensure_ascii=False,indent=2)+'\n')
|
||||
part.replace(out/'collection-report.json')
|
||||
return report
|
||||
except Exception as error:
|
||||
(out/'failed.json').write_text(json.dumps({'status':'failed_not_complete','error_type':type(error).__name__,'database_written':False})+'\n')
|
||||
raise
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog,description=__doc__)
|
||||
parser.add_argument('--snapshots',required=True)
|
||||
parser.add_argument('--out',required=True)
|
||||
args = parser.parse_args(argv)
|
||||
print(json.dumps(collect(args.out,snapshots=args.snapshots),ensure_ascii=False))
|
||||
@@ -0,0 +1,336 @@
|
||||
"""Frozen MM 2026 registered publications with separately attributed program evidence.
|
||||
|
||||
All publications survive normalization. A program metadata match is not a
|
||||
scientific-work identity decision; publisher article type is not research type.
|
||||
"""
|
||||
import argparse
|
||||
from collections import Counter, defaultdict
|
||||
from datetime import datetime
|
||||
import hashlib
|
||||
from html.parser import HTMLParser
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
import unicodedata
|
||||
from urllib.parse import parse_qs, urlsplit
|
||||
|
||||
from paper_atlas.collectors import mm2026 as program
|
||||
|
||||
SOURCE = 'mm_registry_program'
|
||||
SNAPSHOT = 'mm2026-container-registry-v2'
|
||||
MANIFEST_SHA = '128b9d02c72261a66edb9a4537d118bdccab3ed80d49a1e971fefdac399f9855'
|
||||
PARENT_SHA = 'bc3ce7da2914f9ddd198c928b3acb2734a8370add3348c23c5f9f12135381628'
|
||||
PARENT_DOI = '10.1145/3767308'
|
||||
CONTAINER = 'Proceedings of the 34th ACM International Conference on Multimedia'
|
||||
EVENT = "MM '26: The 34th ACM International Conference on Multimedia"
|
||||
TRACKS = {'Main Track':'main','Grand Challenge':'grand_challenge',
|
||||
'Interactive Art':'interactive_art','Datasets':'datasets','Reproducibility':'reproducibility',
|
||||
'BNI':'bni','Open Source':'open_source','Doctoral Symposium':'doctoral_symposium',
|
||||
'Demo and Video':'demo_and_video','Industry Talk':'industry_talk'}
|
||||
BIBLIOGRAPHY = {'DOI','URL','title','original-title','subtitle','short-title','author',
|
||||
'publisher','member','type','container-title','short-container-title','event','abstract',
|
||||
'page','ISBN','isbn-type','published','published-print','published-online','issued',
|
||||
'license','resource','relation','alternative-id'}
|
||||
|
||||
|
||||
def digest(raw):
|
||||
return hashlib.sha256(raw).hexdigest()
|
||||
|
||||
|
||||
def key(text):
|
||||
return ''.join(c for c in unicodedata.normalize('NFKD',text).casefold() if c.isalnum())
|
||||
|
||||
|
||||
def author_names(values):
|
||||
return Counter(key(v) for v in values)
|
||||
|
||||
|
||||
def load_registry(root, parent_path):
|
||||
directory = Path(root)/SNAPSHOT
|
||||
raw_manifest = program.safe_read(directory/'manifest.json', 128*1024)
|
||||
raw_parent = program.safe_read(Path(parent_path), 128*1024)
|
||||
if digest(raw_manifest) != MANIFEST_SHA or digest(raw_parent) != PARENT_SHA:
|
||||
raise ValueError('Reviewed registry or parent hash changed')
|
||||
manifest = json.loads(raw_manifest);parent = json.loads(raw_parent)['message']
|
||||
if (manifest['status'] != 'complete_exact_container_registry_capture_not_work_or_track_approval'
|
||||
or manifest['unique_registry_DOIs'] != 1918 or manifest['expected_registry_items'] != 1918
|
||||
or len(manifest['pages']) != 8 or manifest['database_written'] is not False
|
||||
or manifest['approved_for_import'] != 0 or manifest['parent_body_sha256'] != PARENT_SHA
|
||||
or parent['DOI'] != PARENT_DOI or parent['type'] != 'proceedings'
|
||||
or parent['title'] != [CONTAINER] or parent['event']['name'] != EVENT
|
||||
or parent['member'] != '320' or parent['ISBN'] != ['9798400722134']):
|
||||
raise ValueError('Incomplete reviewed registry or wrong proceedings')
|
||||
evidence = {'manifest.json':raw_manifest,'parent.json':raw_parent};rows=[];seen=set()
|
||||
for number,receipt in enumerate(manifest['pages'],1):
|
||||
name=receipt['file']
|
||||
if name != f'page-{number:03}.json':
|
||||
raise ValueError('Unexpected registry page filename')
|
||||
body=program.safe_read(directory/name,8*1024**2)
|
||||
url=urlsplit(receipt['url']);query=parse_qs(url.query)
|
||||
expected_filter='container-title:'+CONTAINER+',member:320'
|
||||
if (receipt['http_status'] != 200 or receipt['url'] != receipt['final_url']
|
||||
or url.scheme != 'https' or url.netloc != 'api.crossref.org' or url.path != '/works'
|
||||
or set(query) != {'filter','rows','cursor'} or query['filter'] != [expected_filter]
|
||||
or query['rows'] != ['250'] or not query['cursor'][0]
|
||||
or datetime.fromisoformat(receipt['at']).utcoffset() is None
|
||||
or receipt['bytes'] != len(body) or receipt['sha256'] != digest(body)):
|
||||
raise ValueError('Registry receipt and source body disagree')
|
||||
obj=json.loads(body);message=obj['message'];expected=250 if number<8 else 168
|
||||
if (obj['status'] != 'ok' or message['total-results'] != 1918
|
||||
or len(message['items']) != expected or receipt['items'] != expected):
|
||||
raise ValueError('Incomplete registry pagination')
|
||||
evidence[name]=body
|
||||
for position,original in enumerate(message['items']):
|
||||
doi=original.get('DOI')
|
||||
if not isinstance(doi,str) or not re.fullmatch(r'10\.1145/3767308\.[0-9]+',doi) or doi in seen:
|
||||
raise ValueError('Missing, foreign or repeated publication DOI')
|
||||
seen.add(doi)
|
||||
if (original.get('URL') != 'https://doi.org/'+doi or original.get('type') != 'proceedings-article'
|
||||
or original.get('container-title') != [CONTAINER] or original.get('event',{}).get('name') != EVENT
|
||||
or original.get('member') != '320' or original.get('publisher') != 'ACM'
|
||||
or original.get('published') != {'date-parts':[[2026,11,9]]} or original.get('abstract')):
|
||||
raise ValueError('Wrong registered publication context or changed abstract availability')
|
||||
titles=original.get('title');authors=original.get('author')
|
||||
if (not isinstance(titles,list) or len(titles)!=1 or not isinstance(titles[0],str) or not titles[0].strip()
|
||||
or not isinstance(authors,list) or not authors):
|
||||
raise ValueError('Missing publication title or authors')
|
||||
names=[];flags=[]
|
||||
for i,author in enumerate(authors,1):
|
||||
family=author.get('family');given=author.get('given','')
|
||||
if not isinstance(family,str) or not family.strip() or not isinstance(given,str):
|
||||
raise ValueError('Invalid registered author fields')
|
||||
name=' '.join(x.strip() for x in (given,family) if x.strip())
|
||||
if '@' in name or name.casefold()=='anonymous':
|
||||
raise ValueError('Unreviewed author identity')
|
||||
names.append(name)
|
||||
if not given.strip():flags.append({'kind':'single_registered_name_field_not_split','author_position':i})
|
||||
if re.search(r'<[^>]+>|&[A-Za-z#0-9]+;',titles[0]):
|
||||
flags.append({'kind':'literal_title_markup_or_entity_retained_requires_review'})
|
||||
canonical=json.dumps(original,ensure_ascii=False,sort_keys=True,separators=(',',':')).encode()
|
||||
rows.append({'doi':doi,'title':titles[0],'authors':names,'flags':flags,
|
||||
'fields':{k:v for k,v in original.items() if k in BIBLIOGRAPHY},
|
||||
'original_field_names':sorted(original),'original_canonical_json_sha256':digest(canonical),
|
||||
'receipt':receipt,'source_position':position})
|
||||
if len(rows)!=1918:
|
||||
raise ValueError('Incomplete registered publication census')
|
||||
return evidence,rows
|
||||
|
||||
|
||||
|
||||
class _TitleFormatting(HTMLParser):
|
||||
"""Decode text once, removing only balanced, attribute-free reviewed tags."""
|
||||
def __init__(self):
|
||||
super().__init__(convert_charrefs=True)
|
||||
self.parts=[];self.stack=[];self.valid=True
|
||||
|
||||
def handle_data(self, text):
|
||||
self.parts.append(text)
|
||||
|
||||
def handle_starttag(self, tag, attrs):
|
||||
if tag not in {'i','scp','sup','sub'} or attrs:
|
||||
self.valid=False
|
||||
else:
|
||||
self.stack.append(tag)
|
||||
|
||||
def handle_endtag(self, tag):
|
||||
if not self.stack or self.stack[-1]!=tag:
|
||||
self.valid=False
|
||||
else:
|
||||
self.stack.pop()
|
||||
|
||||
def handle_startendtag(self, tag, attrs):
|
||||
self.valid=False
|
||||
|
||||
def handle_comment(self, text):
|
||||
self.valid=False
|
||||
|
||||
def handle_decl(self, text):
|
||||
self.valid=False
|
||||
|
||||
def handle_pi(self, text):
|
||||
self.valid=False
|
||||
|
||||
def unknown_decl(self, text):
|
||||
self.valid=False
|
||||
|
||||
|
||||
def formatting_text(title):
|
||||
# HTMLParser may silently discard a trailing incomplete start tag.
|
||||
# Check lexical coverage before parsing so no title content can disappear.
|
||||
literal=re.sub(r'</?(?:i|scp|sup|sub)\s*>','',title,flags=re.I)
|
||||
if '<' in literal or '>' in literal:
|
||||
return None
|
||||
parser=_TitleFormatting()
|
||||
parser.feed(title);parser.close()
|
||||
text=''.join(parser.parts)
|
||||
return text if parser.valid and not parser.stack and text.strip() and '<' not in text and '>' not in text else None
|
||||
|
||||
|
||||
FORMATTING_MATCHES = {'10.1145/3767308.3834907': 'mfp0053',
|
||||
'10.1145/3767308.3834936': 'mfp0229',
|
||||
'10.1145/3767308.3834946': 'mfp0297',
|
||||
'10.1145/3767308.3834956': 'mfp0339',
|
||||
'10.1145/3767308.3835035': 'mfp0798',
|
||||
'10.1145/3767308.3835055': 'mfp0903',
|
||||
'10.1145/3767308.3835101': 'mfp1175',
|
||||
'10.1145/3767308.3835374': 'mfp2617',
|
||||
'10.1145/3767308.3835424': 'mfp2909',
|
||||
'10.1145/3767308.3835452': 'mfp3048',
|
||||
'10.1145/3767308.3835460': 'mfp3109',
|
||||
'10.1145/3767308.3835544': 'mfp3554',
|
||||
'10.1145/3767308.3835591': 'mfp3783',
|
||||
'10.1145/3767308.3835667': 'mfp4173',
|
||||
'10.1145/3767308.3835763': 'mfp4778',
|
||||
'10.1145/3767308.3835895': 'mfp5537',
|
||||
'10.1145/3767308.3835959': 'mfp5823',
|
||||
'10.1145/3767308.3836215': 'mfp7332',
|
||||
'10.1145/3767308.3836272': 'mfp7635',
|
||||
'10.1145/3767308.3836293': 'mfp7732',
|
||||
'10.1145/3767308.3836447': 'mfp8648',
|
||||
'10.1145/3767308.3836600': 'mfp9815',
|
||||
'10.1145/3767308.3838656': 'mds177',
|
||||
'10.1145/3767308.3839012': 'mde41',
|
||||
'10.1145/3767308.3839024': 'mde15',
|
||||
'10.1145/3767308.3839028': 'mde02'}
|
||||
|
||||
|
||||
def match_program(rows, observations):
|
||||
titles=defaultdict(list);formatted=defaultdict(list)
|
||||
for p in observations:
|
||||
titles[key(p['title'])].append(p)
|
||||
text=formatting_text(p['title'])
|
||||
if text is not None:formatted[key(text)].append(p)
|
||||
matched=set();kinds=Counter()
|
||||
for row in rows:
|
||||
row.pop('program',None);row.pop('formatting_comparison',None)
|
||||
candidates=titles[key(row['title'])]
|
||||
method='literal_title'
|
||||
# A literal-title author disagreement must not be bypassed by fallback.
|
||||
if not candidates:
|
||||
text=formatting_text(row['title'])
|
||||
if text is not None:
|
||||
candidates=formatted[key(text)]
|
||||
if candidates:method='reviewed_title_formatting'
|
||||
exact=[p for p in candidates if author_names(p['authors'])==author_names(row['authors'])]
|
||||
if len(exact)==1:
|
||||
row['kind']=('unique_program_format_title_fullauthor_agreement_not_identity_approval'
|
||||
if method=='reviewed_title_formatting' else 'unique_program_title_fullauthor_agreement_not_identity_approval')
|
||||
row['program']=exact[0];matched.add(exact[0]['paper_id'])
|
||||
if exact[0]['track'] not in TRACKS:
|
||||
raise ValueError('New registered program track requires review')
|
||||
elif len(exact)>1:row['kind']='multiple_program_ids_same_title_authors_hold'
|
||||
elif candidates:
|
||||
row['kind']=('program_format_author_version_requires_review'
|
||||
if method=='reviewed_title_formatting' else 'program_author_version_requires_review')
|
||||
else:row['kind']='no_exact_program_title_requires_review'
|
||||
row['program_title_candidates']=[p['paper_id'] for p in candidates]
|
||||
row['program_exact_candidates']=[p['paper_id'] for p in exact]
|
||||
if method=='reviewed_title_formatting':
|
||||
row['formatting_comparison']={'method':'balanced_attribute_free_i_scp_sup_sub_and_single_entity_decode_then_nfkd_alnum',
|
||||
'registry_comparison_text':formatting_text(row['title']),
|
||||
'program_comparison_texts':{p['paper_id']:formatting_text(p['title']) for p in candidates},
|
||||
'raw_titles_preserved':True,'work_identity_approved':False}
|
||||
kinds[row['kind']]+=1
|
||||
return dict(kinds),[p for p in observations if p['paper_id'] not in matched]
|
||||
|
||||
|
||||
def correlate(rows, observations):
|
||||
kinds,unmatched=match_program(rows,observations)
|
||||
expected={'unique_program_title_fullauthor_agreement_not_identity_approval':1726,
|
||||
'unique_program_format_title_fullauthor_agreement_not_identity_approval':26,
|
||||
'program_author_version_requires_review':122,
|
||||
'program_format_author_version_requires_review':1,
|
||||
'no_exact_program_title_requires_review':43}
|
||||
actual={r['doi']:r['program']['paper_id'] for r in rows
|
||||
if r['kind']=='unique_program_format_title_fullauthor_agreement_not_identity_approval'}
|
||||
if kinds != expected or actual != FORMATTING_MATCHES:
|
||||
raise ValueError('Reviewed program/publication correspondence changed')
|
||||
return kinds,unmatched
|
||||
|
||||
|
||||
def collect(out, *, registry_snapshots, program_snapshots, parent_snapshot):
|
||||
out=Path(out)
|
||||
if out.exists() or out.is_symlink():
|
||||
raise ValueError('Output exists; inspect prior operation, never overwrite')
|
||||
if any(p.is_symlink() for p in out.parents):
|
||||
raise ValueError('Unsafe output parent')
|
||||
out.mkdir(parents=True)
|
||||
try:
|
||||
registry_evidence,publications=load_registry(registry_snapshots,parent_snapshot)
|
||||
program_evidence=program.load_snapshot(program_snapshots)
|
||||
observations,_=program.parse_program(program_evidence['program-data.json'])
|
||||
kinds,unmatched=correlate(publications,observations)
|
||||
for group,files in [('registry',registry_evidence),('program',program_evidence)]:
|
||||
directory=out/'evidence'/group;directory.mkdir(parents=True)
|
||||
for name,body in files.items():(directory/name).write_bytes(body)
|
||||
buckets=defaultdict(list);missing=0;copied=0;known_tracks=Counter()
|
||||
for item in sorted(publications,key=lambda r:r['doi']):
|
||||
p=item.get('program');doi=item['doi'];receipt=item['receipt']
|
||||
track=TRACKS[p['track']] if p else 'proceedings_unclassified'
|
||||
abstract=p['abstract'] if p and p['abstract'] else None
|
||||
missing+=abstract is None;copied+=abstract is not None
|
||||
if p:known_tracks[p['track']]+=1
|
||||
tags=[{'namespace':SOURCE+':crossref:type','value':'proceedings-article'}]
|
||||
if p:tags.append({'namespace':SOURCE+':mm_program:track','value':p['track']})
|
||||
buckets[track].append({'source':SOURCE,'external_id':doi,'url':'https://doi.org/'+doi,
|
||||
'title':item['title'],'authors':item['authors'],'abstract':abstract,
|
||||
'identifiers':{'doi':doi},'submission_identifiers':[{'scheme':'publication_doi','value':doi}],
|
||||
'venue':'MM','year':2026,'track':track,'status':'accepted',
|
||||
'raw_decision':'Registered publication in MM2026 proceedings; not an original signed decision.',
|
||||
'evidence_url':receipt['url'],'observed_at':receipt['at'],'tags':tags,
|
||||
'metadata':{**({'program_title_formatting_comparison':item['formatting_comparison']} if 'formatting_comparison' in item else {}),'registry_fields':item['fields'],'registry_original_field_names':item['original_field_names'],
|
||||
'registry_source_evidence':{'file':'evidence/registry/'+receipt['file'],'sha256':receipt['sha256'],
|
||||
'url':receipt['url'],'json_pointer':'/message/items/'+str(item['source_position']),
|
||||
'canonical_object_json_sha256':item['original_canonical_json_sha256']},
|
||||
'registry_manifest':{'file':'evidence/registry/manifest.json','sha256':MANIFEST_SHA},
|
||||
'parent_evidence':{'file':'evidence/registry/parent.json','sha256':PARENT_SHA},
|
||||
'program_source_evidence':{'file':'evidence/program/program-data.json','sha256':program.BODY_SHA},
|
||||
'program_fields':p,'program_match_kind':item['kind'],
|
||||
'program_title_candidates':item['program_title_candidates'],
|
||||
'program_exact_candidates':item['program_exact_candidates'],
|
||||
'abstract_source':('official_program_reviewed_title_formatting_correspondence' if item.get('formatting_comparison') else 'official_program_exact_metadata_correspondence') if abstract is not None else 'not_available_from_reviewed_sources',
|
||||
'registry_abstract_available':False,'original_decision_signature_verified':False,
|
||||
'scientific_work_identity_approved':False,'research_type_classified':False,
|
||||
'review_flags':item['flags'],
|
||||
'scope':'Derived publication/program metadata with separate original evidence. All registered articles retained, including art/industry/demo; bibliographic type is not research type. Program correlation does not approve work identity or person aliases. Publication dates are not decision dates. DOI is publication evidence, not original OpenReview ID.'}})
|
||||
outputs=[]
|
||||
for track,rows in sorted(buckets.items()):
|
||||
path=out/f'mm-publications-2026-{track}.jsonl'
|
||||
raw=''.join(json.dumps(row,ensure_ascii=False)+'\n' for row in rows).encode();path.write_bytes(raw)
|
||||
manifest={'source':SOURCE,'parser_version':2,'venue':'MM','year':2026,'track':track,'count':len(rows),
|
||||
'file':str(path),'file_sha256':digest(raw),'list_url':'https://doi.org/'+PARENT_DOI,
|
||||
'status':'partial','missing_abstract_count':sum(row['abstract'] is None for row in rows),
|
||||
'scope':'Registered container partition only; not allMMprogram/workshops, independent works or public-negative coverage.'}
|
||||
path.with_suffix('.manifest.json').write_text(json.dumps(manifest,ensure_ascii=False,indent=2)+'\n');outputs.append(manifest)
|
||||
report={'status':'collected_publications_pending_work_identity_review','records':1918,'unique_registered_dois':1918,
|
||||
'files':outputs,'program_ids':2143,'program_observations':4326,'program_match_counts':kinds,
|
||||
'matched_program_track_counts':dict(known_tracks),'unmatched_program':unmatched,
|
||||
'unmatched_program_count':len(unmatched),'registry_missing_abstract_count':1918,
|
||||
'abstracts_from_exact_program_correspondence':copied,'output_missing_abstract_count':missing,
|
||||
'abstracts_from_literal_program_correspondence':copied-26,'abstracts_from_reviewed_title_formatting':26,
|
||||
'registered_single_name_fields':sum(f['kind']=='single_registered_name_field_not_split' for r in publications for f in r['flags']),
|
||||
'registered_title_markup_flags':sum(any(f['kind']=='literal_title_markup_or_entity_retained_requires_review' for f in r['flags']) for r in publications),
|
||||
'database_written':False,'approved_for_import':0,'network_requests':0,'pdfs_downloaded':False,
|
||||
'public_negative_availability':'unknown_no_negative_decisions_inferred',
|
||||
'scope':'1918registeredpublication DOIs, not1918proven distinct scientificworks.166title/authorvariants remainunclassified,391unmatchedprogram items retained.26formattingcorrespondences retainedseparately;rawtitlesunchanged. All28programtracks remain independent census; article type never implies research. Per-work/source/DOI/history review andfinalwholeDBaudit remain.'}
|
||||
if (sum(f['count'] for f in outputs)!=1918 or copied!=1743 or missing!=175 or len(unmatched)!=391):
|
||||
raise ValueError('Reviewed output partition or abstract census changed')
|
||||
part=out/'collection-report.json.part';part.write_text(json.dumps(report,ensure_ascii=False,indent=2)+'\n');part.replace(out/'collection-report.json')
|
||||
return report
|
||||
except Exception as error:
|
||||
(out/'failed.json').write_text(json.dumps({'status':'failed_not_complete','error_type':type(error).__name__,'database_written':False})+'\n')
|
||||
raise
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser=argparse.ArgumentParser(prog=prog,description=__doc__)
|
||||
parser.add_argument('--registry-snapshots',required=True)
|
||||
parser.add_argument('--program-snapshots',required=True)
|
||||
parser.add_argument('--parent-snapshot',required=True)
|
||||
parser.add_argument('--out',required=True)
|
||||
args=parser.parse_args(argv)
|
||||
report=collect(args.out,registry_snapshots=args.registry_snapshots,
|
||||
program_snapshots=args.program_snapshots,parent_snapshot=args.parent_snapshot)
|
||||
print(json.dumps({k:report[k] for k in ('status','records','files','program_match_counts',
|
||||
'abstracts_from_exact_program_correspondence','output_missing_abstract_count',
|
||||
'unmatched_program_count','database_written','approved_for_import')},ensure_ascii=False))
|
||||
@@ -0,0 +1,216 @@
|
||||
"""Source-attributed mirror metadata; no paper/review text or identity promotion."""
|
||||
import argparse
|
||||
from collections import Counter
|
||||
from datetime import datetime, timezone
|
||||
import hashlib
|
||||
from importlib.resources import files
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
|
||||
from paper_atlas.io.parquet_metadata import RangeSource, projection
|
||||
|
||||
DATASET = 'anonymousNeurIPS2026submission4281/reviewarena'
|
||||
COMMIT = 'c2978add17c2099219eaddbc2599974d69d4d09b'
|
||||
TREE_SHA256 = 'bbc75f5774b019c4a9ee3efe5c0f37b14761bc3c5a054570db1df8f5603eac20'
|
||||
SOURCE = 'reviewarena_mirror'
|
||||
SPLITS = ('colm', 'corl', 'emnlp', 'iclr', 'icml', 'neurips', 'tmlr')
|
||||
COLUMNS = {'forum_id': 'string', 'conference': 'string', 'year': 'integer',
|
||||
'track': 'string', 'venue_id': 'string', 'paper_number': 'integer',
|
||||
'title': 'string', 'abstract': 'string', 'authors': 'strings',
|
||||
'keywords': 'strings', 'venue': 'string', 'decision': 'string'}
|
||||
|
||||
|
||||
def inventory(split):
|
||||
if split not in SPLITS:
|
||||
raise ValueError('Explicit supported mirror split required')
|
||||
raw = files('paper_atlas.resources').joinpath('reviewarena-files.json').read_bytes()
|
||||
if hashlib.sha256(raw).hexdigest() != TREE_SHA256:
|
||||
raise ValueError('Pinned mirror file inventory changed')
|
||||
items = json.loads(raw)
|
||||
if len(items) != 15:
|
||||
raise ValueError('Pinned mirror shard inventory changed')
|
||||
selected = []
|
||||
for item in items:
|
||||
path = item['path']
|
||||
if not re.fullmatch(r'data/(colm|corl|emnlp|iclr|icml|neurips|tmlr)-\d{5}-of-\d{5}\.parquet', path):
|
||||
raise ValueError('Unexpected mirror shard path')
|
||||
if (item['type'] != 'file' or type(item['size']) is not int or item['size'] < 12
|
||||
or item['lfs']['size'] != item['size'] or not re.fullmatch('[0-9a-f]{64}', item['lfs']['oid'])):
|
||||
raise ValueError('Missing source object size/checksum claim')
|
||||
if path.startswith('data/'+split+'-'):
|
||||
selected.append(item)
|
||||
parts = sorted(selected, key=lambda x: x['path'])
|
||||
if not parts or any(int(x['path'][-13:-8]) != len(parts) for x in parts):
|
||||
raise ValueError('Incomplete selected mirror split inventory')
|
||||
if [int(x['path'].split('-')[-3]) for x in parts] != list(range(len(parts))):
|
||||
raise ValueError('Mirror shard sequence changed')
|
||||
return parts
|
||||
|
||||
|
||||
def decision_claim(raw):
|
||||
value = raw.strip().casefold() if isinstance(raw, str) else ''
|
||||
if value in {'reject', 'rejected'}:
|
||||
return 'rejected'
|
||||
if value in {'desk reject', 'desk rejected', 'desk_rejected'}:
|
||||
return 'desk_rejected'
|
||||
if value in {'withdraw', 'withdrawn'}:
|
||||
return 'withdrawn'
|
||||
if value in {'accept', 'accepted', 'accept-main', 'accept-findings', 'accept (main)', 'accept (findings)',
|
||||
'accept (poster)', 'accept (oral)', 'accept (spotlight)', 'accept (talk)'}:
|
||||
return 'accepted'
|
||||
return 'unknown'
|
||||
|
||||
|
||||
def decision_track_claim(raw):
|
||||
value = raw.strip().casefold() if isinstance(raw, str) else ''
|
||||
if value in {'accept-findings', 'accept (findings)'}:
|
||||
return 'findings'
|
||||
if value in {'accept-main', 'accept (main)'}:
|
||||
return 'main'
|
||||
return None
|
||||
|
||||
|
||||
def record(row, split, item, row_index):
|
||||
if set(row) != set(COLUMNS):
|
||||
raise ValueError('Unexpected projected metadata fields')
|
||||
if (not isinstance(row['conference'], str) or row['conference'].casefold() != split
|
||||
or type(row['year']) is not int or not 2020 <= row['year'] <= 2026):
|
||||
raise ValueError('Mirror conference/year outside pinned source scope')
|
||||
locator = {'dataset': DATASET, 'commit': COMMIT, 'file': item['path'], 'row_index': row_index,
|
||||
'lfs_sha256_claim_not_full_file_verification': item['lfs']['oid']}
|
||||
return {'source': SOURCE, 'external_id': COMMIT+':'+item['path']+':'+str(row_index),
|
||||
'url': 'https://huggingface.co/datasets/'+DATASET+'/blob/'+COMMIT+'/'+item['path'],
|
||||
'title': row['title'], 'abstract': row['abstract'], 'authors': row['authors'],
|
||||
'conference': row['conference'], 'year': row['year'], 'track': row['track'],
|
||||
'mirror_decision_category': decision_claim(row['decision']),
|
||||
'mirror_decision_track_claim': decision_track_claim(row['decision']), 'raw_fields': row,
|
||||
'source_locator': locator, 'identifiers': {},
|
||||
'metadata': {'forum_id_claim': row['forum_id'], 'original_forum_id_verified': False,
|
||||
'original_decision_signature_verified': False, 'source_visibility': 'public_mirror',
|
||||
'scientific_work_identity_approved': False, 'research_type_classified': False,
|
||||
'approved_for_import': False,
|
||||
'scope': 'Mirror row only. No original forum/person/decision verification; absence is not rejection or proof of no later acceptance.'}}
|
||||
|
||||
|
||||
def cached_projections(path, expected_sha256, split, items):
|
||||
path = Path(path)
|
||||
raw = path.read_bytes()
|
||||
if (not isinstance(expected_sha256, str) or not re.fullmatch('[0-9a-f]{64}', expected_sha256)
|
||||
or hashlib.sha256(raw).hexdigest() != expected_sha256):
|
||||
raise ValueError('Cached projection report hash differs')
|
||||
report = json.loads(raw)
|
||||
if (report.get('dataset') != DATASET or report.get('commit') != COMMIT
|
||||
or report.get('split') != split or report.get('inventory_sha256') != TREE_SHA256
|
||||
or report.get('raw_parquet_saved') is not False or report.get('database_written') is not False
|
||||
or report.get('paper_body_saved') is not False or report.get('review_text_saved') is not False
|
||||
or report.get('approved_for_import') != 0):
|
||||
raise ValueError('Cached report scope differs from permitted mirror projection')
|
||||
declared = report.get('files', [])
|
||||
if len(declared) != len(items) or len({f['source_file'] for f in declared}) != len(items):
|
||||
raise ValueError('Incomplete cached mirror shard set')
|
||||
result = {}
|
||||
for item in items:
|
||||
fm = next((f for f in declared if f['source_file'] == item['path']), None)
|
||||
filename = Path(item['path']).stem+'.metadata.jsonl'
|
||||
if not fm or Path(fm['file']).name != filename or fm['source_lfs_sha256_claim'] != item['lfs']['oid']:
|
||||
raise ValueError('Cached mirror source differs')
|
||||
file = path.parent/filename
|
||||
if file.is_symlink():
|
||||
raise ValueError('Symlink metadata cache forbidden')
|
||||
content = file.read_bytes()
|
||||
if hashlib.sha256(content).hexdigest() != fm['file_sha256']:
|
||||
raise ValueError('Cached metadata file hash differs')
|
||||
rows = []
|
||||
for index, line in enumerate(content.splitlines()):
|
||||
previous = json.loads(line); row = previous['raw_fields']; expected = record(row, split, item, index)
|
||||
for key in ('source', 'external_id', 'url', 'title', 'abstract', 'authors', 'conference', 'year', 'track', 'raw_fields', 'source_locator', 'identifiers', 'metadata'):
|
||||
if previous.get(key) != expected[key]:
|
||||
raise ValueError('Cached raw metadata/identity differs: '+key)
|
||||
rows.append(row)
|
||||
proof = fm['projection']
|
||||
if (len(rows) != fm['records'] or len(rows) != proof['rows'] or proof['columns'] != COLUMNS
|
||||
or proof['source_bytes'] != item['size'] or proof['raw_parquet_saved'] is not False
|
||||
or proof['footer_saved'] is not False or proof['excluded_payload_pages_downloaded'] is not False
|
||||
or proof['full_file_checksum_verified'] is not False):
|
||||
raise ValueError('Cached projection proof differs')
|
||||
result[item['path']] = (rows, proof)
|
||||
if sum(len(rows) for rows, _ in result.values()) != report['records']:
|
||||
raise ValueError('Cached projection row count differs')
|
||||
return result
|
||||
|
||||
|
||||
def collect(split, out, *, source_factory=RangeSource, progress=None, from_projection=None, projection_sha256=None):
|
||||
items = inventory(split); out = Path(out)
|
||||
if bool(from_projection) != bool(projection_sha256):
|
||||
raise ValueError('Cached projection and exact report hash required together')
|
||||
if out.exists() or not out.parent.is_dir() or out.is_symlink():
|
||||
raise ValueError('New output directory in existing parent required')
|
||||
out.mkdir()
|
||||
counts = Counter(); track_claims = Counter(); raw_counts = Counter(); years = Counter(); records = 0; reports = []
|
||||
observed = datetime.now(timezone.utc).isoformat()
|
||||
try:
|
||||
cached = cached_projections(from_projection, projection_sha256, split, items) if from_projection else None
|
||||
for item in items:
|
||||
if cached is None:
|
||||
source_url = 'https://huggingface.co/datasets/'+DATASET+'/resolve/'+COMMIT+'/'+item['path']
|
||||
source = source_factory(source_url, item['size'])
|
||||
rows, proof = projection(source, COLUMNS, progress=progress)
|
||||
else:
|
||||
rows, proof = cached[item['path']]
|
||||
if progress:
|
||||
progress({'phase': 'cached_metadata_projection', 'file': item['path'], 'rows': len(rows)})
|
||||
path = out/(Path(item['path']).stem+'.metadata.jsonl')
|
||||
part = path.with_suffix(path.suffix+'.part'); h = hashlib.sha256()
|
||||
with part.open('wb') as stream:
|
||||
for index, row in enumerate(rows):
|
||||
item_record = record(row, split, item, index)
|
||||
raw = (json.dumps(item_record, ensure_ascii=False)+'\n').encode()
|
||||
stream.write(raw); h.update(raw); records += 1
|
||||
counts[item_record['mirror_decision_category']] += 1
|
||||
track_claims[item_record['mirror_decision_track_claim'] or 'unspecified'] += 1
|
||||
raw_counts[json.dumps(row['decision'], ensure_ascii=False)] += 1
|
||||
years[(row['year'], row['track'])] += 1
|
||||
part.rename(path)
|
||||
reports.append({'source_file': item['path'], 'file': str(path), 'records': len(rows),
|
||||
'file_sha256': h.hexdigest(), 'source_lfs_sha256_claim': item['lfs']['oid'],
|
||||
'projection': proof})
|
||||
report = {'status': 'collected_mirror_metadata_pending_original_provenance_and_work_identity',
|
||||
'dataset': DATASET, 'commit': COMMIT, 'inventory_sha256': TREE_SHA256,
|
||||
'split': split, 'observed_at': observed, 'records': records, 'files': reports,
|
||||
'mirror_decision_counts': dict(counts), 'raw_mirror_decision_counts': dict(raw_counts),
|
||||
'mirror_decision_track_claim_counts': dict(track_claims),
|
||||
'year_track_counts': [{'year': y, 'track': t, 'rows': n} for (y, t), n in sorted(years.items(), key=lambda x: (x[0][0], str(x[0][1])))],
|
||||
'metadata_bytes_read': sum(x['projection']['metadata_bytes_read'] for x in reports) if cached is None else 0,
|
||||
'source_projection_bytes': sum(x['projection']['metadata_bytes_read'] for x in reports),
|
||||
'cached_projection_report_sha256': projection_sha256,
|
||||
'excluded_payload_pages_downloaded': False, 'raw_parquet_saved': False,
|
||||
'paper_body_saved': False, 'review_text_saved': False, 'database_written': False,
|
||||
'approved_for_import': 0,
|
||||
'limitations': ['Public mirror copies do not verify original forum IDs, visibility or signed decisions.',
|
||||
'Original source LFS SHA256 is a pinned claim, not a computed full-Parquet checksum.',
|
||||
'Every selected row and row locator is retained, including missing or anonymous author data.',
|
||||
'Decision categories describe only explicit mirror literals; venue labels and missing decisions do not imply rejection.',
|
||||
'Rows are not independent scientific works or a complete venue/year public negative census.',
|
||||
'Cross-source identity, chronological rejection/acceptance history and final dedup remain.']}
|
||||
(out/'collection-report.json').write_text(json.dumps(report, ensure_ascii=False, indent=2)+'\n')
|
||||
return report
|
||||
except BaseException as error:
|
||||
# No source-specific body or signed redirect URLs in failure reports.
|
||||
failure = {'status': 'incomplete_do_not_import', 'split': split, 'commit': COMMIT,
|
||||
'completed_files': reports, 'error_type': type(error).__name__,
|
||||
'database_written': False, 'approved_for_import': 0, 'no_automatic_retry': True}
|
||||
(out/'failure.json').write_text(json.dumps(failure, ensure_ascii=False, indent=2)+'\n')
|
||||
raise
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog)
|
||||
parser.add_argument('--split', required=True, choices=SPLITS)
|
||||
parser.add_argument('--out', required=True)
|
||||
parser.add_argument('--from-projection', help='Offline replay of a complete permitted metadata projection')
|
||||
parser.add_argument('--projection-sha256', help='Exact cached collection-report.json SHA256')
|
||||
args = parser.parse_args(argv)
|
||||
report = collect(args.split, args.out, from_projection=args.from_projection, projection_sha256=args.projection_sha256,
|
||||
progress=lambda row: print(json.dumps(row), flush=True))
|
||||
print(json.dumps({'status': report['status'], 'records': report['records'], 'out': args.out}), flush=True)
|
||||
@@ -1,16 +1,14 @@
|
||||
"""Reviewed postconference selected volumes from frozen public HTML metadata."""
|
||||
import argparse
|
||||
from collections import defaultdict
|
||||
import fcntl
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
import tempfile
|
||||
from urllib.parse import parse_qs, urlsplit
|
||||
|
||||
from paper_atlas.db.submissions import timestamp
|
||||
from paper_atlas.io.html import Document, clean
|
||||
from paper_atlas.io.springer import parse_chapter as parse_citation_chapter, read_page, snapshot_index
|
||||
|
||||
SOURCE = 'springer_selected_volume'
|
||||
BOOK = '10.1007/978-1-4612-2660-4'
|
||||
@@ -26,50 +24,6 @@ SCOPE = {
|
||||
}
|
||||
|
||||
|
||||
def snapshot_index(root):
|
||||
root = Path(root).resolve()
|
||||
receipts = json.loads((root/'receipts.json').read_bytes())['responses']
|
||||
index = {}
|
||||
for receipt in receipts:
|
||||
url = receipt['url']
|
||||
if url in index:
|
||||
raise ValueError('Duplicate snapshot URL')
|
||||
index[url] = (root, receipt)
|
||||
return index
|
||||
|
||||
|
||||
def read_page(index, url):
|
||||
if url not in index:
|
||||
raise ValueError('Missing required public HTML snapshot: '+url)
|
||||
root, receipt = index[url]
|
||||
filename = Path(receipt['file'])
|
||||
path = root/filename
|
||||
content_type = {k.lower():v for k,v in receipt.get('headers', {}).items()}.get('content-type', '')
|
||||
requested, final = urlsplit(url), urlsplit(receipt.get('final_url', ''))
|
||||
query = parse_qs(final.query, keep_blank_values=True)
|
||||
original_query = parse_qs(requested.query, keep_blank_values=True)
|
||||
# Springer returns the public metadata with an explicit cookie diagnostic.
|
||||
# Retain that final URL verbatim; accept no other redirect or changed page.
|
||||
if query != original_query and requested.netloc == 'link.springer.com':
|
||||
if (query.pop('error', None) != ['cookies_not_supported']
|
||||
or not re.fullmatch(r'[0-9a-f-]{36}', ''.join(query.pop('code', [])))):
|
||||
raise ValueError('Unreviewed Springer redirect query')
|
||||
if (filename.is_absolute() or '..' in filename.parts or not path.resolve().is_relative_to(root)
|
||||
or receipt.get('http_status') != 200
|
||||
or (final.scheme, final.netloc, final.path) != (requested.scheme, requested.netloc, requested.path)
|
||||
or final.fragment or query != original_query
|
||||
or 'text/html' not in content_type.lower()):
|
||||
raise ValueError('Unsafe, redirected or unsuccessful public HTML snapshot')
|
||||
raw = path.read_bytes()
|
||||
digest = hashlib.sha256(raw).hexdigest()
|
||||
if len(raw) > 2_000_000 or len(raw) != receipt['bytes'] or digest != receipt['sha256']:
|
||||
raise ValueError('Snapshot checksum or size differs')
|
||||
evidence = {'url':url, 'sha256':digest, 'bytes':len(raw),
|
||||
'file':'evidence/'+digest+'.html', 'http_status':200,
|
||||
'observed_at':timestamp(receipt['at']), 'response_receipt':receipt}
|
||||
return raw, evidence
|
||||
|
||||
|
||||
def parse_toc(html, *, excluded_materials=None):
|
||||
entries = {}
|
||||
for node in Document(html).root.walk():
|
||||
@@ -97,40 +51,8 @@ def parse_toc(html, *, excluded_materials=None):
|
||||
|
||||
|
||||
def parse_chapter(html, doi):
|
||||
nodes = list(Document(html).root.walk())
|
||||
meta = defaultdict(list)
|
||||
for node in nodes:
|
||||
if node.tag == 'meta' and node.attrs.get('name', '').startswith('citation_'):
|
||||
meta[node.attrs['name']].append(node.attrs.get('content', ''))
|
||||
def one(key):
|
||||
values = meta[key]
|
||||
if len(values) != 1 or not values[0]:
|
||||
raise ValueError('Missing or ambiguous citation field: '+key)
|
||||
return values[0]
|
||||
if (one('citation_doi') != doi or one('citation_inbook_title') != SCOPE['book_title']
|
||||
or one('citation_publication_date') != str(SCOPE['publication_year'])):
|
||||
raise ValueError('Chapter DOI, book or publication year differs from reviewed scope')
|
||||
raw_authors = meta['citation_author']
|
||||
if not raw_authors:
|
||||
raise ValueError('Missing named chapter authors')
|
||||
authors = []
|
||||
for raw in raw_authors:
|
||||
if raw.count(',') != 1:
|
||||
raise ValueError('Unreviewed citation author format')
|
||||
family, given = [clean(s) for s in raw.split(',', 1)]
|
||||
if not family or not given:
|
||||
raise ValueError('Incomplete citation author')
|
||||
authors.append({'given':given, 'family':family})
|
||||
abstract_nodes = [n for n in nodes if n.attrs.get('id') == 'Abs1-content']
|
||||
if len(abstract_nodes) > 1:
|
||||
raise ValueError('Ambiguous abstract section')
|
||||
abstract = clean(abstract_nodes[0].text()) if abstract_nodes else None
|
||||
first, last = one('citation_firstpage'), one('citation_lastpage')
|
||||
if not first.isdigit() or not last.isdigit() or int(first) > int(last):
|
||||
raise ValueError('Invalid chapter pages')
|
||||
return {'doi':doi, 'title':one('citation_title'), 'authors':authors,
|
||||
'authors_citation_raw':list(raw_authors), 'abstract':abstract or None,
|
||||
'first_page':first, 'last_page':last}
|
||||
return parse_citation_chapter(html, doi, book_title=SCOPE['book_title'],
|
||||
publication_year=SCOPE['publication_year'])
|
||||
|
||||
|
||||
def record_for(chapter, evidence, scope_evidence):
|
||||
|
||||
@@ -0,0 +1,128 @@
|
||||
"""Versioned publication-source scopes and guards against stale source refreshes."""
|
||||
import json
|
||||
import re
|
||||
|
||||
TABLE = 'source_scope_revisions'
|
||||
MARKER = 'source_scope_revisions_v1'
|
||||
COLUMNS = ('version_sha256', 'source_id', 'venue_id', 'year', 'event_key', 'list_url',
|
||||
'parent_sha256', 'records_json', 'plan_json', 'applied_at')
|
||||
|
||||
|
||||
def available(c):
|
||||
if getattr(c, 'postgres', False):
|
||||
return bool(c.execute("SELECT 1 FROM information_schema.tables WHERE table_schema=current_schema() AND table_name=?", (TABLE,)).fetchone())
|
||||
return bool(c.execute("SELECT 1 FROM sqlite_master WHERE type='table' AND name=?", (TABLE,)).fetchone())
|
||||
|
||||
|
||||
def initialize(c):
|
||||
exists = available(c)
|
||||
marker = bool(c.execute('SELECT 1 FROM schema_migrations WHERE name=?', (MARKER,)).fetchone())
|
||||
if exists != marker:
|
||||
raise ValueError('Source revision table/marker mismatch')
|
||||
if exists:
|
||||
result = c.execute('SELECT * FROM source_scope_revisions LIMIT 0')
|
||||
description = result.cursor.description if getattr(c, 'postgres', False) else result.description
|
||||
if tuple(x[0] for x in description) != COLUMNS:
|
||||
raise ValueError('Unexpected source revision schema')
|
||||
return
|
||||
ddl = '''CREATE TABLE source_scope_revisions (
|
||||
version_sha256 TEXT PRIMARY KEY,
|
||||
source_id INTEGER NOT NULL REFERENCES sources(id),
|
||||
venue_id INTEGER NOT NULL REFERENCES venues(id),
|
||||
year INTEGER NOT NULL, event_key TEXT NOT NULL, list_url TEXT NOT NULL,
|
||||
parent_sha256 TEXT UNIQUE REFERENCES source_scope_revisions(version_sha256),
|
||||
records_json TEXT NOT NULL, plan_json TEXT NOT NULL,
|
||||
applied_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP)'''
|
||||
if getattr(c, 'postgres', False):
|
||||
ddl = re.sub(r'\bINTEGER\b', 'BIGINT', ddl)
|
||||
ddl = re.sub(r'\bTEXT\b', 'TEXT COLLATE "C"', ddl)
|
||||
c.execute(ddl)
|
||||
c.execute('CREATE INDEX source_scope_revisions_scope ON source_scope_revisions(source_id,venue_id,year,event_key,list_url)')
|
||||
c.execute('INSERT INTO schema_migrations(name) VALUES(?)', (MARKER,))
|
||||
|
||||
|
||||
def heads(c, source_id):
|
||||
if not available(c):
|
||||
return []
|
||||
if not c.execute('SELECT 1 FROM schema_migrations WHERE name=?', (MARKER,)).fetchone():
|
||||
raise ValueError('Source revision table/marker mismatch')
|
||||
return c.execute('''SELECT v.version_sha256,v.venue_id,v.year,v.event_key,v.list_url
|
||||
FROM source_scope_revisions v WHERE v.source_id=? AND NOT EXISTS
|
||||
(SELECT 1 FROM source_scope_revisions child WHERE child.parent_sha256=v.version_sha256)
|
||||
ORDER BY v.version_sha256''', (source_id,)).fetchall()
|
||||
|
||||
|
||||
def current_rows(c, source_id, cache):
|
||||
active = heads(c, source_id)
|
||||
signatures = tuple(r[0] for r in active)
|
||||
# A new head invalidates the cache, even when maintenance uses the same Store.
|
||||
if cache.get('signatures') != (source_id, signatures):
|
||||
records = {}; scopes = {}
|
||||
for signature in signatures:
|
||||
raw = c.execute('SELECT records_json FROM source_scope_revisions WHERE version_sha256=?', (signature,)).fetchone()[0]
|
||||
scopes[signature] = json.loads(raw)
|
||||
for row in scopes[signature]:
|
||||
ext = row['external_id']
|
||||
if ext in records and records[ext] != row:
|
||||
raise ValueError('Conflicting active source snapshots')
|
||||
records[ext] = row
|
||||
cache.clear()
|
||||
cache.update(signatures=(source_id, signatures), records=records, scopes=scopes)
|
||||
return cache['records']
|
||||
|
||||
|
||||
def subset(expected, actual):
|
||||
"""Reviewed source fields must survive additional work-identity/curation evidence."""
|
||||
return isinstance(actual, dict) and all(k in actual and actual[k] == v for k, v in expected.items())
|
||||
|
||||
|
||||
def require_record(c, source_id, external, url, title, abstract, authors, identifiers, metadata, cache):
|
||||
expected = current_rows(c, source_id, cache).get(external)
|
||||
if expected is None:
|
||||
return
|
||||
actual = {'url': url, 'title': title, 'abstract': abstract, 'authors': authors}
|
||||
if any(actual[k] != expected.get(k) for k in actual):
|
||||
raise ValueError('Stale source snapshot: reconcile the source version before refresh')
|
||||
if not subset(expected.get('metadata', {}), metadata or {}) or not subset(expected.get('identifiers', {}), identifiers or {}):
|
||||
raise ValueError('Stale source snapshot metadata or identifiers')
|
||||
|
||||
|
||||
def scope_head(c, source_id, manifest):
|
||||
venue = c.execute('SELECT id FROM venues WHERE acronym=?', (manifest.get('venue'),)).fetchone()
|
||||
if not venue:
|
||||
return None
|
||||
matches = [r for r in heads(c, source_id)
|
||||
if [r[i] for i in range(1, 5)] == [venue[0], manifest.get('year'), manifest.get('event_key', ''), manifest.get('list_url')]]
|
||||
if len(matches) > 1:
|
||||
raise ValueError('Forked source revision history')
|
||||
return matches[0][0] if matches else None
|
||||
|
||||
|
||||
def require_manifest(c, source_id, manifest, cache):
|
||||
current_rows(c, source_id, cache)
|
||||
signature = scope_head(c, source_id, manifest)
|
||||
if signature is None:
|
||||
return
|
||||
expected = sum(r['track'] == manifest.get('track') for r in cache['scopes'][signature])
|
||||
bound = manifest.get('coverage_expected_count') if manifest.get('coverage_mode') == 'incremental' else manifest.get('count')
|
||||
if expected < 1 or bound != expected or manifest.get('status') != 'partial':
|
||||
raise ValueError('Source coverage manifest differs from current source snapshot')
|
||||
|
||||
|
||||
def require_row(c, source_id, row, manifest, cache):
|
||||
expected = current_rows(c, source_id, cache).get(row['external_id'])
|
||||
scoped_manifest = {**manifest, 'venue': row['venue'], 'year': row['year'], 'event_key': row.get('event_key', '')}
|
||||
signature = scope_head(c, source_id, scoped_manifest)
|
||||
if expected is None:
|
||||
if signature is not None:
|
||||
raise ValueError('Publication absent from current source snapshot; reconcile source membership first')
|
||||
return
|
||||
if signature is None or not any(r['external_id'] == row['external_id'] for r in cache['scopes'][signature]):
|
||||
raise ValueError('Source listing differs from current source snapshot')
|
||||
for key, value in expected.items():
|
||||
if key in ('metadata', 'identifiers'):
|
||||
okay = subset(value, row.get(key))
|
||||
else:
|
||||
okay = row.get(key) == value
|
||||
if not okay:
|
||||
raise ValueError('Stale source snapshot field: ' + key)
|
||||
@@ -96,6 +96,10 @@ class Store:
|
||||
if found:
|
||||
matches.add(found[0])
|
||||
old = c.execute('SELECT paper_id,title,abstract,retrieved_at,metadata_json FROM source_records WHERE source_id=? AND external_id=?',(source_id:=self.source(source,url),external_id)).fetchone()
|
||||
from paper_atlas.db.source_revisions import require_record
|
||||
if not hasattr(self, '_source_revision_cache'):
|
||||
self._source_revision_cache = {}
|
||||
require_record(c, source_id, external_id, url, title, abstract, authors, identifiers, metadata, self._source_revision_cache)
|
||||
from paper_atlas.db.edition_scopes import preserve
|
||||
metadata = preserve(dict(metadata or {}), old[4] if old else None)
|
||||
if old:
|
||||
|
||||
@@ -66,7 +66,7 @@ def require_incremental_parent(c, source, manifest, options):
|
||||
if entries:
|
||||
if len(entries) != 1 or entries[0][0] != manifest['list_url']:
|
||||
raise ValueError('Incremental coverage requires one unambiguous source listing')
|
||||
if tuple(entries[0][1:]) != ('partial', prior_expected, previous):
|
||||
if tuple(entries[0][i] for i in (1, 2, 3)) != ('partial', prior_expected, previous):
|
||||
raise ValueError('Incremental coverage parent changed or is not partial')
|
||||
elif previous != 0 or prior_expected is not None:
|
||||
raise ValueError('Incremental coverage parent is absent')
|
||||
@@ -101,6 +101,9 @@ def ingest_file(db,path, *, pg_service=None, pg_schema=None):
|
||||
c.commit()
|
||||
if incremental is not None:
|
||||
require_incremental_parent(c, source_id, manifest, incremental)
|
||||
from paper_atlas.db.source_revisions import require_manifest
|
||||
store._source_revision_cache = {}
|
||||
require_manifest(c, source_id, manifest, store._source_revision_cache)
|
||||
digest = hashlib.sha256()
|
||||
with path.open('rb') as stream:
|
||||
for line in stream:
|
||||
@@ -113,6 +116,10 @@ def ingest_file(db,path, *, pg_service=None, pg_schema=None):
|
||||
require_track_record(row, manifest)
|
||||
from paper_atlas.db.events import validate_event_key
|
||||
validate_event_key(row.get('event_key'))
|
||||
from paper_atlas.db.source_revisions import require_row
|
||||
if not hasattr(store, '_source_revision_cache'):
|
||||
store._source_revision_cache = {}
|
||||
require_row(c, source_id, row, manifest, store._source_revision_cache)
|
||||
external = row['external_id']
|
||||
if external in seen:
|
||||
raise ValueError('Duplicate external ID in one collected file: '+external)
|
||||
|
||||
@@ -0,0 +1,327 @@
|
||||
"""Plan-bound publication scope revisions with complete before-images, never reimport."""
|
||||
import argparse
|
||||
from collections import Counter
|
||||
from contextlib import closing
|
||||
import copy
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
|
||||
from paper_atlas.db import source_revisions as versions
|
||||
from paper_atlas.db.backend import database_arguments, open_read_connection, open_store
|
||||
from paper_atlas.db.edition_scopes import KEY
|
||||
from paper_atlas.ingestion.edition_tracks import curation
|
||||
from paper_atlas.ingestion.event_identity import database_identity, digest, values
|
||||
|
||||
MAX_ROWS = 10000
|
||||
|
||||
|
||||
def validate(inventory):
|
||||
required = {'format', 'source', 'venue', 'year', 'event_key', 'list_url',
|
||||
'previous_rows', 'rows', 'evidence'}
|
||||
if not isinstance(inventory, dict) or set(inventory) != required or inventory['format'] != 1:
|
||||
raise ValueError('Unsupported source revision inventory')
|
||||
if (type(inventory['year']) is not int or inventory['year'] < 2000
|
||||
or not all(isinstance(inventory[k], str) and inventory[k] for k in ('source', 'venue', 'list_url'))
|
||||
or not isinstance(inventory['event_key'], str)):
|
||||
raise ValueError('Explicit publication source scope required')
|
||||
if not inventory['list_url'].startswith('https://'):
|
||||
raise ValueError('HTTPS source listing required')
|
||||
evidence = inventory['evidence']
|
||||
if not isinstance(evidence, list) or not 1 <= len(evidence) <= 100:
|
||||
raise ValueError('Bounded source evidence required')
|
||||
for item in evidence:
|
||||
if (not isinstance(item, dict) or set(item) != {'url', 'sha256'}
|
||||
or not isinstance(item['url'], str) or not item['url'].startswith('https://')
|
||||
or not isinstance(item['sha256'], str) or not re.fullmatch('[0-9a-f]{64}', item['sha256'])):
|
||||
raise ValueError('Source evidence URL/hash required')
|
||||
snapshots = []
|
||||
for key in ('previous_rows', 'rows'):
|
||||
items = inventory[key]
|
||||
if not isinstance(items, list) or not 1 <= len(items) <= MAX_ROWS:
|
||||
raise ValueError('Complete bounded source snapshots required')
|
||||
records = {}
|
||||
for row in items:
|
||||
if not isinstance(row, dict) or any(row.get(k) != inventory[k] for k in ('source', 'venue', 'year')):
|
||||
raise ValueError('Source snapshot scope mismatch')
|
||||
ext = row.get('external_id')
|
||||
if (not isinstance(ext, str) or ext in records or row.get('status') != 'accepted'
|
||||
or not re.fullmatch(r'10\.\d{4,9}/\S+', ext)
|
||||
or row.get('identifiers', {}).get('doi') != ext
|
||||
or not isinstance(row.get('track'), str) or not row['track']
|
||||
or row.get('event_key', '') != inventory['event_key']
|
||||
or not isinstance(row.get('title'), str) or not row['title']
|
||||
or not isinstance(row.get('authors'), list)
|
||||
or not isinstance(row.get('metadata'), dict)):
|
||||
raise ValueError('Unique DOI-bearing accepted publication records required')
|
||||
records[ext] = row
|
||||
snapshots.append(records)
|
||||
previous, current = snapshots
|
||||
if set(previous) != set(current):
|
||||
raise ValueError('Source membership changed; this operation cannot add or remove publications')
|
||||
for ext, row in current.items():
|
||||
before = previous[ext]
|
||||
stable = (set(before) | set(row)) - {'abstract', 'track', 'tags', 'metadata'}
|
||||
if any(before.get(k) != row.get(k) for k in stable):
|
||||
raise ValueError('Stable publication identity or decision changed')
|
||||
return previous, current
|
||||
|
||||
|
||||
def bounded(c, sql, params):
|
||||
result = values(c.execute(sql + ' LIMIT 100001', params))
|
||||
if len(result) > 100000:
|
||||
raise ValueError('Source revision protection exceeds bounded scope')
|
||||
return result
|
||||
|
||||
|
||||
def capture(c, inventory):
|
||||
previous, current = validate(inventory)
|
||||
source = c.execute('SELECT id FROM sources WHERE name=?', (inventory['source'],)).fetchone()
|
||||
venue = c.execute('SELECT id FROM venues WHERE acronym=?', (inventory['venue'],)).fetchone()
|
||||
if not source or not venue:
|
||||
raise ValueError('Existing source and venue required')
|
||||
sid, vid = source[0], venue[0]
|
||||
editions = bounded(c, 'SELECT * FROM editions WHERE venue_id=? AND year=? AND event_key=? ORDER BY id',
|
||||
(vid, inventory['year'], inventory['event_key']))
|
||||
by_track = {r[3]: r[0] for r in editions}
|
||||
# The reviewed scope includes all its existing target attempts, even when
|
||||
# they originate from another source. No empty-target shortcut is used.
|
||||
scope_subs = 'SELECT s.id FROM submissions s JOIN editions e ON e.id=s.edition_id WHERE e.venue_id=? AND e.year=? AND e.event_key=?'
|
||||
scope_params = (vid, inventory['year'], inventory['event_key'])
|
||||
scoped = bounded(c, 'SELECT * FROM submissions WHERE id IN (' + scope_subs + ') ORDER BY id', scope_params)
|
||||
own_records = bounded(c, '''SELECT DISTINCT r.* FROM source_records r JOIN submission_sources ss ON ss.source_record_id=r.id
|
||||
WHERE r.source_id=? AND ss.submission_id IN (''' + scope_subs + ') ORDER BY r.id', (sid, *scope_params))
|
||||
if not own_records:
|
||||
raise ValueError('Existing source observations required')
|
||||
if any(r[2] not in current for r in own_records):
|
||||
raise ValueError('Snapshot omits an existing source observation')
|
||||
pids = sorted({r[1] for r in scoped})
|
||||
if not pids or len(pids) > MAX_ROWS:
|
||||
raise ValueError('Bounded nonempty protected paper scope required')
|
||||
marks = ','.join('?' for _ in pids)
|
||||
papers = 'SELECT id FROM papers WHERE id IN (' + marks + ')'
|
||||
subs = 'SELECT id FROM submissions WHERE paper_id IN (' + papers + ')'
|
||||
decisions = 'SELECT id FROM decisions WHERE submission_id IN (' + subs + ')'
|
||||
queries = {
|
||||
'papers': 'SELECT * FROM papers WHERE id IN (' + papers + ') ORDER BY id',
|
||||
'identifiers': 'SELECT * FROM identifiers WHERE paper_id IN (' + papers + ') ORDER BY scheme,value',
|
||||
'source_records': 'SELECT * FROM source_records WHERE paper_id IN (' + papers + ') ORDER BY id',
|
||||
'submissions': 'SELECT * FROM submissions WHERE paper_id IN (' + papers + ') ORDER BY id',
|
||||
'decisions': 'SELECT * FROM decisions WHERE submission_id IN (' + subs + ') ORDER BY id',
|
||||
'submission_sources': 'SELECT * FROM submission_sources WHERE submission_id IN (' + subs + ') ORDER BY submission_id,source_record_id,source_submission_id',
|
||||
'submission_identifiers': 'SELECT * FROM submission_identifiers WHERE submission_id IN (' + subs + ') ORDER BY edition_id,scheme,value',
|
||||
'decision_sources': 'SELECT * FROM decision_sources WHERE decision_id IN (' + decisions + ') ORDER BY decision_id,source_record_id,source_decision_id',
|
||||
'paper_tags': 'SELECT * FROM paper_tags WHERE paper_id IN (' + papers + ') ORDER BY paper_id,tag_id,evidence_key',
|
||||
}
|
||||
from paper_atlas.curation.annotations import available
|
||||
if available(c):
|
||||
queries['annotation_events'] = 'SELECT * FROM annotation_events WHERE paper_id IN (' + papers + ') ORDER BY id'
|
||||
packet = {k: bounded(c, q, pids) for k, q in queries.items()}
|
||||
coverage = bounded(c, '''SELECT es.* FROM edition_sources es JOIN editions e ON e.id=es.edition_id
|
||||
WHERE es.source_id=? AND e.venue_id=? AND e.year=? AND e.event_key=? ORDER BY es.edition_id,es.list_url''', (sid, *scope_params))
|
||||
if not coverage or any(r[2] != inventory['list_url'] for r in coverage):
|
||||
raise ValueError('Unambiguous source listing required')
|
||||
if len({r[0] for r in coverage}) != len(coverage):
|
||||
raise ValueError('One source coverage entry per edition required')
|
||||
heads = [[r[i] for i in range(len(r))] for r in versions.heads(c, sid)]
|
||||
own_heads = [r for r in heads if r[1:] == [vid, inventory['year'], inventory['event_key'], inventory['list_url']]]
|
||||
if len(own_heads) > 1:
|
||||
raise ValueError('Forked source revision history')
|
||||
parent = own_heads[0][0] if own_heads else None
|
||||
return {'source_id': sid, 'venue_id': vid, 'editions': editions, 'coverage': coverage,
|
||||
'packet': packet, 'own_records': own_records, 'revision_heads': heads,
|
||||
'parent_sha256': parent, 'by_track': by_track,
|
||||
'venue_names': sorted({inventory['venue']} | {r[0] for r in c.execute('SELECT alias FROM venue_aliases WHERE venue_id=?', (vid,))})}
|
||||
|
||||
|
||||
def changes(c, inventory, before):
|
||||
previous, current = validate(inventory)
|
||||
packet = before['packet']; result = []; sid = before['source_id']
|
||||
editions = {r[0]: r for r in before['editions']}
|
||||
for raw in before['own_records']:
|
||||
ext = raw[2]; old = previous[ext]; new = current[ext]
|
||||
metadata = json.loads(raw[8])
|
||||
candidates = [old, new]
|
||||
if not any(raw[4] == r['url'] and raw[5] == r['title'] and raw[6] == r.get('abstract')
|
||||
and json.loads(raw[7]) == r['authors'] and versions.subset(r['metadata'], metadata) for r in candidates):
|
||||
raise ValueError('Current raw source differs from both reviewed snapshots')
|
||||
if metadata.get('record_identifiers', {}).get('doi') != ext:
|
||||
raise ValueError('Current stable DOI differs from source evidence')
|
||||
attempts = [s for s in packet['submissions'] if s[3] == raw[0] and s[2] in editions]
|
||||
if len(attempts) != 1:
|
||||
raise ValueError('Exactly one publication attempt per source record required')
|
||||
sub = attempts[0]; origin = editions[sub[2]]
|
||||
if origin[3] not in {old['track'], new['track']} or sub[4] != ext or sub[1] != raw[3]:
|
||||
raise ValueError('Current source attempt identity/scope differs')
|
||||
if [x for x in packet['submission_sources'] if x[0] == sub[0]] != [[sub[0], raw[0], ext]]:
|
||||
raise ValueError('Shared-source attempt requires separate reconciliation')
|
||||
decisions = [d for d in packet['decisions'] if d[1] == sub[0]]
|
||||
if not decisions or any(d[2] != 'accepted' for d in decisions):
|
||||
raise ValueError('Only accepted publication observations can be revised')
|
||||
target = before['by_track'].get(new['track'])
|
||||
if target is None:
|
||||
raise ValueError('Reviewed target edition must already exist')
|
||||
identities = [x for x in packet['submission_identifiers'] if x[3] == sub[0]]
|
||||
if not identities or any(x[0] != sub[2] for x in identities):
|
||||
raise ValueError('Inconsistent scoped submission identifiers')
|
||||
for _, scheme, value, _ in identities:
|
||||
conflict = c.execute('SELECT submission_id FROM submission_identifiers WHERE edition_id=? AND scheme=? AND value=?', (target, scheme, value)).fetchone()
|
||||
if conflict and conflict[0] != sub[0]:
|
||||
raise ValueError('Target submission identifier collision')
|
||||
replacement = copy.deepcopy(metadata); replacement.update(new['metadata'])
|
||||
if sub[2] != target and KEY in metadata:
|
||||
raise ValueError('An already curated attempt needs explicit scope-chain reconciliation')
|
||||
if sub[2] != target or raw[6] != new.get('abstract') or replacement != metadata:
|
||||
result.append({'source_record_id': raw[0], 'paper_id': raw[3], 'submission_id': sub[0],
|
||||
'external_id': ext, 'from_edition_id': sub[2], 'target_edition_id': target,
|
||||
'from_track': origin[3], 'track': new['track'],
|
||||
'new_abstract': new.get('abstract'), 'metadata': replacement})
|
||||
old_expected = Counter(r['track'] for r in previous.values()); new_expected = Counter(r['track'] for r in current.values())
|
||||
observed = Counter()
|
||||
own_ids = {r[0] for r in before['own_records']}
|
||||
for s in packet['submissions']:
|
||||
if s[3] in own_ids and s[2] in editions:
|
||||
observed[s[2]] += 1
|
||||
coverage = []
|
||||
for raw in before['coverage']:
|
||||
track = editions[raw[0]][3]
|
||||
if raw[3] != 'partial' or raw[8] is not None or raw[5] != observed[raw[0]]:
|
||||
raise ValueError('Coverage parent is not the verified partial source observation count')
|
||||
if raw[4] not in {old_expected[track], new_expected[track]}:
|
||||
raise ValueError('Coverage expected count differs from both source versions')
|
||||
count = raw[5] + sum((r['target_edition_id'] == raw[0]) - (r['from_edition_id'] == raw[0]) for r in result)
|
||||
if count < 0 or count > new_expected[track]:
|
||||
raise ValueError('Corrected coverage exceeds current source membership')
|
||||
coverage.append({'before': raw, 'expected_count': new_expected[track], 'imported_count': count})
|
||||
covered = {r['before'][0] for r in coverage}
|
||||
if any(r['target_edition_id'] not in covered for r in result):
|
||||
raise ValueError('Target source coverage must already exist')
|
||||
return result, coverage
|
||||
|
||||
|
||||
def proposal(c, inventory, identity):
|
||||
before = capture(c, inventory)
|
||||
if before['parent_sha256']:
|
||||
saved = json.loads(c.execute('SELECT records_json FROM source_scope_revisions WHERE version_sha256=?', (before['parent_sha256'],)).fetchone()[0])
|
||||
if sorted(saved, key=lambda r: r['external_id']) != sorted(inventory['previous_rows'], key=lambda r: r['external_id']):
|
||||
raise ValueError('Previous snapshot differs from active source version')
|
||||
records, coverage = changes(c, inventory, before)
|
||||
return {'format': 1, 'database': identity, 'inventory': inventory, 'before': before,
|
||||
'changes': records, 'coverage': coverage}
|
||||
|
||||
|
||||
def verify(c, plan, signature):
|
||||
after = capture(c, plan['inventory']); expected = copy.deepcopy(plan['before']['packet'])
|
||||
for change in plan['changes']:
|
||||
for row in expected['submissions']:
|
||||
if row[0] == change['submission_id']:
|
||||
row[2] = change['target_edition_id']
|
||||
for row in expected['submission_identifiers']:
|
||||
if row[3] == change['submission_id']:
|
||||
row[0] = change['target_edition_id']
|
||||
for row in expected['source_records']:
|
||||
if row[0] == change['source_record_id']:
|
||||
row[6] = change['new_abstract']
|
||||
row[8] = encoded_metadata(plan, change, signature)
|
||||
expected['submission_identifiers'].sort(key=lambda r: (r[0], r[1], r[2]))
|
||||
if after['packet'] != expected:
|
||||
raise ValueError('Protected profiles/history/tags/identifiers differ; roll back')
|
||||
if after['editions'] != plan['before']['editions']:
|
||||
raise ValueError('Edition identities changed')
|
||||
expected_heads = [r for r in plan['before']['revision_heads'] if r[0] != plan['before']['parent_sha256']]
|
||||
inventory = plan['inventory']
|
||||
expected_heads.append([signature, plan['before']['venue_id'], inventory['year'], inventory['event_key'], inventory['list_url']])
|
||||
if after['revision_heads'] != sorted(expected_heads):
|
||||
raise ValueError('Active source version differs')
|
||||
expected_coverage = []
|
||||
for item in plan['coverage']:
|
||||
row = item['before'][:]; row[4] = item['expected_count']; row[5] = item['imported_count']
|
||||
expected_coverage.append(row)
|
||||
if after['coverage'] != expected_coverage:
|
||||
raise ValueError('Reconciled coverage differs; roll back')
|
||||
stored = c.execute('SELECT records_json,plan_json,parent_sha256 FROM source_scope_revisions WHERE version_sha256=?', (signature,)).fetchone()
|
||||
if (not stored or json.loads(stored[0]) != plan['inventory']['rows']
|
||||
or json.loads(stored[1]) != plan or stored[2] != plan['before']['parent_sha256']):
|
||||
raise ValueError('Versioned source or complete before-images changed')
|
||||
return {'records_changed': len(plan['changes']), 'events_moved': sum(r['from_edition_id'] != r['target_edition_id'] for r in plan['changes']),
|
||||
'protected_papers': len(expected['papers']), 'protected_decisions': len(expected['decisions'])}
|
||||
|
||||
|
||||
def encoded_metadata(plan, change, signature):
|
||||
metadata = copy.deepcopy(change['metadata'])
|
||||
if change['from_edition_id'] != change['target_edition_id']:
|
||||
inventory = plan['inventory']
|
||||
evidence = inventory['evidence'][0]
|
||||
row = {**change, 'source': inventory['source'], 'doi': change['external_id'],
|
||||
'venue_id': plan['before']['venue_id'], 'year': inventory['year'],
|
||||
'event_key': inventory['event_key'], 'venue_names': plan['before']['venue_names'],
|
||||
'section': change['track'], 'evidence_url': evidence['url'], 'evidence_sha256': evidence['sha256']}
|
||||
metadata[KEY] = [curation(row, change['target_edition_id'], signature)]
|
||||
return json.dumps(metadata, ensure_ascii=False)
|
||||
|
||||
|
||||
def apply_plan(c, plan):
|
||||
if plan.get('format') != 1:
|
||||
raise ValueError('Unsupported source revision plan')
|
||||
if not getattr(c, 'postgres', False) and not c.in_transaction:
|
||||
c.execute('BEGIN IMMEDIATE')
|
||||
signature = digest(plan)
|
||||
if versions.available(c) and c.execute('SELECT 1 FROM source_scope_revisions WHERE version_sha256=?', (signature,)).fetchone():
|
||||
return {'status': 'already_applied', 'version_sha256': signature, **verify(c, plan, signature)}
|
||||
if getattr(c, 'postgres', False):
|
||||
c.execute('LOCK TABLE sources,papers,identifiers,source_records,venues,venue_aliases,editions,edition_sources,submissions,decisions,submission_sources,submission_identifiers,decision_sources,tags,paper_tags IN SHARE ROW EXCLUSIVE MODE')
|
||||
if proposal(c, plan['inventory'], plan['database']) != plan:
|
||||
raise ValueError('Stale source revision plan; create and review a fresh plan')
|
||||
versions.initialize(c)
|
||||
for row in plan['changes']:
|
||||
if row['from_edition_id'] != row['target_edition_id']:
|
||||
c.execute('UPDATE submission_identifiers SET edition_id=? WHERE submission_id=?', (row['target_edition_id'], row['submission_id']))
|
||||
c.execute('UPDATE submissions SET edition_id=? WHERE id=? AND edition_id=?', (row['target_edition_id'], row['submission_id'], row['from_edition_id']))
|
||||
c.execute('UPDATE source_records SET abstract=?,metadata_json=? WHERE id=?',
|
||||
(row['new_abstract'], encoded_metadata(plan, row, signature), row['source_record_id']))
|
||||
for row in plan['coverage']:
|
||||
before = row['before']
|
||||
c.execute('UPDATE edition_sources SET expected_count=?,imported_count=? WHERE edition_id=? AND source_id=? AND list_url=?',
|
||||
(row['expected_count'], row['imported_count'], before[0], before[1], before[2]))
|
||||
inventory = plan['inventory']
|
||||
c.execute('''INSERT INTO source_scope_revisions(version_sha256,source_id,venue_id,year,event_key,list_url,parent_sha256,records_json,plan_json)
|
||||
VALUES(?,?,?,?,?,?,?,?,?)''', (signature, plan['before']['source_id'], plan['before']['venue_id'], inventory['year'], inventory['event_key'],
|
||||
inventory['list_url'], plan['before']['parent_sha256'], json.dumps(inventory['rows'], ensure_ascii=False), json.dumps(plan, ensure_ascii=False)))
|
||||
return {'status': 'applied', 'version_sha256': signature, **verify(c, plan, signature)}
|
||||
|
||||
|
||||
def run(db, out, *, pg_service=None, pg_schema=None, inventory=None, plan=None, plan_sha256=None):
|
||||
out = Path(out)
|
||||
if out.exists() or not out.parent.is_dir():
|
||||
raise ValueError('New output in existing directory required')
|
||||
if bool(inventory) == bool(plan) or bool(plan) != bool(plan_sha256):
|
||||
raise ValueError('Choose inventory preflight or plan plus exact file hash')
|
||||
if inventory:
|
||||
with closing(open_read_connection(db, pg_service=pg_service, pg_schema=pg_schema)) as c:
|
||||
result = proposal(c, json.loads(Path(inventory).read_bytes()), database_identity(c, db))
|
||||
else:
|
||||
raw = Path(plan).read_bytes()
|
||||
if hashlib.sha256(raw).hexdigest() != plan_sha256:
|
||||
raise ValueError('Plan file hash differs')
|
||||
proposed = json.loads(raw)
|
||||
with closing(open_store(db, pg_service=pg_service, pg_schema=pg_schema).conn) as c:
|
||||
try:
|
||||
if proposed['database'] != database_identity(c, db):
|
||||
raise ValueError('Plan belongs to another database')
|
||||
if not getattr(c, 'postgres', False):
|
||||
c.commit(); c.execute('BEGIN IMMEDIATE')
|
||||
result = apply_plan(c, proposed); c.commit()
|
||||
except BaseException:
|
||||
c.rollback(); raise
|
||||
out.write_text(json.dumps(result, ensure_ascii=False, indent=2) + '\n')
|
||||
return result
|
||||
|
||||
|
||||
def main(argv=None, *, prog=None):
|
||||
parser = argparse.ArgumentParser(prog=prog); database_arguments(parser)
|
||||
parser.add_argument('--inventory'); parser.add_argument('--plan'); parser.add_argument('--plan-sha256'); parser.add_argument('--out', required=True)
|
||||
args = parser.parse_args(argv)
|
||||
report = run(args.db, args.out, pg_service=args.pg_service, pg_schema=args.pg_schema,
|
||||
inventory=args.inventory, plan=args.plan, plan_sha256=args.plan_sha256)
|
||||
print(json.dumps({'status': report.get('status', 'planned'), 'out': args.out}), flush=True)
|
||||
@@ -0,0 +1,195 @@
|
||||
"""Read only selected Parquet column chunks; never fetch other payload pages."""
|
||||
import hashlib
|
||||
import io
|
||||
import re
|
||||
import struct
|
||||
import urllib.error
|
||||
import urllib.request
|
||||
|
||||
MAX_FOOTER = 4 * 1024**2
|
||||
MAX_SELECTED_BYTES = 64 * 1024**2
|
||||
MAX_REQUESTS = 512
|
||||
|
||||
|
||||
class RangeSource:
|
||||
"""Strict bounded HTTP ranges, no full-file fallback, redirects kept private."""
|
||||
def __init__(self, url, size, *, opener=None):
|
||||
self.url, self.size = url, size
|
||||
self.opener = opener or urllib.request.build_opener()
|
||||
self.etag = None
|
||||
self.receipts = []
|
||||
self.bytes_read = 0
|
||||
|
||||
def read(self, start, length):
|
||||
if (type(start) is not int or type(length) is not int or length <= 0
|
||||
or start < 0 or start + length > self.size
|
||||
or self.bytes_read + length > MAX_SELECTED_BYTES
|
||||
or len(self.receipts) >= MAX_REQUESTS):
|
||||
raise ValueError('Bounded metadata range budget exceeded')
|
||||
headers = {'Range': f'bytes={start}-{start + length - 1}', 'Accept-Encoding': 'identity'}
|
||||
if self.etag and not self.etag.startswith('W/'):
|
||||
headers['If-Match'] = self.etag
|
||||
request = urllib.request.Request(self.url, headers=headers)
|
||||
try:
|
||||
with self.opener.open(request, timeout=45) as response:
|
||||
if response.status != 206:
|
||||
raise ValueError('Server did not honor metadata range; full download forbidden')
|
||||
if response.headers.get('Content-Range') != f'bytes {start}-{start + length - 1}/{self.size}':
|
||||
raise ValueError('Object size or range differs from pinned inventory')
|
||||
encoding = response.headers.get('Content-Encoding', 'identity')
|
||||
if encoding != 'identity':
|
||||
raise ValueError('Encoded HTTP ranges are unsupported')
|
||||
etag = response.headers.get('ETag')
|
||||
if self.receipts and etag != self.etag:
|
||||
raise ValueError('Object ETag changed between metadata ranges')
|
||||
raw = response.read(length + 1)
|
||||
if len(raw) != length:
|
||||
raise ValueError('Truncated or oversized metadata range')
|
||||
if not self.receipts:
|
||||
self.etag = etag
|
||||
except urllib.error.HTTPError as error:
|
||||
# Do not expose signed CDN redirect URLs or retry denied resources.
|
||||
raise RuntimeError('Metadata source HTTP ' + str(error.code) + '; no retry or full-file fallback') from None
|
||||
self.receipts.append({'start': start, 'length': length, 'sha256': hashlib.sha256(raw).hexdigest()})
|
||||
self.bytes_read += length
|
||||
return raw
|
||||
|
||||
|
||||
class SparseFile(io.RawIOBase):
|
||||
"""A logical file backed only by approved bytes; all other positions are zeros.
|
||||
|
||||
Arrow may read a 64 KiB footer window. Synthesizing its unused prefix avoids
|
||||
fetching the preceding OCR/review page. Data pages are explicitly selected
|
||||
from footer offsets before they can be downloaded.
|
||||
"""
|
||||
def __init__(self, size, segments):
|
||||
self.size, self.segments, self.position = size, segments, 0
|
||||
|
||||
def readable(self):
|
||||
return True
|
||||
|
||||
def seekable(self):
|
||||
return True
|
||||
|
||||
def tell(self):
|
||||
return self.position
|
||||
|
||||
def seek(self, offset, whence=0):
|
||||
base = {0: 0, 1: self.position, 2: self.size}.get(whence)
|
||||
if base is None or base + offset < 0:
|
||||
raise ValueError('Invalid sparse metadata seek')
|
||||
self.position = base + offset
|
||||
return self.position
|
||||
|
||||
def read(self, length=-1):
|
||||
if length < 0:
|
||||
length = self.size - self.position
|
||||
length = min(length, max(0, self.size - self.position))
|
||||
if length > MAX_SELECTED_BYTES:
|
||||
raise ValueError('Unbounded sparse read forbidden')
|
||||
start, end = self.position, self.position + length
|
||||
raw = bytearray(length)
|
||||
for offset, data in self.segments:
|
||||
left, right = max(offset, start), min(offset + len(data), end)
|
||||
if left < right:
|
||||
raw[left-start:right-start] = data[left-offset:right-offset]
|
||||
self.position = end
|
||||
return bytes(raw)
|
||||
|
||||
def readinto(self, buffer):
|
||||
raw = self.read(len(buffer)); buffer[:len(raw)] = raw
|
||||
return len(raw)
|
||||
|
||||
|
||||
def arrow():
|
||||
try:
|
||||
import pyarrow as pa
|
||||
import pyarrow.parquet as pq
|
||||
except ImportError:
|
||||
raise RuntimeError('Install paper-atlas[mirrors] only on the approved remote runner') from None
|
||||
return pa, pq
|
||||
|
||||
|
||||
def projection(source, columns, *, progress=None):
|
||||
"""Return complete metadata rows and range receipts, no full Parquet snapshot."""
|
||||
pa, pq = arrow()
|
||||
if type(source.size) is not int or not 12 <= source.size <= 4 * 1024**3:
|
||||
raise ValueError('Bounded pinned Parquet size required')
|
||||
if (not isinstance(columns, dict) or not columns
|
||||
or any(not re.fullmatch('[a-z][a-z0-9_]*', key) for key in columns)):
|
||||
raise ValueError('Explicit top-level metadata columns required')
|
||||
header = source.read(0, 4); trailer = source.read(source.size - 8, 8)
|
||||
if header != b'PAR1' or trailer[4:] != b'PAR1':
|
||||
raise ValueError('Expected unencrypted Parquet framing')
|
||||
length = struct.unpack('<I', trailer[:4])[0]
|
||||
if not 0 < length <= min(MAX_FOOTER, source.size - 12):
|
||||
raise ValueError('Invalid or oversized Parquet footer')
|
||||
footer_start = source.size - 8 - length
|
||||
footer = source.read(footer_start, length)
|
||||
segments = [(0, header), (footer_start, footer), (source.size - 8, trailer)]
|
||||
sparse = SparseFile(source.size, segments)
|
||||
parquet = pq.ParquetFile(sparse, pre_buffer=False, buffer_size=0)
|
||||
try:
|
||||
schema = parquet.schema_arrow
|
||||
if len(set(schema.names)) != len(schema.names) or not set(columns) <= set(schema.names):
|
||||
raise ValueError('Missing or repeated metadata columns')
|
||||
for key, kind in columns.items():
|
||||
value = schema.field(key).type
|
||||
if kind == 'string':
|
||||
okay = pa.types.is_string(value) or pa.types.is_large_string(value)
|
||||
elif kind == 'integer':
|
||||
okay = pa.types.is_integer(value)
|
||||
elif kind == 'strings':
|
||||
okay = (pa.types.is_list(value) or pa.types.is_large_list(value)) and (
|
||||
pa.types.is_string(value.value_type) or pa.types.is_large_string(value.value_type))
|
||||
else:
|
||||
raise ValueError('Unsupported safe metadata type')
|
||||
if not okay:
|
||||
raise ValueError('Metadata column type differs: ' + key)
|
||||
metadata = parquet.metadata
|
||||
if not 0 < metadata.num_rows <= 100000 or metadata.num_row_groups > 1000:
|
||||
raise ValueError('Bounded metadata row count required')
|
||||
wanted, excluded = [], []
|
||||
for group in range(metadata.num_row_groups):
|
||||
rg = metadata.row_group(group)
|
||||
for i in range(rg.num_columns):
|
||||
col = rg.column(i)
|
||||
if col.file_path:
|
||||
raise ValueError('External Parquet column files forbidden')
|
||||
offsets = [x for x in (col.dictionary_page_offset, col.data_page_offset) if x is not None and x > 0]
|
||||
if not offsets:
|
||||
raise ValueError('Missing column chunk offset')
|
||||
start = min(offsets); end = start + col.total_compressed_size
|
||||
if not 4 <= start < end <= footer_start:
|
||||
raise ValueError('Column chunk lies outside the pinned payload')
|
||||
item = (start, end, group, col.path_in_schema)
|
||||
(wanted if col.path_in_schema.split('.')[0] in columns else excluded).append(item)
|
||||
if any(max(a, c) < min(b, d) for a, b, _, _ in wanted for c, d, _, _ in excluded):
|
||||
raise ValueError('Selected range overlaps an excluded payload column')
|
||||
if sum(end-start for start, end, _, _ in wanted) + length + 12 > MAX_SELECTED_BYTES:
|
||||
raise ValueError('Metadata projection exceeds byte budget')
|
||||
if len(wanted) + 3 > MAX_REQUESTS:
|
||||
raise ValueError('Metadata projection exceeds request budget')
|
||||
for start, end, group, path in sorted(wanted):
|
||||
segments.append((start, source.read(start, end-start)))
|
||||
if progress:
|
||||
progress({'phase': 'metadata_column_range', 'row_group': group,
|
||||
'column': path, 'metadata_bytes_read': source.bytes_read})
|
||||
rows = []
|
||||
for group in range(metadata.num_row_groups):
|
||||
table = parquet.read_row_group(group, columns=list(columns), use_threads=False)
|
||||
if set(table.column_names) != set(columns) or len(table.column_names) != len(columns):
|
||||
raise ValueError('Unexpected projected metadata fields')
|
||||
rows.extend(table.select(list(columns)).to_pylist())
|
||||
if len(rows) != metadata.num_rows:
|
||||
raise ValueError('Incomplete metadata projection')
|
||||
report = {'rows': len(rows), 'row_groups': metadata.num_row_groups,
|
||||
'columns': dict(columns), 'footer_sha256': hashlib.sha256(footer).hexdigest(),
|
||||
'column_ranges': [{'start': a, 'length': b-a, 'row_group': g, 'column': p} for a, b, g, p in wanted],
|
||||
'range_receipts': source.receipts, 'metadata_bytes_read': source.bytes_read,
|
||||
'source_bytes': source.size, 'full_file_checksum_verified': False,
|
||||
'excluded_payload_pages_downloaded': False,
|
||||
'footer_saved': False, 'raw_parquet_saved': False}
|
||||
return rows, report
|
||||
finally:
|
||||
parquet.close()
|
||||
@@ -0,0 +1,91 @@
|
||||
"""Frozen Springer public HTML receipts and chapter citation metadata."""
|
||||
from collections import defaultdict
|
||||
import hashlib
|
||||
from pathlib import Path
|
||||
import json
|
||||
import re
|
||||
from urllib.parse import parse_qs, urlsplit
|
||||
|
||||
from paper_atlas.db.submissions import timestamp
|
||||
from paper_atlas.io.html import Document, clean
|
||||
|
||||
|
||||
def snapshot_index(root):
|
||||
root = Path(root).resolve()
|
||||
receipts = json.loads((root/'receipts.json').read_bytes())['responses']
|
||||
index = {}
|
||||
for receipt in receipts:
|
||||
url = receipt['url']
|
||||
if url in index:
|
||||
raise ValueError('Duplicate snapshot URL')
|
||||
index[url] = (root, receipt)
|
||||
return index
|
||||
|
||||
|
||||
def read_page(index, url):
|
||||
if url not in index:
|
||||
raise ValueError('Missing required public HTML snapshot: '+url)
|
||||
root, receipt = index[url]
|
||||
filename = Path(receipt['file'])
|
||||
path = root/filename
|
||||
content_type = {k.lower():v for k,v in receipt.get('headers', {}).items()}.get('content-type', '')
|
||||
requested, final = urlsplit(url), urlsplit(receipt.get('final_url', ''))
|
||||
query = parse_qs(final.query, keep_blank_values=True)
|
||||
original_query = parse_qs(requested.query, keep_blank_values=True)
|
||||
# Springer returns the public metadata with an explicit cookie diagnostic.
|
||||
# Retain that final URL verbatim; accept no other redirect or changed page.
|
||||
if query != original_query and requested.netloc == 'link.springer.com':
|
||||
if (query.pop('error', None) != ['cookies_not_supported']
|
||||
or not re.fullmatch(r'[0-9a-f-]{36}', ''.join(query.pop('code', [])))):
|
||||
raise ValueError('Unreviewed Springer redirect query')
|
||||
if (filename.is_absolute() or '..' in filename.parts or not path.resolve().is_relative_to(root)
|
||||
or receipt.get('http_status') != 200
|
||||
or (final.scheme, final.netloc, final.path) != (requested.scheme, requested.netloc, requested.path)
|
||||
or final.fragment or query != original_query
|
||||
or 'text/html' not in content_type.lower()):
|
||||
raise ValueError('Unsafe, redirected or unsuccessful public HTML snapshot')
|
||||
raw = path.read_bytes()
|
||||
digest = hashlib.sha256(raw).hexdigest()
|
||||
if len(raw) > 2_000_000 or len(raw) != receipt['bytes'] or digest != receipt['sha256']:
|
||||
raise ValueError('Snapshot checksum or size differs')
|
||||
evidence = {'url':url, 'sha256':digest, 'bytes':len(raw),
|
||||
'file':'evidence/'+digest+'.html', 'http_status':200,
|
||||
'observed_at':timestamp(receipt['at']), 'response_receipt':receipt}
|
||||
return raw, evidence
|
||||
|
||||
|
||||
def parse_chapter(html, doi, *, book_title, publication_year):
|
||||
nodes = list(Document(html).root.walk())
|
||||
meta = defaultdict(list)
|
||||
for node in nodes:
|
||||
if node.tag == 'meta' and node.attrs.get('name', '').startswith('citation_'):
|
||||
meta[node.attrs['name']].append(node.attrs.get('content', ''))
|
||||
def one(key):
|
||||
values = meta[key]
|
||||
if len(values) != 1 or not values[0]:
|
||||
raise ValueError('Missing or ambiguous citation field: '+key)
|
||||
return values[0]
|
||||
if (one('citation_doi') != doi or one('citation_inbook_title') != book_title
|
||||
or one('citation_publication_date') != str(publication_year)):
|
||||
raise ValueError('Chapter DOI, book or publication year differs from reviewed scope')
|
||||
raw_authors = meta['citation_author']
|
||||
if not raw_authors:
|
||||
raise ValueError('Missing named chapter authors')
|
||||
authors = []
|
||||
for raw in raw_authors:
|
||||
if raw.count(',') != 1:
|
||||
raise ValueError('Unreviewed citation author format')
|
||||
family, given = [clean(s) for s in raw.split(',', 1)]
|
||||
if not family or not given:
|
||||
raise ValueError('Incomplete citation author')
|
||||
authors.append({'given':given, 'family':family})
|
||||
abstract_nodes = [n for n in nodes if n.attrs.get('id') == 'Abs1-content']
|
||||
if len(abstract_nodes) > 1:
|
||||
raise ValueError('Ambiguous abstract section')
|
||||
abstract = clean(abstract_nodes[0].text()) if abstract_nodes else None
|
||||
first, last = one('citation_firstpage'), one('citation_lastpage')
|
||||
if not first.isdigit() or not last.isdigit() or int(first) > int(last):
|
||||
raise ValueError('Invalid chapter pages')
|
||||
return {'doi':doi, 'title':one('citation_title'), 'authors':authors,
|
||||
'authors_citation_raw':list(raw_authors), 'abstract':abstract or None,
|
||||
'first_page':first, 'last_page':last}
|
||||
@@ -8,7 +8,7 @@ TABLES = ('sources', 'papers', 'identifiers', 'source_records', 'venues', 'editi
|
||||
'submission_sources', 'submission_identifiers', 'decision_sources',
|
||||
'submission_merges', 'decision_merges', 'schema_migrations', 'tags',
|
||||
'paper_tags', 'match_candidates', 'paper_merges', 'ingest_runs')
|
||||
OPTIONAL_TABLES = ('annotation_events',)
|
||||
OPTIONAL_TABLES = ('annotation_events', 'source_scope_revisions')
|
||||
CHECKS = {
|
||||
'annotation_events': ["kind IN ('tag','research_type')", 'retracts_id IS NULL OR retracts_id<>id'],
|
||||
'venues': ["venue_type IN ('conference','journal','review_platform')"],
|
||||
|
||||
@@ -0,0 +1 @@
|
||||
[{"type":"file","oid":"f87639c8a74ab77de91ea6bb4c88718a52e4250f","size":37291354,"lfs":{"oid":"bcb446a0398dfe62593e1db617f82f43bc792fc413d6d4658b0e0d32c358b888","size":37291354,"pointerSize":133},"xetHash":"c9a6d5d70b6f3a9ee09b4e085327e4f61cad91b9b61244e6bf99031fef9dd85b","path":"data/colm-00000-of-00001.parquet"},{"type":"file","oid":"27ef0f337b76d92ac1aaa4e1dc41a1253768545b","size":28627916,"lfs":{"oid":"930dd4ef90fa352fe6bccca7b73eb36eeabfb99f1e77e783572e2db83e15d779","size":28627916,"pointerSize":133},"xetHash":"468b49fe8b9a921e852c1995c4902e175ad7b54fcefb048a485e3fbad8b62bfd","path":"data/corl-00000-of-00001.parquet"},{"type":"file","oid":"c8de3418de080cbd24667e28c55abf684a8a3d9f","size":90794691,"lfs":{"oid":"e9e3aab42d28ca32a02287ed0bd0f1f9e1f3af5cb8ed8d6a9bf28d4f630773da","size":90794691,"pointerSize":133},"xetHash":"c21b83e5820cea8b7a6bcac34bbd334a0a6ac5b7acb1c4c681715a52ec2cceda","path":"data/emnlp-00000-of-00001.parquet"},{"type":"file","oid":"0a226079bb3fe5256a51750aa874c0e7325f60fa","size":167662887,"lfs":{"oid":"3a65801f95fe490b30282df2cb668fa33fe4182e9d80544148ea66da35215ad8","size":167662887,"pointerSize":134},"xetHash":"2225125f6d18015bb3c44037c82a7f53363eeb636d309de3445e1f1b49f51a03","path":"data/iclr-00000-of-00005.parquet"},{"type":"file","oid":"7b53c49c39f2c287a778e64b18969690926710a8","size":196573710,"lfs":{"oid":"736c05be45eed9db00848471d88dc9cd5a1b1a7d5a0db6432639a21dfb8b4562","size":196573710,"pointerSize":134},"xetHash":"48da5a74c33ca8b04411f4d6f41e34a4ba57d8c4aecb1a72be5dcad89a72bd65","path":"data/iclr-00001-of-00005.parquet"},{"type":"file","oid":"cbf09e34f7f00e33f5bb501461a915ab4bda2ec2","size":210807163,"lfs":{"oid":"1c0b77c370b8735e3d4ebd578706ec8dfbb5e53eb6a71e8cbfeaec36c78ee13f","size":210807163,"pointerSize":134},"xetHash":"fa8bf4848b55d274464b6d06413e1183321e6ffd467e1efa369ee32e0749e8af","path":"data/iclr-00002-of-00005.parquet"},{"type":"file","oid":"4082a290b2e2785413340920e3475c4b5a58b05f","size":250525528,"lfs":{"oid":"078d441ef3854be30f8fd73d34e8442a2a119cd26dbaddf9ab8aba2008df507f","size":250525528,"pointerSize":134},"xetHash":"63ba7c9757b53b02b1793f2186ddc55b79149dbda9a37bacedaf2b0466604d61","path":"data/iclr-00003-of-00005.parquet"},{"type":"file","oid":"d4528dcfcd6d7f56619d2bed75d4705c11811bfb","size":263725824,"lfs":{"oid":"de5efe8578b744f59c8ff9d7c1171cdaa760008920dde190c04b73e5c45901a5","size":263725824,"pointerSize":134},"xetHash":"5934eb773ec49ebf9f48c89d9d7e88ae0d2eb5f5e7f83bfd13f51080aa6d4090","path":"data/iclr-00004-of-00005.parquet"},{"type":"file","oid":"68cd41fa1983b2ba666deb74d1e6da2f9fa9003d","size":204752048,"lfs":{"oid":"4f50e2a0fd7e40a0fdb239595b636c753ca89af1ad9692b7d63f366e7a25eda0","size":204752048,"pointerSize":134},"xetHash":"de8ac963c1738ffe68728255bf7f32fbfff3863b3b07ca1f7026f4c76f123ec0","path":"data/icml-00000-of-00001.parquet"},{"type":"file","oid":"580a98d667a2fc7252fd2bcc2eeeaff116f7030d","size":126503813,"lfs":{"oid":"d2c4359745d30900bd19a44762c9fb6b8415554e2aead6f22f902362aca3e61a","size":126503813,"pointerSize":134},"xetHash":"27cd8f2318a0d096cf3a6aa7178fc502ce52c5d026887f64661b46af74b4b9d5","path":"data/neurips-00000-of-00005.parquet"},{"type":"file","oid":"e90e69bdd1985a36f9a72b4b3b3a9b54d90b7b9d","size":166685127,"lfs":{"oid":"32be2572bc42d33ff227ceadbdcc64e27c996b31e82195d25eced11ec4617963","size":166685127,"pointerSize":134},"xetHash":"2e3528c026d585513df96b23efe39d174b368abd3088e494d85fcf806504a241","path":"data/neurips-00001-of-00005.parquet"},{"type":"file","oid":"b7be2e96d3c3726b87c4a997fcb2c144288b70b4","size":229115194,"lfs":{"oid":"7f9dc8025aebe1d78f74bfccc68e1792f6a076142a920305065d805ff04ab248","size":229115194,"pointerSize":134},"xetHash":"77a4b411199fee5bd1b3fb8752ad4bf1ac090d560a4bfde54b75cf19d48c545b","path":"data/neurips-00002-of-00005.parquet"},{"type":"file","oid":"127763711c2fe76631137432291fa839746536a3","size":264212756,"lfs":{"oid":"32d86e74f00d3fce8eef1cf1db1dd78eed341ddda0e09a924ba17d08c3d566b2","size":264212756,"pointerSize":134},"xetHash":"abe6a12cf6f6f2db52ba28b84a0b91fcfab1411dfa0eb0ab7cdae540b86c5c9d","path":"data/neurips-00003-of-00005.parquet"},{"type":"file","oid":"89eb22ff72a8f44d461e601176239d4f997bbec7","size":279556753,"lfs":{"oid":"b4a536cdb6d306049a32121272b8224ef0e37360423ca04c274f0da74d4dc64c","size":279556753,"pointerSize":134},"xetHash":"d2b68be869a02a2cbe81c74e50fcac3c35585422dbaffcb19985e2376a90e0a7","path":"data/neurips-00004-of-00005.parquet"},{"type":"file","oid":"2b6048e08f894bb47b874a485d3bc5655a43a449","size":211604858,"lfs":{"oid":"df21b892ec531bc223b114d03cd51be6f4f2c48563edde0bdd1b153c36a74a98","size":211604858,"pointerSize":134},"xetHash":"1e81fc6612959ef821e97b7b5ed06a9ebbb3748f0ecc2fe2027b41f7626ef863","path":"data/tmlr-00000-of-00001.parquet"}]
|
||||
+2
-1
@@ -12,6 +12,7 @@ dependencies = ["PyYAML==6.0.3"]
|
||||
|
||||
[project.optional-dependencies]
|
||||
postgres = ["psycopg[binary]==3.3.8"]
|
||||
mirrors = ["pyarrow==26.0.1"]
|
||||
|
||||
[project.scripts]
|
||||
paper-atlas = "paper_atlas.cli:main"
|
||||
@@ -22,4 +23,4 @@ exclude = ["tests*"]
|
||||
|
||||
[tool.setuptools.package-data]
|
||||
"paper_atlas.db" = ["schema.sql", "annotations.sql"]
|
||||
"paper_atlas.resources" = ["venues.json"]
|
||||
"paper_atlas.resources" = ["venues.json", "reviewarena-files.json"]
|
||||
|
||||
BIN
Binary file not shown.
BIN
Binary file not shown.
BIN
Binary file not shown.
Vendored
BIN
Binary file not shown.
+706
@@ -0,0 +1,706 @@
|
||||
{
|
||||
"status": "reviewed26_presentation_only_title_correspondences_not_work_identity_approval",
|
||||
"gold": [
|
||||
{
|
||||
"doi": "10.1145/3767308.3834907",
|
||||
"title": "H\n <sup>2</sup>\n AL: Hyperbolic Hierarchy-aware Aggregative Learning for Registration-based Few-shot Medical Image Segmentation",
|
||||
"authors": [
|
||||
"Jia Wang",
|
||||
"Jiaming Cai",
|
||||
"Zunying Hu",
|
||||
"Zhanjie Wu",
|
||||
"Jinyuan Liu",
|
||||
"Hua Cheng",
|
||||
"Yun Peng",
|
||||
"Xin Fan"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-004.json",
|
||||
"source_position": 246,
|
||||
"registry_object_sha256": "dd4a627c13ca806e15e59d3ae894c9216d14d33c85bc5ce7362b51fad8c8a692",
|
||||
"program_id": "mfp0053",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "516fac88f1b7a2821c1e2b97cc41530f32abbe2514b65b5823427ff3ecdc9a23",
|
||||
"program_context_pointers": [
|
||||
"/entries/129/papers/19",
|
||||
"/entries/129/subsessions/1/items/7"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3834936",
|
||||
"title": "M\n <sup>3</sup>\n TR: Temporal Retrieval Enhanced Multi-Modal Micro-video Popularity Prediction",
|
||||
"authors": [
|
||||
"Jiacheng Lu",
|
||||
"Weijian Wang",
|
||||
"Mingyuan Xiao",
|
||||
"Yang Hua",
|
||||
"Tao Song",
|
||||
"Bo Peng",
|
||||
"Cheng Hua",
|
||||
"Haibing Guan"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-005.json",
|
||||
"source_position": 25,
|
||||
"registry_object_sha256": "93340f8b923d7cc0d848941107351fb661d95effa35b1869921e7163da36bae4",
|
||||
"program_id": "mfp0229",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "6968e6a2d364075020e0b6c95c9076742531681232c40df511a2e8992c6b25f9",
|
||||
"program_context_pointers": [
|
||||
"/entries/109/papers/27",
|
||||
"/entries/109/subsessions/2/items/3"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3834946",
|
||||
"title": "<i>PAST:</i>\n Prompt-Adaptive Sampling Termination for Efficient Diffusion Model",
|
||||
"authors": [
|
||||
"Renye Yan",
|
||||
"Jikang Cheng",
|
||||
"You Wu",
|
||||
"Wei Peng",
|
||||
"Zongwei Wang",
|
||||
"Ling Liang",
|
||||
"Yimao Cai"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-005.json",
|
||||
"source_position": 35,
|
||||
"registry_object_sha256": "4210191c2871090711820aef9aff06289694380c9d0194ec9b8b1e5e40fad00d",
|
||||
"program_id": "mfp0297",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "b47bd7502bf3a8e9e9dc87e3d9552efe928cc455f843d527a4860c430c5e9490",
|
||||
"program_context_pointers": [
|
||||
"/entries/61/papers/40",
|
||||
"/entries/61/subsessions/3/items/4"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3834956",
|
||||
"title": "<i>SwiftExplorer:</i>\n Training-free diffusion model alignment with Swift Diversity Exploration",
|
||||
"authors": [
|
||||
"Renye Yan",
|
||||
"Jikang Cheng",
|
||||
"You Wu",
|
||||
"Baojin Huang",
|
||||
"Wei Peng",
|
||||
"Zongwei Wang",
|
||||
"Ling Liang",
|
||||
"Yimao Cai"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-006.json",
|
||||
"source_position": 235,
|
||||
"registry_object_sha256": "7f2812ca4072437ecaaa585e85139e7cba0e0d721f9a2c0ac4e9897739faed47",
|
||||
"program_id": "mfp0339",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "f7809f57290e44fe291ffdc07ea956e5484aa491390cd385db75d056f95b840a",
|
||||
"program_context_pointers": [
|
||||
"/entries/88/papers/48",
|
||||
"/entries/88/subsessions/4/items/0"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835035",
|
||||
"title": "<i>Balancing Efficiency and Efficacy:</i>\n Training-Free Attention-Guided Switching Between Explicit and Latent Thoughts for MLLMs",
|
||||
"authors": [
|
||||
"Haoqian Kang",
|
||||
"Liupeng Li",
|
||||
"Kuofeng Gao",
|
||||
"Jinpeng Wang",
|
||||
"Zhenyu Lu",
|
||||
"Bin Chen",
|
||||
"Ke Chen",
|
||||
"Yaowei Wang"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-004.json",
|
||||
"source_position": 8,
|
||||
"registry_object_sha256": "3d76f692754dc159642eeb1706491631267d437a95c4459674361ebf35f416b9",
|
||||
"program_id": "mfp0798",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "477e42fa3d5c8236d6415ff18ba26d799d659b9e2f3f09c1d80bc0971dbca125",
|
||||
"program_context_pointers": [
|
||||
"/entries/88/papers/39",
|
||||
"/entries/88/subsessions/3/items/3"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835055",
|
||||
"title": "TM\n <sup>2</sup>\n Q-Det: Boosting Video Object Detection via Text-Guided Spatiotemporal Mamba and Multimodal Query Learning",
|
||||
"authors": [
|
||||
"Qiang Qi",
|
||||
"Wenqi Shang",
|
||||
"Xiao Wang",
|
||||
"Shuyuan Lin"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-004.json",
|
||||
"source_position": 28,
|
||||
"registry_object_sha256": "7e3c3e6acdad588970c681bdb92969f5134086baa2a5a7b9414871135d5ea137",
|
||||
"program_id": "mfp0903",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "8549d38e781896079ab04ef6a771a4061a0fe42df2fa5c5c03a037d59cc9f548",
|
||||
"program_context_pointers": [
|
||||
"/entries/129/papers/55",
|
||||
"/entries/129/subsessions/4/items/7"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835101",
|
||||
"title": "<scp>ConfAvatar:</scp>\n Confidence-Aware Self-Bootstrapping for Monocular 4D Head Avatar Reconstruction",
|
||||
"authors": [
|
||||
"Zifan Zheng",
|
||||
"Yu Luo",
|
||||
"Yuxin Huang",
|
||||
"Chuancheng Shi",
|
||||
"Mohammad Masbaul Alam Polash",
|
||||
"Zhiyong Wang",
|
||||
"Kun Hu"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-008.json",
|
||||
"source_position": 82,
|
||||
"registry_object_sha256": "b838756bd0b6af036e408936f8f13a6398d83e524ce0282167199eca092bc7d8",
|
||||
"program_id": "mfp1175",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "cb8dbe67e16e163e5b136ad3401fc57c827905e9704d7c0a84961bc8f12565f4",
|
||||
"program_context_pointers": [
|
||||
"/entries/109/papers/86",
|
||||
"/entries/109/subsessions/7/items/2"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835374",
|
||||
"title": "<scp>DateGBC:</scp>\n Deep Fuzzy Granular-Ball Clustering via Adaptive Topology Evolution for Unsupervised Image Segmentation",
|
||||
"authors": [
|
||||
"Qianpeng Chong",
|
||||
"Jiakun Wen",
|
||||
"Guangyi Wei",
|
||||
"Wenyi Zeng"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-006.json",
|
||||
"source_position": 59,
|
||||
"registry_object_sha256": "a7fe694a04449e841be5181888ae35786d0a4938d87eca1fffb0776d4d350e30",
|
||||
"program_id": "mfp2617",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "b57b7026c750aeebb47598860b82980215c9e4e4935f6e8611a31b2fe64f73b9",
|
||||
"program_context_pointers": [
|
||||
"/entries/61/papers/80",
|
||||
"/entries/61/subsessions/6/items/8"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835424",
|
||||
"title": "<i>RTCBench</i>\n : A Scalable Benchmarking Framework for Multi-Party Video Conferencing",
|
||||
"authors": [
|
||||
"Evan Ram",
|
||||
"Tuan Tran",
|
||||
"Youngwook Son",
|
||||
"Sangtae Ha"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-005.json",
|
||||
"source_position": 108,
|
||||
"registry_object_sha256": "b21198e0eba35d175bc83292be8c87bca2b80a9c8e01f40b640a4e9efc7ecde5",
|
||||
"program_id": "mfp2909",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "2398ec6539c50b14c7ed4fa773277cc860151ab77b5430b49d935f88e90a487e",
|
||||
"program_context_pointers": [
|
||||
"/entries/129/papers/130",
|
||||
"/entries/129/subsessions/8/items/22"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835452",
|
||||
"title": "<scp>HeadRouter:</scp>\n Dynamic Head-Weight Routing for Task-Adaptive Audio Token Pruning in Large Audio Language Models",
|
||||
"authors": [
|
||||
"He Peize",
|
||||
"Yaodi Luo",
|
||||
"Xiaoqian Liu",
|
||||
"Xuyang Liu",
|
||||
"Jiahang Deng",
|
||||
"Yaosong Du",
|
||||
"Li Bangyu",
|
||||
"Xiyan Gui",
|
||||
"Yuxuan Chen",
|
||||
"Linfeng Zhang"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-004.json",
|
||||
"source_position": 167,
|
||||
"registry_object_sha256": "ee31b12972ea5844a50c6c63a6fc3430875c3388a4bf261f8e76cab31112d351",
|
||||
"program_id": "mfp3048",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "c455de7c27f43d0fb49990ead5fca795334c0b613eafa2b3c6150eb147d9c3fa",
|
||||
"program_context_pointers": [
|
||||
"/entries/46/papers/78",
|
||||
"/entries/46/subsessions/6/items/6"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835460",
|
||||
"title": "<i>CoAnchor:</i>\n Robust Collaborative Perception under Spatio-Temporal Misalignment via Object-Level Anchors",
|
||||
"authors": [
|
||||
"Chi Li",
|
||||
"Rui Lin",
|
||||
"Aobo Ji",
|
||||
"Dongzhu Xu"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-004.json",
|
||||
"source_position": 175,
|
||||
"registry_object_sha256": "7c41b0dcb65ad27a1f0961f095889ccc73044776250ec816cf95fc25c30da7f1",
|
||||
"program_id": "mfp3109",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "5f3d97775d83528d6f41fb2fc481d1848cac7c3f9de1d469039450f9b2ecbff1",
|
||||
"program_context_pointers": [
|
||||
"/entries/109/papers/121",
|
||||
"/entries/109/subsessions/8/items/13"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835544",
|
||||
"title": "<i>OC-GS:</i>\n Object-Centric Representation and Demand Alignment for Immersive 3D Gaussian Streaming",
|
||||
"authors": [
|
||||
"Boyan Li",
|
||||
"Ruijun Mao",
|
||||
"Zhicheng Liang",
|
||||
"Dayou Zhang",
|
||||
"Fangxin Wang"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-006.json",
|
||||
"source_position": 129,
|
||||
"registry_object_sha256": "afc0c7db304251511eef5e223f027dc6314dd70895955ff1a1cb0a461dbfdbe9",
|
||||
"program_id": "mfp3554",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "00b2ffe1073feede8447bac439ac7842c1ba0841cb6ee38ec1b5e06c3a3dc22e",
|
||||
"program_context_pointers": [
|
||||
"/entries/46/papers/153",
|
||||
"/entries/46/subsessions/9/items/21"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835591",
|
||||
"title": "DP\n <sup>2</sup>\n -NeRF: Dual-Prior Probabilistic Neural Radiance Fields for Event-Guided Motion Deblurring",
|
||||
"authors": [
|
||||
"Jiafeng Huang",
|
||||
"Wenzhen Jia",
|
||||
"Shilong Dong",
|
||||
"Shengjie Zhao"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-007.json",
|
||||
"source_position": 97,
|
||||
"registry_object_sha256": "145025e784ec251247f8cd86d0dbee1749666ec71f35566c20e41edd0a2347ce",
|
||||
"program_id": "mfp3783",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "0b2d267736cfe506de666618820e2934c23395bd25dab63c5a347710445a361c",
|
||||
"program_context_pointers": [
|
||||
"/entries/33/papers/158",
|
||||
"/entries/33/subsessions/10/items/2"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835667",
|
||||
"title": "MMCoT\n <sup>3</sup>\n : Empowering 3D Scene Understanding with 3D-Text Interleaved Multi-Modal Chain-of-Thought Reasoning",
|
||||
"authors": [
|
||||
"Yongqi Shan",
|
||||
"Yifan Wang",
|
||||
"Yunzhi Zhuge",
|
||||
"Lijun Wang",
|
||||
"Huchuan Lu"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-005.json",
|
||||
"source_position": 209,
|
||||
"registry_object_sha256": "8b5ad8d891692e441b6e210304dcd35103835fd35d086c860c7611bf5385a6d9",
|
||||
"program_id": "mfp4173",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "4924721a2c3328e6bca661654569bca86d3fa69aa5f6c9e3719f827c5f0d7639",
|
||||
"program_context_pointers": [
|
||||
"/entries/88/papers/114",
|
||||
"/entries/88/subsessions/8/items/6"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835763",
|
||||
"title": "<scp>AudioKV:</scp>\n KV Cache Eviction in Efficient Large Audio Language Models",
|
||||
"authors": [
|
||||
"Yuxuan Wang",
|
||||
"He Peize",
|
||||
"Xiyan Gui",
|
||||
"Xiaoqian Liu",
|
||||
"Junhao He",
|
||||
"Xuyang Liu",
|
||||
"Xuming Hu",
|
||||
"Linfeng Zhang"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-006.json",
|
||||
"source_position": 192,
|
||||
"registry_object_sha256": "70054b0a5a2586fd08e75494eee08a26498730554601a5fe2e4efddcaad309de",
|
||||
"program_id": "mfp4778",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "274b360da553dfc7afdb9072505c2957cec4213e096373e2edbf58236a3fc3a2",
|
||||
"program_context_pointers": [
|
||||
"/entries/129/papers/100",
|
||||
"/entries/129/subsessions/7/items/16"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835895",
|
||||
"title": "F\n <sup>2</sup>\n -Adapter: A Focus-and-Forget Framework for Generalizable CLIP-based Face Forgery Detection",
|
||||
"authors": [
|
||||
"Zan Gao",
|
||||
"Xin Duan",
|
||||
"Nian Hu",
|
||||
"Hao Jia",
|
||||
"Yibo Zhao",
|
||||
"Cong Liu"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-001.json",
|
||||
"source_position": 39,
|
||||
"registry_object_sha256": "16bdbcc4398df4d9556c14d4d4203127349070669f5d0e1c4bf87756e241b698",
|
||||
"program_id": "mfp5537",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "b3ef626b66d93ac37a73caf15a4cf9ff92241209c80ee66be5575e6b56f9c899",
|
||||
"program_context_pointers": [
|
||||
"/entries/88/papers/184",
|
||||
"/entries/88/subsessions/11/items/6"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3835959",
|
||||
"title": "UPR-VAD:\n <i>U</i>\n ncertainty-guided Signal\n <i>P</i>\n urification and\n <i>R</i>\n egularization for Weakly-Supervised\n <i>V</i>\n ideo\n <i>A</i>\n nomaly\n <i>D</i>\n etection",
|
||||
"authors": [
|
||||
"Weiliang Huang",
|
||||
"Weichao Cai",
|
||||
"Bob Zhang",
|
||||
"Chao Huang",
|
||||
"Biao Xue",
|
||||
"Rongxin Zhang"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-007.json",
|
||||
"source_position": 21,
|
||||
"registry_object_sha256": "e96f9e9d67cce31a9235bdc7dc5dba551676f3d36ba72654ae3d5a98ae8d5174",
|
||||
"program_id": "mfp5823",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "d8469bcb68f8cc19070f3bf9a866c3377ba27fe6445b5bf9337b7599a6881592",
|
||||
"program_context_pointers": [
|
||||
"/entries/88/papers/66",
|
||||
"/entries/88/subsessions/5/items/6"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3836215",
|
||||
"title": "BLM\n <sub>1</sub>\n : A Boundless Large Model for Cross-Space, Cross-Task, and Cross-Embodiment Learning",
|
||||
"authors": [
|
||||
"Wentao Tan",
|
||||
"Bowen Wang",
|
||||
"Heng Zhi",
|
||||
"Chenyv Liu",
|
||||
"Zhe Li",
|
||||
"Jian Liu",
|
||||
"Zengrong Lin",
|
||||
"Yukun Dai",
|
||||
"Yipeng Chen",
|
||||
"Wenjie Yang",
|
||||
"Enci Xie",
|
||||
"Hao Xue",
|
||||
"Baixu Ji",
|
||||
"Chen Xu",
|
||||
"Zhibin Wang",
|
||||
"Tianshi Wang",
|
||||
"Lei Zhu",
|
||||
"Heng Tao Shen"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-003.json",
|
||||
"source_position": 46,
|
||||
"registry_object_sha256": "b4b51e5d0a4c0a330d01e7638c0bcc73871d750312081bafbd3b8a10529153f2",
|
||||
"program_id": "mfp7332",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "aa458905209ad562117d72dd33d4d61e000b0d9efaa55dd9672919f40e6cc2fd",
|
||||
"program_context_pointers": [
|
||||
"/entries/33/papers/13",
|
||||
"/entries/33/subsessions/1/items/1",
|
||||
"/entries/47/papers/0"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3836272",
|
||||
"title": "SEE: Structure-aware Exploring & Exploiting for Long-horizon GUI Agent Trajectory Synthesis",
|
||||
"authors": [
|
||||
"Zhuohang Fan",
|
||||
"Beichen Zhang",
|
||||
"Yuanfa Li",
|
||||
"Changqiao Wu",
|
||||
"Wei Liu",
|
||||
"Jian Luan",
|
||||
"Weigang Zhang"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-002.json",
|
||||
"source_position": 127,
|
||||
"registry_object_sha256": "66eebbdef9e2d07fe08f3cb41cd577095f282f1256bdb73fe1e0d4c5d6e55d51",
|
||||
"program_id": "mfp7635",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "df11d727337f397bab72977bcdce4246b388ef0c8e1198ac47fdc32e68a3aff4",
|
||||
"program_context_pointers": [
|
||||
"/entries/88/papers/131",
|
||||
"/entries/88/subsessions/8/items/23"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3836293",
|
||||
"title": "Video\n <sup>*</sup>\n Bench: Benchmarking Fine-grained Video Perception with \"Think with Videos\"",
|
||||
"authors": [
|
||||
"Ruizhe Chen",
|
||||
"Zhiting Fan",
|
||||
"Yang Shi",
|
||||
"Enxin Song",
|
||||
"Wenhao Chai",
|
||||
"Tongkun Guan",
|
||||
"Songtao Jiang",
|
||||
"Ruilin Luo",
|
||||
"Yuanxing Zhang",
|
||||
"Shuai Bai",
|
||||
"Sibo Song",
|
||||
"Zhibo Yang",
|
||||
"Zuozhu Liu"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-002.json",
|
||||
"source_position": 148,
|
||||
"registry_object_sha256": "7332426edb3d15499f175fb98c093b78df21ea95fdd9cfea33da5e925cbec171",
|
||||
"program_id": "mfp7732",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "f6e4df110fe85d7f6d34eb712cd338443c2e34e924393a4e65e07a0c7b152c9e",
|
||||
"program_context_pointers": [
|
||||
"/entries/46/papers/198",
|
||||
"/entries/46/subsessions/11/items/20"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3836447",
|
||||
"title": "<i>Insert In Style</i>\n : A Zero-Shot Generative Framework for Harmonious Cross-Domain Object Composition",
|
||||
"authors": [
|
||||
"Raghu Vamsi Chittersu",
|
||||
"Yuvraj Singh Rathore",
|
||||
"Pranav Adlinge",
|
||||
"Eswar Sai Pashaula",
|
||||
"Kunal Swami"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-002.json",
|
||||
"source_position": 183,
|
||||
"registry_object_sha256": "92b2c094b96cebbace0150805b65b0a1bc3ca99ed42124bcb8de442a81094e1b",
|
||||
"program_id": "mfp8648",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "39d3f21f774010980f8b30f2fd800b5243ea7ddc6e091b4d8769de3dee622273",
|
||||
"program_context_pointers": [
|
||||
"/entries/107/papers/4",
|
||||
"/entries/117/papers/30",
|
||||
"/entries/117/subsessions/2/items/6"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3836600",
|
||||
"title": "<scp>EEGaze</scp>\n : Multimodal Attention Prediction in Visual Search via EEG-Based Saliency Refinement",
|
||||
"authors": [
|
||||
"Mansi Sharma",
|
||||
"Dinar Orazgaliyev",
|
||||
"Antonio Krüger",
|
||||
"Philipp Müller"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-003.json",
|
||||
"source_position": 140,
|
||||
"registry_object_sha256": "30b99b234aa2b944d6890119b2524d3b3c3c425bde22f20e813b9517b8f8536b",
|
||||
"program_id": "mfp9815",
|
||||
"track_literal": "Main Track",
|
||||
"track": "main",
|
||||
"abstract_sha256": "47dcb456aee600663d0df59429c1745b335dbae9bdef2dd1063265429fff49fb",
|
||||
"program_context_pointers": [
|
||||
"/entries/46/papers/219",
|
||||
"/entries/46/subsessions/12/items/17"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3838656",
|
||||
"title": "<scp>MedMKG:</scp>\n Benchmarking Medical Knowledge Exploitation with Multimodal Knowledge Graph",
|
||||
"authors": [
|
||||
"Xiaochen Wang",
|
||||
"Yuan Zhong",
|
||||
"Lingwei Zhang",
|
||||
"Lisong Dai",
|
||||
"Ting Wang",
|
||||
"Fenglong Ma"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-003.json",
|
||||
"source_position": 225,
|
||||
"registry_object_sha256": "b910b66541ca65127646d35d0cd5737a56ae2e0f311a79526005dec75fca739f",
|
||||
"program_id": "mds177",
|
||||
"track_literal": "Datasets",
|
||||
"track": "datasets",
|
||||
"abstract_sha256": "5bee3b5e9ff60690509d1f34953f3dcc185c50861e112d2c4a3baf5673f6b945",
|
||||
"program_context_pointers": [
|
||||
"/entries/46/papers/23",
|
||||
"/entries/46/subsessions/1/items/11"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3839012",
|
||||
"title": "<i>EULALens</i>\n : SEE What YOU Agree!",
|
||||
"authors": [
|
||||
"Abhijeet Narang",
|
||||
"Hagar Elbatanouny",
|
||||
"Vanishri Shekhawat",
|
||||
"Hasan Al-Nashash",
|
||||
"Abhinav Dhall",
|
||||
"Usman Tariq"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-007.json",
|
||||
"source_position": 238,
|
||||
"registry_object_sha256": "81dd29dafd6228f51ae2f32e2bb713272a88df78c78347b9170b54f83db00c84",
|
||||
"program_id": "mde41",
|
||||
"track_literal": "Demo and Video",
|
||||
"track": "demo_and_video",
|
||||
"abstract_sha256": "9acbb264b8019916aad0845aeae4eb5a9c94903ac5bca3cfb612c857306f6d55",
|
||||
"program_context_pointers": [
|
||||
"/entries/94/papers/7"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3839024",
|
||||
"title": "<scp>ThreatVerse:</scp>\n A Closed-Loop World Model Simulator for Autonomous Driving Safety Evaluation",
|
||||
"authors": [
|
||||
"Jiangfan Liu",
|
||||
"Zexuan Cui",
|
||||
"Tianyuan Zhang",
|
||||
"Zonglei Jing",
|
||||
"Yaoyuan Zhang",
|
||||
"Jiakai Wang",
|
||||
"Jinyang Guo",
|
||||
"Yuqing Ma",
|
||||
"Lu Sheng",
|
||||
"Aishan Liu",
|
||||
"Xianglong Liu"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-008.json",
|
||||
"source_position": 0,
|
||||
"registry_object_sha256": "d0571f4562b30e4e7fb2e3aa826068ac14a540d1ae70e8cf0b11fd73c67f26ee",
|
||||
"program_id": "mde15",
|
||||
"track_literal": "Demo and Video",
|
||||
"track": "demo_and_video",
|
||||
"abstract_sha256": "02fc9b31d2b7667ea13f00c74f923a375373042171aaa1ff1f2010ee9697211a",
|
||||
"program_context_pointers": [
|
||||
"/entries/49/papers/4"
|
||||
],
|
||||
"single_name_positions": []
|
||||
},
|
||||
{
|
||||
"doi": "10.1145/3767308.3839028",
|
||||
"title": "<scp>FinAlogy:</scp>\n A Visual Analogy Retrieval System for Financial K-Line Analysis",
|
||||
"authors": [
|
||||
"Xiaoyan Yu",
|
||||
"Ziya Zhao",
|
||||
"Yunshan Ma",
|
||||
"Yifan Wei",
|
||||
"Dianbo Sui",
|
||||
"Tat-Seng Chua"
|
||||
],
|
||||
"kind": "unique_program_format_title_fullauthor_agreement_not_identity_approval",
|
||||
"source_page": "page-008.json",
|
||||
"source_position": 4,
|
||||
"registry_object_sha256": "b18c8e2821873a59318f6f09d1f453dd02195ccedad5c8e0ce1da60ac07466fb",
|
||||
"program_id": "mde02",
|
||||
"track_literal": "Demo and Video",
|
||||
"track": "demo_and_video",
|
||||
"abstract_sha256": "22f978dd02c05439663c5724f25bb997098fd85f5558d01d836711252d119931",
|
||||
"program_context_pointers": [
|
||||
"/entries/49/papers/0"
|
||||
],
|
||||
"single_name_positions": []
|
||||
}
|
||||
],
|
||||
"author_disagreement_hold": {
|
||||
"doi": "10.1145/3767308.3836308",
|
||||
"program_id": "mfp7801"
|
||||
},
|
||||
"scope": "Onlytitlecomparisonchanges;originalregistryandprogramfixturesremainunchanged;sourceABandtrackcomefromtheexplicitprogramID.",
|
||||
"held_gold": {
|
||||
"doi": "10.1145/3767308.3836308",
|
||||
"title": "Dynamic Late Fusion Multiple Kernel\n <i>k</i>\n -means Clustering based on Adaptive Sample Selection and Alignment",
|
||||
"authors": [
|
||||
"Zhirong Tian",
|
||||
"Yi Zhang",
|
||||
"Weixuan Liang",
|
||||
"Miaomiao Li",
|
||||
"Xinwang Liu"
|
||||
],
|
||||
"kind": "program_format_author_version_requires_review",
|
||||
"source_page": "page-001.json",
|
||||
"source_position": 82,
|
||||
"registry_object_sha256": "26a3c1715d276b5fbe988596e3115091086158f6ef12600a36e5206785fc9dfb",
|
||||
"program_id": null,
|
||||
"track_literal": null,
|
||||
"track": "proceedings_unclassified",
|
||||
"abstract_sha256": null,
|
||||
"program_context_pointers": [],
|
||||
"single_name_positions": []
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,124 @@
|
||||
"""Checks for published ECCV metadata and preserved submission histories."""
|
||||
from hashlib import sha256
|
||||
import json
|
||||
from pathlib import Path
|
||||
import sqlite3
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
from paper_atlas.collectors.eccv2026_publications import BOOK_TITLE, SOURCE, collect
|
||||
from paper_atlas.db.store import Store
|
||||
from paper_atlas.ingestion.records import ingest_file
|
||||
|
||||
|
||||
FIRST_BOOK = '10.1007/978-3-032-10000-0'
|
||||
FIRST_DOI = FIRST_BOOK + '_1'
|
||||
|
||||
|
||||
def snapshots(root):
|
||||
toc, chapters = root / 'toc', root / 'chapters'
|
||||
toc.mkdir()
|
||||
chapters.mkdir()
|
||||
receipts, volumes = [], {}
|
||||
for offset in range(78):
|
||||
volume = 17001 + offset
|
||||
book = f'10.1007/978-3-032-{10000 + offset}-0'
|
||||
book_url = 'https://link.springer.com/book/' + book
|
||||
count = 37 if offset < 27 else 36
|
||||
files, chapter_dois = {}, []
|
||||
for page, indices in ((1, range(1, 20)), (2, range(20, count + 1))):
|
||||
filename = f'{volume}-{page}.html'
|
||||
html = f'<h1>{BOOK_TITLE}</h1><p>Malmö (LNCS, volume {volume})</p>'
|
||||
html += '<li data-test="chapter"><h3>Front Matter</h3></li>'
|
||||
for i in indices:
|
||||
doi = f'{book}_{i}'
|
||||
html += f'<li data-test="chapter"><h3><a href="/chapter/{doi}">Title {volume}-{i}</a></h3></li>'
|
||||
chapter_dois.append(doi)
|
||||
raw = html.encode()
|
||||
(toc / filename).write_bytes(raw)
|
||||
files[str(page)] = filename
|
||||
receipts.append({'file': filename, 'status': 200, 'bytes': len(raw), 'sha256': sha256(raw).hexdigest(),
|
||||
'url': book_url + ('?page=2' if page == 2 else '')})
|
||||
volumes[str(volume)] = {'book': book_url, 'files': files, 'chapter_dois': chapter_dois}
|
||||
data = ''.join(json.dumps(row) + '\n' for row in receipts).encode()
|
||||
(toc / 'receipts.jsonl').write_bytes(data)
|
||||
checkpoint = json.dumps(volumes).encode()
|
||||
(toc / 'volume-checkpoint.json').write_bytes(checkpoint)
|
||||
(toc / 'summary.json').write_text(json.dumps({
|
||||
'raw_receipts_sha256': sha256(data).hexdigest(),
|
||||
'volume_checkpoint_sha256': sha256(checkpoint).hexdigest(),
|
||||
'chapter_link_count': 2835,
|
||||
}))
|
||||
url = 'https://link.springer.com/chapter/' + FIRST_DOI
|
||||
citation = {
|
||||
'citation_doi': FIRST_DOI, 'citation_title': 'Title 17001-1',
|
||||
'citation_inbook_title': BOOK_TITLE, 'citation_publication_date': '2026',
|
||||
'citation_firstpage': '1', 'citation_lastpage': '10',
|
||||
'citation_author': 'Chen, Lei',
|
||||
}
|
||||
body = ''.join(f'<meta name="{k}" content="{v}">' for k, v in citation.items())
|
||||
body += '<div id="Abs1-content"><p>Published abstract.</p></div>'
|
||||
raw = body.encode()
|
||||
(chapters / 'first.html').write_bytes(raw)
|
||||
(chapters / 'receipts.json').write_text(json.dumps({'responses': [{
|
||||
'url': url, 'final_url': url, 'file': 'first.html', 'at': '2026-11-18T00:00:00Z',
|
||||
'http_status': 200, 'bytes': len(raw), 'sha256': sha256(raw).hexdigest(),
|
||||
'headers': {'Content-Type': 'text/html'},
|
||||
}]}))
|
||||
return {'toc_snapshots': toc, 'chapter_snapshots': chapters}
|
||||
|
||||
|
||||
class ECCV2026PublicationsTests(unittest.TestCase):
|
||||
def test_frozen_snapshot_and_existing_history(self):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root = Path(tmp)
|
||||
source = snapshots(root)
|
||||
report = collect(root / 'out', **source)
|
||||
self.assertEqual((report['toc_chapters'], report['chapter_details_missing']), (2835, 2834))
|
||||
manifest, = report['files']
|
||||
self.assertEqual((manifest['track'], manifest['status']), ('main', 'partial'))
|
||||
row = json.loads(Path(manifest['file']).read_text())
|
||||
self.assertEqual(row['identifiers'], {'doi': FIRST_DOI})
|
||||
self.assertEqual(row['submission_identifiers'], [{'scheme': 'publication_doi', 'value': FIRST_DOI}])
|
||||
self.assertEqual(row['authors'], [{'family': 'Chen', 'given': 'Lei'}])
|
||||
self.assertEqual(row['abstract'], 'Published abstract.')
|
||||
self.assertEqual(row['metadata']['lncs_volume'], 17001)
|
||||
self.assertFalse(row['metadata']['original_decision_signature_verified'])
|
||||
detail = row['metadata']['publisher_chapter_evidence']
|
||||
self.assertEqual(sha256((root / 'out' / detail['file']).read_bytes()).hexdigest(), detail['sha256'])
|
||||
db = root / 'example.sqlite'
|
||||
store = Store(db)
|
||||
pid, rid = store.record('prior', 'work', 'https://example.org/work',
|
||||
row['title'], 'Original preprint abstract.', row['authors'],
|
||||
{'arxiv': '2601.00001'})
|
||||
store.decision(pid, rid, 'ICLR', 2026, 'main', 'work', 'https://example.org/decision',
|
||||
'rejected', 'Original ICLR result.')
|
||||
store.conn.commit()
|
||||
store.conn.close()
|
||||
row['identifiers']['arxiv'] = '2601.00001'
|
||||
path = Path(manifest['file'])
|
||||
raw = (json.dumps(row) + '\n').encode()
|
||||
path.write_bytes(raw)
|
||||
manifest['file_sha256'] = sha256(raw).hexdigest()
|
||||
path.with_suffix('.manifest.json').write_text(json.dumps(manifest))
|
||||
ingest_file(db, path)
|
||||
ingest_file(db, path)
|
||||
with sqlite3.connect(db) as conn:
|
||||
self.assertEqual(conn.execute('SELECT id,abstract FROM papers').fetchall(),
|
||||
[(pid, 'Original preprint abstract.')])
|
||||
self.assertEqual(conn.execute('SELECT status FROM decisions ORDER BY id').fetchall(),
|
||||
[('rejected',), ('accepted',)])
|
||||
self.assertEqual(conn.execute('SELECT scheme,value FROM submission_identifiers').fetchall(),
|
||||
[('publication_doi', FIRST_DOI)])
|
||||
self.assertEqual(conn.execute('SELECT name FROM sources ORDER BY id').fetchall(), [('prior',), (SOURCE,)])
|
||||
with self.assertRaises(FileExistsError):
|
||||
collect(root / 'out', **source)
|
||||
|
||||
def test_changed_contents_receipt_stops_before_output(self):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root = Path(tmp)
|
||||
source = snapshots(root)
|
||||
(source['toc_snapshots'] / '17001-1.html').write_text('unverified bytes')
|
||||
with self.assertRaisesRegex(ValueError, 'contents bytes disagree'):
|
||||
collect(root / 'out', **source)
|
||||
self.assertFalse((root / 'out').exists())
|
||||
@@ -0,0 +1,95 @@
|
||||
import base64
|
||||
import copy
|
||||
import gzip
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
from paper_atlas.collectors import icassp2026 as c
|
||||
|
||||
|
||||
class ICASSP2026Test(unittest.TestCase):
|
||||
@classmethod
|
||||
def setUpClass(cls):
|
||||
cls.fixture = json.loads(gzip.decompress((Path(__file__).parent/'fixtures/icassp2026-source.json.gz').read_bytes()))
|
||||
cls.files = {name: base64.b64decode(body) for name, body in cls.fixture['files'].items()}
|
||||
cls.items = []
|
||||
cls.manifest = json.loads(cls.files[c.SNAPSHOT+'/manifest.json'])
|
||||
for receipt in cls.manifest['pages']:
|
||||
cls.items.extend((r, receipt) for r in json.loads(cls.files[c.SNAPSHOT+'/'+receipt['file']])['message']['items'])
|
||||
|
||||
def materialize(self, root):
|
||||
for name, body in self.files.items():
|
||||
path = root/name; path.parent.mkdir(parents=True, exist_ok=True); path.write_bytes(body)
|
||||
|
||||
def test_full_census_matches_independent_gold_and_preserves_source(self):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root = Path(tmp); src = root/'src'; self.materialize(src)
|
||||
report = c.collect(root/'out', snapshots=src)
|
||||
self.assertEqual((report['records'],report['missing_abstract_count'],report['single_name_author_fields']), (4582,4582,23))
|
||||
self.assertEqual(len(report['excluded_nonpaper_items']),7); self.assertEqual(len(report['excluded_other_conference_dois']),56)
|
||||
self.assertFalse(report['database_written']); self.assertFalse(report['pdfs_downloaded']); self.assertEqual(report['network_requests'],0)
|
||||
self.assertEqual(report['approved_for_import'],0); self.assertEqual(report['public_rejected_availability'],'unknown')
|
||||
manifest, = report['files']; path = Path(manifest['file']); raw = path.read_bytes()
|
||||
self.assertEqual(hashlib.sha256(raw).hexdigest(),manifest['file_sha256']); self.assertEqual(manifest['status'],'partial')
|
||||
rows = [json.loads(l) for l in raw.splitlines()]
|
||||
actual = [{'doi':r['external_id'],'title':r['title'],'authors':r['authors']} for r in rows]
|
||||
gold = sorted(self.fixture['independent_gold'],key=lambda r:r['doi'])
|
||||
self.assertEqual(actual,gold)
|
||||
original = {r['DOI']:r for r,_ in self.items}
|
||||
evidence_hashes = {str(p.relative_to(root/'out')):hashlib.sha256(p.read_bytes()).hexdigest() for p in (root/'out/evidence').iterdir()}
|
||||
for r in rows:
|
||||
doi = r['external_id']; self.assertEqual(r['metadata']['registry_fields'],original[doi])
|
||||
self.assertEqual(r['identifiers'],{'doi':doi}); self.assertEqual(r['submission_identifiers'],[{'scheme':'publication_doi','value':doi}])
|
||||
self.assertEqual(r['track'],'proceedings_unclassified'); self.assertFalse(r['tags']); self.assertFalse(r['abstract'])
|
||||
self.assertFalse(r['metadata']['original_decision_signature_verified'])
|
||||
evidence = r['metadata']['source_evidence']
|
||||
self.assertEqual(evidence_hashes[evidence['file']],evidence['sha256'])
|
||||
self.assertIn('Fareedha',next(r for r in rows if r['external_id'].endswith('.11461474'))['authors'])
|
||||
with self.assertRaisesRegex(ValueError,'Output exists'):
|
||||
c.collect(root/'out',snapshots=src)
|
||||
|
||||
def test_wrong_conference_doi_date_and_unexpected_abstract_rejected(self):
|
||||
for kind in ('container','event','doi','date','author','sequence','abstract'):
|
||||
with self.subTest(kind=kind):
|
||||
items = copy.deepcopy(self.items); r = next(r for r,_ in items if r.get('author') and r['DOI'].startswith(c.PREFIX))
|
||||
if kind=='container': r['container-title']=['Foreign']
|
||||
if kind=='event': r['event']['name']='Other'
|
||||
if kind=='doi': r['DOI']='10.1109/foreign.2026.1'
|
||||
if kind=='date': r['published']['date-parts']=[[2025,5,3]]
|
||||
if kind=='author': r['author']=[]
|
||||
if kind=='sequence': r['author'][0]['sequence']='additional'
|
||||
if kind=='abstract': r['abstract']='new source version'
|
||||
with self.assertRaises(ValueError): c.registry_papers(items)
|
||||
|
||||
def test_nonpaper_exclusion_requires_exact_reviewed_evidence(self):
|
||||
for kind in ('title','page','author'):
|
||||
items = copy.deepcopy(self.items); r = next(r for r,_ in items if r['DOI'] in c.NONPAPERS)
|
||||
if kind=='title': r['title']=['A research paper']
|
||||
if kind=='page': r['page']='1-5'
|
||||
if kind=='author': r['author']=[{'family':'Smith'}]
|
||||
with self.assertRaisesRegex(ValueError,'nonpaper'): c.registry_papers(items)
|
||||
|
||||
def test_duplicate_and_incomplete_census_rejected(self):
|
||||
with self.assertRaisesRegex(ValueError,'repeated'): c.registry_papers(self.items+[self.items[0]])
|
||||
with self.assertRaisesRegex(ValueError,'partition'): c.registry_papers(self.items[:-1])
|
||||
|
||||
def test_corruption_and_partial_input_preserve_failure_without_success(self):
|
||||
for kind in ('manifest','body','missing'):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp); src=root/'src'; self.materialize(src)
|
||||
path=src/c.SNAPSHOT/('manifest.json' if kind=='manifest' else 'page-002.json')
|
||||
if kind=='missing': path.unlink()
|
||||
else: path.write_bytes(path.read_bytes()+b' ')
|
||||
with self.assertRaises((ValueError,FileNotFoundError)): c.collect(root/'out',snapshots=src)
|
||||
self.assertTrue((root/'out/failed.json').exists()); self.assertFalse((root/'out/collection-report.json').exists())
|
||||
|
||||
def test_symlink_page_and_snapshot_rejected(self):
|
||||
for kind in ('page','directory'):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp); src=root/'src'; self.materialize(src)
|
||||
path=src/c.SNAPSHOT/('page-001.json' if kind=='page' else '')
|
||||
renamed=path.with_name(path.name+'-original'); path.rename(renamed); path.symlink_to(renamed,target_is_directory=kind=='directory')
|
||||
with self.assertRaisesRegex(ValueError,'Unsafe'): c.collect(root/'out',snapshots=src)
|
||||
@@ -0,0 +1,92 @@
|
||||
import base64
|
||||
import copy
|
||||
import gzip
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
from paper_atlas.collectors import ismir2026 as c
|
||||
|
||||
|
||||
class ISMIR2026Test(unittest.TestCase):
|
||||
@classmethod
|
||||
def setUpClass(cls):
|
||||
cls.fixture=json.loads(gzip.decompress((Path(__file__).parent/'fixtures/ismir2026-source.json.gz').read_bytes()))
|
||||
cls.files={name:base64.b64decode(raw) for name,raw in cls.fixture['files'].items()}
|
||||
cls.program=json.loads(cls.files['ismir2026-json-research-v1/papers.json'])
|
||||
cls.accepted=cls.files['remaining2026-listings-research-v2/ismir-accepted.html']
|
||||
|
||||
def materialize(self,root):
|
||||
for name,raw in self.files.items():
|
||||
p=root/name;p.parent.mkdir(parents=True,exist_ok=True);p.write_bytes(raw)
|
||||
|
||||
def test140_source_versions_and_review_optins_against_independent_gold(self):
|
||||
listing=c.accepted_listing(self.accepted);observations=c.program_observations(self.program,listing)
|
||||
self.assertEqual(len(observations),140)
|
||||
for observation,gold in zip(observations,self.fixture['independent_gold']):
|
||||
r=observation['original'];v=r['content'];official=observation['accepted_listing']
|
||||
self.assertEqual((r['id'],r['forum'],v['title'],v['authors'],v['abstract']),(gold['program_id'],gold['forum_literal'],gold['title'],gold['authors'],gold['abstract']))
|
||||
self.assertEqual(official,{'title':gold['official_title'],'authors_literal':gold['official_authors_literal'],'author_names':gold['official_author_names']})
|
||||
self.assertEqual(v['session'],gold['session_literal']);self.assertEqual(v['paper_presentation'],gold['presentation_literal']);self.assertEqual(v['keywords'],gold['keywords'])
|
||||
self.assertEqual(sum('program_and_accepted_title_literal_versions_differ' in r['review_flags'] for r in observations),2)
|
||||
self.assertEqual({r['original']['id'] for r in observations if 'program_and_accepted_author_sequence_versions_differ' in r['review_flags']},{'60','126','433'})
|
||||
self.assertEqual(sum(r['original']['reviews']['publish_reviews']=='Yes' for r in observations),101)
|
||||
|
||||
def test_full_collection_keeps_abstract_bytes_review_text_tags_and_no_inferred_ids(self):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);src=root/'src';self.materialize(src);out=root/'out';report=c.collect(out,snapshots=src)
|
||||
self.assertEqual((report['records'],report['missing_abstract_count'],report['review_optin_yes'],report['review_optin_no']),(140,0,101,39))
|
||||
self.assertFalse(report['database_written']);self.assertFalse(report['pdfs_downloaded']);self.assertEqual(report['network_requests'],0);self.assertEqual(report['approved_for_import'],0)
|
||||
self.assertEqual(report['public_rejected_availability'],'unknown_accepted_program_reviews_are_not_a_rejected_export')
|
||||
manifest,=report['files'];raw=Path(manifest['file']).read_bytes();self.assertEqual(hashlib.sha256(raw).hexdigest(),manifest['file_sha256']);self.assertEqual(manifest['status'],'partial')
|
||||
rows=[json.loads(l) for l in raw.splitlines()];self.assertEqual(len(rows),140)
|
||||
for row,original,gold in zip(rows,self.program,self.fixture['independent_gold']):
|
||||
self.assertEqual(row['metadata']['program_entry'],original);self.assertEqual(row['abstract'],gold['abstract']);self.assertEqual(row['title'],gold['title']);self.assertEqual(row['authors'],gold['authors'])
|
||||
self.assertEqual(row['external_id'],'2026/'+gold['program_id']);self.assertEqual(row['identifiers'],{});self.assertEqual(row['submission_identifiers'],[])
|
||||
self.assertEqual(row['status'],'accepted');self.assertEqual(row['track'],'program_unclassified');self.assertIsNone(row['metadata']['paper_type']);self.assertFalse(row['metadata']['original_decision_signature_verified'])
|
||||
tags=[{'namespace':'ismir_program:presentation','value':gold['presentation_literal']}]+[{'namespace':'ismir_program:keyword','value':v} for v in dict.fromkeys(' '.join(k.split()) for k in gold['keywords'] if k.strip())]
|
||||
self.assertEqual(row['tags'],tags);self.assertTrue(all(t['value'] for t in row['tags']))
|
||||
if original['reviews']['publish_reviews']=='No':self.assertFalse(any(v for k,v in row['metadata']['program_entry']['reviews'].items() if k!='publish_reviews'))
|
||||
for name,body in self.files.items():self.assertEqual((out/'evidence'/name).read_bytes(),body)
|
||||
with self.assertRaisesRegex(ValueError,'Output exists'):c.collect(out,snapshots=src)
|
||||
|
||||
def test_duplicate_and_malformed_locator_or_incomplete_census_rejected(self):
|
||||
listing=c.accepted_listing(self.accepted)
|
||||
for kind in ('repeat','forum','non_numeric','missing_row'):
|
||||
rows=copy.deepcopy(self.program)
|
||||
if kind=='repeat':rows[1]['id']=rows[0]['id'];rows[1]['forum']=rows[0]['forum']
|
||||
if kind=='forum':rows[0]['forum']='OriginalOpenReviewID'
|
||||
if kind=='non_numeric':rows[0]['id']='abc';rows[0]['forum']='abc'
|
||||
if kind=='missing_row':rows.pop()
|
||||
with self.subTest(kind=kind),self.assertRaises(ValueError):c.program_observations(rows,listing)
|
||||
|
||||
def test_empty_fields_foreign_title_and_session_disagree_rejected(self):
|
||||
listing=c.accepted_listing(self.accepted)
|
||||
for field,value in [('authors',[]),('abstract',''),('title','A foreign paper'),('session',['99']),('keywords',[42])]:
|
||||
rows=copy.deepcopy(self.program);rows[0]['content'][field]=value
|
||||
with self.subTest(field=field),self.assertRaises(ValueError):c.program_observations(rows,listing)
|
||||
|
||||
def test_source_optin_no_cannot_publish_review_text(self):
|
||||
rows=copy.deepcopy(self.program);r=next(r for r in rows if r['reviews']['publish_reviews']=='No');r['reviews']['meta_review']='Do not infer a negative decision'
|
||||
with self.assertRaisesRegex(ValueError,'preference'):c.program_observations(rows,c.accepted_listing(self.accepted))
|
||||
|
||||
def test_parenthesized_affiliation_semicolons_do_not_split_people(self):
|
||||
self.assertEqual(c.author_names('A Person (University A; Lab (X)); B Person (Institute B)'),['A Person','B Person'])
|
||||
for raw in ['A Person (Broken','; A Person (Lab)','A Person )']:
|
||||
with self.assertRaises(ValueError):c.author_names(raw)
|
||||
|
||||
def test_ambiguous_official_title_is_not_fused(self):
|
||||
source=self.accepted.replace(b'FusID: Modality-Fused Semantic IDs for Generative Music Recommendation',b'Off-Manifold Robustness in Synthesizer Inversion with Joint Distribution Flow Matching',1)
|
||||
with self.assertRaisesRegex(ValueError,'ambiguous'):c.accepted_listing(source)
|
||||
|
||||
def test_tamper_missing_file_and_symlink_fail_without_success_report(self):
|
||||
for kind in ('manifest','body','missing','symlink'):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);src=root/'src';self.materialize(src);path=src/'ismir2026-json-research-v1'/('manifest.json' if kind=='manifest' else 'papers.json')
|
||||
if kind=='missing':path.unlink()
|
||||
elif kind=='symlink':other=root/'actual.json';path.rename(other);path.symlink_to(other)
|
||||
else:path.write_bytes(path.read_bytes()+b' ')
|
||||
with self.subTest(kind=kind),self.assertRaises((ValueError,FileNotFoundError)):c.collect(root/'out',snapshots=src)
|
||||
self.assertTrue((root/'out/failed.json').exists());self.assertFalse((root/'out/collection-report.json').exists())
|
||||
@@ -0,0 +1,146 @@
|
||||
import copy
|
||||
import gzip
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
from paper_atlas.collectors import mm2026 as c
|
||||
|
||||
|
||||
class MM2026Test(unittest.TestCase):
|
||||
@classmethod
|
||||
def setUpClass(cls):
|
||||
cls.fixture = json.loads(gzip.decompress((Path(__file__).parent/'fixtures/mm2026-source.json.gz').read_bytes()))
|
||||
cls.files = {name:body.encode('utf-8') for name,body in cls.fixture['files'].items()}
|
||||
cls.raw = cls.files[c.SNAPSHOT+'/program-data.json']
|
||||
cls.data = json.loads(cls.raw)
|
||||
cls.gold = cls.fixture['independent_inventory']
|
||||
|
||||
def materialize(self, root):
|
||||
for name,body in self.files.items():
|
||||
path = root/name; path.parent.mkdir(parents=True,exist_ok=True); path.write_bytes(body)
|
||||
|
||||
def parse_changed(self, data):
|
||||
return c.parse_program(json.dumps(data,ensure_ascii=False).encode())
|
||||
|
||||
def test_all2143_ids_and4326_locations_against_independent_inventory(self):
|
||||
rows,sessions = c.parse_program(self.raw)
|
||||
self.assertEqual(len(rows),2143); self.assertEqual(len(sessions),157)
|
||||
self.assertEqual(sum(len(r['contexts']) for r in rows),4326)
|
||||
actual = {r['paper_id']:r for r in rows}
|
||||
self.assertEqual(set(actual),{g['paper_id'] for g in self.gold['inventory']})
|
||||
for gold in self.gold['inventory']:
|
||||
r = actual[gold['paper_id']]
|
||||
self.assertEqual((r['title'],r['authors'],r['track'],r['affiliations']),
|
||||
(gold['title'],gold['authors'],gold['track_literal'],gold['affiliations']))
|
||||
self.assertEqual(hashlib.sha256(r['abstract'].encode()).hexdigest(),gold['abstract_sha256'])
|
||||
self.assertEqual([x['source_pointer'] for x in r['contexts']],gold['locations'])
|
||||
|
||||
def test_every_presentation_and_session_context_is_preserved(self):
|
||||
rows,sessions = c.parse_program(self.raw)
|
||||
for r in rows:
|
||||
for context in r['contexts']:
|
||||
node = self.data
|
||||
for key in context['source_pointer'].split('/')[1:]:
|
||||
node = node[int(key)] if isinstance(node,list) else node[key]
|
||||
self.assertEqual(context['presentation'],{k:node[k] for k in c.PRESENTATION if k in node})
|
||||
self.assertEqual(r['paper_id'],node['paper_id'])
|
||||
for i,session in enumerate(sessions):
|
||||
original = self.data['entries'][i]
|
||||
self.assertEqual(session['source_pointer'],'/entries/'+str(i))
|
||||
self.assertEqual(session['track'],'')
|
||||
for key in original.keys()-{'papers','subsessions'}:
|
||||
self.assertEqual(session[key],original[key])
|
||||
for j,sub in enumerate(original.get('subsessions') or []):
|
||||
saved = session['subsessions'][j]
|
||||
self.assertEqual(saved['title'],sub['title'])
|
||||
expected = [x for x in sub.get('items') or [] if 'paper_id' not in x]
|
||||
self.assertEqual([x['fields'] for x in saved['nonpaper_items']],expected)
|
||||
|
||||
def test_all_tracks_retained_without_publication_or_negative_inference(self):
|
||||
rows,_ = c.parse_program(self.raw)
|
||||
self.assertEqual({r['track'] for r in rows},set(self.gold['unique_id_track_counts']))
|
||||
for r in rows:
|
||||
self.assertFalse(r['approved_for_import']); self.assertFalse(r['final_decision_verified'])
|
||||
self.assertFalse(r['publication_identity_verified'])
|
||||
self.assertEqual(r['source_locator'],{'namespace':'mm_program:program_id','value':r['paper_id']})
|
||||
self.assertEqual(r['tags'],[{'namespace':'mm_program:track','value':r['track']}])
|
||||
self.assertEqual(r['observed_status'],'listed_in_official_program')
|
||||
self.assertTrue({'status','identifiers','submission_identifiers','paper_type','emails'}.isdisjoint(r))
|
||||
missing = [r for r in rows if not r['authors']]
|
||||
self.assertEqual({r['paper_id'] for r in missing},{'mfp4971','mfp7801'})
|
||||
self.assertTrue(all(r['missing_fields']==['authors','abstract'] for r in missing))
|
||||
self.assertEqual(sum(not r['abstract'] for r in rows),216)
|
||||
|
||||
def test_collect_evidence_hashes_and_no_contact_arrays_in_normalized_output(self):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);self.materialize(root/'src')
|
||||
report=c.collect(root/'out',snapshots=root/'src')
|
||||
self.assertEqual((report['records'],report['program_observations'],report['extra_presentation_observations']),(2143,4326,2183))
|
||||
self.assertFalse(report['database_written']);self.assertFalse(report['pdfs_downloaded'])
|
||||
self.assertEqual(report['network_requests'],0);self.assertEqual(report['approved_for_import'],0)
|
||||
self.assertEqual(report['public_negative_access']['http_statuses'],[403]*4)
|
||||
self.assertEqual(report['track_counts'],self.gold['unique_id_track_counts'])
|
||||
for name,body in self.files.items():
|
||||
self.assertEqual((root/'out/evidence'/Path(name).name).read_bytes(),body)
|
||||
n=0
|
||||
for item in report['files']:
|
||||
raw=Path(item['file']).read_bytes();self.assertEqual(hashlib.sha256(raw).hexdigest(),item['file_sha256'])
|
||||
self.assertNotIn(b'"emails"',raw);self.assertEqual(len(raw.splitlines()),item['count']);n+=item['count']
|
||||
self.assertEqual(item['format'],'mm_program_observation_v1_not_ingestion_record')
|
||||
self.assertEqual(n,2143)
|
||||
raw=(root/'out/sessions.jsonl').read_bytes()
|
||||
self.assertEqual(hashlib.sha256(raw).hexdigest(),report['sessions_file_sha256']);self.assertNotIn(b'"emails"',raw)
|
||||
with self.assertRaisesRegex(ValueError,'Output exists'):c.collect(root/'out',snapshots=root/'src')
|
||||
|
||||
def test_conflicting_core_version_is_not_silently_merged(self):
|
||||
data=copy.deepcopy(self.data)
|
||||
first=data['entries'][0]['papers'][0]
|
||||
data['entries'][0]['papers'].append({**first,'abstract':'conflicting abstract'})
|
||||
with self.assertRaisesRegex(ValueError,'Conflicting core'):self.parse_changed(data)
|
||||
|
||||
def test_unknown_nesting_cannot_silently_lose_observations(self):
|
||||
data=copy.deepcopy(self.data)
|
||||
moved=data['entries'][0]['papers'].pop()
|
||||
data['entries'][0]['unreviewed_bucket']=[moved]
|
||||
with self.assertRaisesRegex(ValueError,'Unhandled'):self.parse_changed(data)
|
||||
|
||||
def test_track_drift_unknown_fields_or_missing_title_require_review(self):
|
||||
for kind in ('outer_track','paper_track','field','title','authors'):
|
||||
data=copy.deepcopy(self.data);paper=data['entries'][0]['papers'][0]
|
||||
if kind=='outer_track':data['entries'][0]['track']='Main Track'
|
||||
elif kind=='paper_track':paper['track']='New Track'
|
||||
elif kind=='field':paper['doi']='unreviewed'
|
||||
elif kind=='title':paper['title']=''
|
||||
else:paper['authors']='Unparsed author string'
|
||||
with self.subTest(kind=kind),self.assertRaises(ValueError):self.parse_changed(data)
|
||||
|
||||
def test_edition_or_census_change_requires_review(self):
|
||||
for kind in ('edition','count'):
|
||||
data=copy.deepcopy(self.data)
|
||||
if kind=='edition':data['generated_at']='2025-11-14T18:36:26.860677+00:00'
|
||||
else:data['entries'][0]['papers'].pop()
|
||||
with self.subTest(kind=kind),self.assertRaises(ValueError):self.parse_changed(data)
|
||||
|
||||
def test_corruption_missing_evidence_and_symlink_fail_without_completed_report(self):
|
||||
for kind in ('manifest','program','probe','missing','symlink'):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);self.materialize(root/'src');directory=root/'src'/c.SNAPSHOT
|
||||
name='manifest.json' if kind=='manifest' else 'openreview-rejected-probe.json' if kind=='probe' else 'program-data.json'
|
||||
path=directory/name
|
||||
if kind=='missing':path.unlink()
|
||||
elif kind=='symlink':
|
||||
original=directory/'original.json';path.rename(original);path.symlink_to(original)
|
||||
else:path.write_bytes(path.read_bytes()+b' ')
|
||||
with self.subTest(kind=kind),self.assertRaises((ValueError,FileNotFoundError)):c.collect(root/'out',snapshots=root/'src')
|
||||
self.assertTrue((root/'out/failed.json').exists());self.assertFalse((root/'out/collection-report.json').exists())
|
||||
|
||||
def test_literal_abstract_math_and_nonpaper_session_events_retained(self):
|
||||
rows,sessions=c.parse_program(self.raw)
|
||||
self.assertTrue(any('\\' in r['abstract'] for r in rows))
|
||||
self.assertTrue(any('\n' in r['abstract'] for r in rows))
|
||||
self.assertTrue(any(s['type']=='coffee_break' for s in sessions))
|
||||
self.assertTrue(any(s['type']=='industry_talk' for s in sessions))
|
||||
self.assertEqual(c.without_emails({'emails':['contact'],'items':[{'emails':['contact'],'title':'talk'}]}),{'items':[{'title':'talk'}]})
|
||||
@@ -0,0 +1,273 @@
|
||||
from contextlib import redirect_stdout
|
||||
import io
|
||||
import os
|
||||
import uuid
|
||||
import gzip
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
from paper_atlas.collectors import mm_publications2026 as c
|
||||
|
||||
|
||||
class MMPublications2026Tests(unittest.TestCase):
|
||||
@classmethod
|
||||
def setUpClass(cls):
|
||||
fixtures=Path(__file__).parent/'fixtures'
|
||||
cls.fixture=json.loads(gzip.decompress((fixtures/'mm2026-publications.json.gz').read_bytes()))
|
||||
cls.program_fixture=json.loads(gzip.decompress((fixtures/'mm2026-source.json.gz').read_bytes()))
|
||||
cls.format_gold=json.loads((fixtures/'mm2026-title-formatting-review.json').read_text())
|
||||
overrides={g['doi']:g for g in [*cls.format_gold['gold'],cls.format_gold['held_gold']]}
|
||||
cls.gold=[overrides.get(g['doi'],g) for g in cls.fixture['independent_gold']]
|
||||
cls.temp=tempfile.TemporaryDirectory();cls.root=Path(cls.temp.name)
|
||||
cls.materialize(cls.root/'registry',cls.fixture['files'])
|
||||
cls.materialize(cls.root/'program',cls.program_fixture['files'])
|
||||
cls.out=cls.root/'out'
|
||||
cls.report=c.collect(cls.out,registry_snapshots=cls.root/'registry',program_snapshots=cls.root/'program',parent_snapshot=cls.root/'registry/parent.json')
|
||||
cls.rows=[json.loads(line) for f in cls.report['files'] for line in Path(f['file']).read_bytes().splitlines()]
|
||||
cls.by_doi={r['external_id']:r for r in cls.rows}
|
||||
|
||||
@classmethod
|
||||
def tearDownClass(cls):
|
||||
cls.temp.cleanup()
|
||||
|
||||
@staticmethod
|
||||
def materialize(root,files):
|
||||
for name,body in files.items():
|
||||
path=root/name;path.parent.mkdir(parents=True,exist_ok=True);path.write_bytes(body.encode('utf-8'))
|
||||
|
||||
def test_all1918_publications_against_independent_registry_program_gold(self):
|
||||
self.assertEqual(len(self.rows),1918);self.assertEqual(len(self.by_doi),1918)
|
||||
self.assertEqual(set(self.by_doi),{g['doi'] for g in self.gold})
|
||||
for g in self.gold:
|
||||
r=self.by_doi[g['doi']];m=r['metadata'];p=m['program_fields']
|
||||
self.assertEqual((r['title'],r['authors'],r['track']),(g['title'],g['authors'],g['track']))
|
||||
self.assertEqual(m['program_match_kind'],g['kind'])
|
||||
self.assertEqual(m['registry_source_evidence']['json_pointer'],'/message/items/'+str(g['source_position']))
|
||||
self.assertEqual(m['registry_source_evidence']['file'],'evidence/registry/'+g['source_page'])
|
||||
self.assertEqual(m['registry_source_evidence']['canonical_object_json_sha256'],g['registry_object_sha256'])
|
||||
self.assertEqual(p['paper_id'] if p else None,g['program_id'])
|
||||
self.assertEqual([x['source_pointer'] for x in p['contexts']] if p else [],g['program_context_pointers'])
|
||||
|
||||
def test_abstract_source_is_separate_and_verbatim_not_fabricated_registry_abstract(self):
|
||||
count=0
|
||||
for g in self.gold:
|
||||
r=self.by_doi[g['doi']];m=r['metadata']
|
||||
self.assertFalse(m['registry_abstract_available']);self.assertFalse(m['registry_fields'].get('abstract'))
|
||||
if g['abstract_sha256'] is not None:
|
||||
count+=1;self.assertEqual(hashlib.sha256(r['abstract'].encode()).hexdigest(),g['abstract_sha256'])
|
||||
self.assertEqual(r['abstract'],m['program_fields']['abstract'])
|
||||
self.assertEqual(m['abstract_source'],'official_program_reviewed_title_formatting_correspondence' if g['kind'].startswith('unique_program_format_') else 'official_program_exact_metadata_correspondence')
|
||||
else:
|
||||
self.assertIsNone(r['abstract']);self.assertEqual(m['abstract_source'],'not_available_from_reviewed_sources')
|
||||
self.assertEqual(count,1743);self.assertEqual(self.report['output_missing_abstract_count'],175)
|
||||
|
||||
def test_all_program_tracks_and_4326_contexts_survive_in_matches_or_unmatched(self):
|
||||
programs={r['metadata']['program_fields']['paper_id']:r['metadata']['program_fields'] for r in self.rows if r['metadata']['program_fields']}
|
||||
for p in self.report['unmatched_program']:
|
||||
self.assertNotIn(p['paper_id'],programs);programs[p['paper_id']]=p
|
||||
self.assertEqual(len(programs),2143);self.assertEqual(self.report['unmatched_program_count'],391)
|
||||
gold=self.program_fixture['independent_inventory']['inventory']
|
||||
self.assertEqual(set(programs),{g['paper_id'] for g in gold})
|
||||
for g in gold:
|
||||
p=programs[g['paper_id']]
|
||||
self.assertEqual((p['title'],p['authors'],p['track']),(g['title'],g['authors'],g['track_literal']))
|
||||
self.assertEqual([x['source_pointer'] for x in p['contexts']],g['locations'])
|
||||
self.assertEqual(sum(len(p['contexts']) for p in programs.values()),4326)
|
||||
self.assertEqual(len({p['track'] for p in programs.values()}),28)
|
||||
|
||||
def test_no_person_alias_original_submission_decision_or_research_type_inference(self):
|
||||
for r in self.rows:
|
||||
doi=r['external_id'];m=r['metadata']
|
||||
self.assertEqual(r['identifiers'],{'doi':doi})
|
||||
self.assertEqual(r['submission_identifiers'],[{'scheme':'publication_doi','value':doi}])
|
||||
self.assertEqual((r['source'],r['venue'],r['year']),(c.SOURCE,'MM',2026))
|
||||
self.assertEqual(r['status'],'accepted');self.assertIn('not an original signed decision',r['raw_decision'])
|
||||
self.assertFalse(m['original_decision_signature_verified']);self.assertFalse(m['research_type_classified'])
|
||||
self.assertFalse(m['scientific_work_identity_approved'])
|
||||
self.assertTrue({'decided_at','decision_date','source_decision_id','paper_type'}.isdisjoint(r))
|
||||
self.assertIn({'namespace':c.SOURCE+':crossref:type','value':'proceedings-article'},r['tags'])
|
||||
self.assertTrue({x['namespace'] for x in r['tags']} <= {c.SOURCE+':crossref:type',c.SOURCE+':mm_program:track'})
|
||||
for track in ('interactive_art','industry_talk','demo_and_video'):
|
||||
self.assertTrue(any(r['track']==track for r in self.rows))
|
||||
variants=[r for r in self.rows if r['track']=='proceedings_unclassified']
|
||||
self.assertEqual(len(variants),166)
|
||||
self.assertTrue(all(r['metadata']['program_fields'] is None and r['abstract'] is None for r in variants))
|
||||
|
||||
def test_original_bibliography_evidence_kept_without_duplicating_reference_payload(self):
|
||||
authors_by_page={}
|
||||
for name,body in self.fixture['files'].items():
|
||||
actual=self.out/'evidence/registry'/('parent.json' if name=='parent.json' else Path(name).name)
|
||||
self.assertEqual(actual.read_bytes(),body.encode())
|
||||
if re.fullmatch(r'page-[0-9]+\.json',Path(name).name):
|
||||
authors_by_page[str(actual.relative_to(self.out))]=[row['author'] for row in json.loads(body)['message']['items']]
|
||||
for name,body in self.program_fixture['files'].items():
|
||||
self.assertEqual((self.out/'evidence/program'/Path(name).name).read_bytes(),body.encode())
|
||||
for r in self.rows:
|
||||
m=r['metadata'];self.assertNotIn('reference',m['registry_fields'])
|
||||
self.assertEqual(m['registry_fields']['title'],[r['title']])
|
||||
self.assertEqual(m['registry_fields']['author'],authors_by_page[m['registry_source_evidence']['file']][int(m['registry_source_evidence']['json_pointer'].rsplit('/',1)[-1])])
|
||||
for f in self.report['files']:
|
||||
raw=Path(f['file']).read_bytes();self.assertEqual(hashlib.sha256(raw).hexdigest(),f['file_sha256'])
|
||||
self.assertEqual(len(raw.splitlines()),f['count']);self.assertEqual(f['status'],'partial')
|
||||
|
||||
def test_eight_single_name_fields_preserved_without_splitting(self):
|
||||
n=0
|
||||
for g in self.gold:
|
||||
r=self.by_doi[g['doi']];flags=r['metadata']['review_flags']
|
||||
actual=[x['author_position'] for x in flags if x['kind']=='single_registered_name_field_not_split']
|
||||
self.assertEqual(actual,g['single_name_positions']);n+=len(actual)
|
||||
for i in actual:
|
||||
self.assertEqual(r['authors'][i-1],r['metadata']['registry_fields']['author'][i-1]['family'].strip())
|
||||
self.assertEqual(n,8)
|
||||
|
||||
def test_structured_title_markup_retained_and_flagged_without_flattening(self):
|
||||
count=0
|
||||
for g in self.gold:
|
||||
if re.search(r'<[^>]+>|&[A-Za-z#0-9]+;',g['title']):
|
||||
count+=1;r=self.by_doi[g['doi']]
|
||||
self.assertEqual(r['title'],g['title'])
|
||||
self.assertIn({'kind':'literal_title_markup_or_entity_retained_requires_review'},r['metadata']['review_flags'])
|
||||
self.assertGreater(count,0);self.assertEqual(count,self.report['registered_title_markup_flags'])
|
||||
|
||||
|
||||
def test_formatting_gold_preserves_raw_fields_and_copies_exact_program_evidence(self):
|
||||
self.assertEqual(len(self.format_gold['gold']),26)
|
||||
for gold in self.format_gold['gold']:
|
||||
row=self.by_doi[gold['doi']];m=row['metadata'];p=m['program_fields']
|
||||
self.assertEqual((row['title'],row['authors']),(gold['title'],gold['authors']))
|
||||
self.assertEqual((p['paper_id'],row['track']),(gold['program_id'],gold['track']))
|
||||
self.assertEqual(hashlib.sha256(row['abstract'].encode()).hexdigest(),gold['abstract_sha256'])
|
||||
comparison=m['program_title_formatting_comparison']
|
||||
self.assertTrue(comparison['raw_titles_preserved']);self.assertFalse(comparison['work_identity_approved'])
|
||||
self.assertEqual(set(comparison['program_comparison_texts']),{p['paper_id']})
|
||||
self.assertFalse(m['scientific_work_identity_approved'])
|
||||
held=self.by_doi[self.format_gold['author_disagreement_hold']['doi']]
|
||||
self.assertEqual(held['metadata']['program_title_candidates'],['mfp7801'])
|
||||
self.assertEqual(held['metadata']['program_exact_candidates'],[])
|
||||
self.assertEqual(held['metadata']['program_match_kind'],'program_format_author_version_requires_review')
|
||||
self.assertIsNone(held['abstract']);self.assertEqual(held['track'],'proceedings_unclassified')
|
||||
|
||||
def test_reviewed_formatting_keeps_superscript_subscript_and_decodes_entities_once(self):
|
||||
self.assertEqual(c.formatting_text('<i>A</i> H<sup>2</sup> B<sub>1</sub> & C'),'A H2 B1 & C')
|
||||
self.assertEqual(c.formatting_text('A &amp; B'),'A & B')
|
||||
|
||||
def test_unknown_attributes_unbalanced_or_encoded_markup_not_discarded(self):
|
||||
for title in ['A <model>B</model>','A <i class="x">B</i>','A <i>B','A </i>B',
|
||||
'A <sup/>B','A <!--hidden-->B','<!DOCTYPE html>A','A <i>B</i>',
|
||||
'A <sup','A <i>B</sup>','<![CDATA[A]]>']:
|
||||
with self.subTest(title=title):self.assertIsNone(c.formatting_text(title))
|
||||
|
||||
def test_ambiguous_program_and_author_multiset_differences_remain_held(self):
|
||||
def observation(pid,authors):
|
||||
return {'paper_id':pid,'title':'M3 Model','authors':authors,'track':'Main Track'}
|
||||
rows=[{'title':'M<sup>3</sup> Model','authors':['A','B']}]
|
||||
kinds,_=c.match_program(rows,[observation('p1',['A','B']),observation('p2',['B','A'])])
|
||||
self.assertEqual(kinds,{'multiple_program_ids_same_title_authors_hold':1})
|
||||
self.assertNotIn('program',rows[0])
|
||||
rows=[{'title':'M<sup>3</sup> Model','authors':['A','A','B']}]
|
||||
c.match_program(rows,[observation('p1',['A','B'])])
|
||||
self.assertEqual(rows[0]['kind'],'program_format_author_version_requires_review')
|
||||
self.assertNotIn('program',rows[0])
|
||||
|
||||
def test_literal_title_author_disagreement_does_not_use_formatting_bypass(self):
|
||||
rows=[{'title':'M<sup>3</sup> Model','authors':['A','B']}]
|
||||
observations=[{'paper_id':'literal','title':'M<sup>3</sup> Model','authors':['Other'],'track':'Main Track'},
|
||||
{'paper_id':'formatted','title':'M3 Model','authors':['A','B'],'track':'Main Track'}]
|
||||
c.match_program(rows,observations)
|
||||
self.assertEqual(rows[0]['kind'],'program_author_version_requires_review')
|
||||
self.assertEqual(rows[0]['program_title_candidates'],['literal']);self.assertNotIn('program',rows[0])
|
||||
|
||||
|
||||
def test_corrupt_registry_parent_or_program_keeps_failure_not_complete_output(self):
|
||||
for kind in ('registry_manifest','registry_page','parent','program'):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);self.materialize(root/'registry',self.fixture['files']);self.materialize(root/'program',self.program_fixture['files'])
|
||||
path={'registry_manifest':root/'registry'/c.SNAPSHOT/'manifest.json','registry_page':root/'registry'/c.SNAPSHOT/'page-001.json','parent':root/'registry/parent.json','program':root/'program'/c.program.SNAPSHOT/'program-data.json'}[kind]
|
||||
path.write_bytes(path.read_bytes()+b' ')
|
||||
with self.subTest(kind=kind),self.assertRaises(ValueError):
|
||||
c.collect(root/'out',registry_snapshots=root/'registry',program_snapshots=root/'program',parent_snapshot=root/'registry/parent.json')
|
||||
self.assertTrue((root/'out/failed.json').exists());self.assertFalse((root/'out/collection-report.json').exists())
|
||||
|
||||
def test_symlink_source_rejected_and_no_previous_output_overwrite(self):
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);self.materialize(root/'registry',self.fixture['files'])
|
||||
path=root/'registry/parent.json';original=root/'registry/actual.json';path.rename(original);path.symlink_to(original)
|
||||
with self.assertRaisesRegex(ValueError,'Unsafe'):
|
||||
c.collect(root/'out',registry_snapshots=root/'registry',program_snapshots=self.root/'program',parent_snapshot=path)
|
||||
with self.assertRaisesRegex(ValueError,'Output exists'):
|
||||
c.collect(self.out,registry_snapshots=self.root/'registry',program_snapshots=self.root/'program',parent_snapshot=self.root/'registry/parent.json')
|
||||
|
||||
def test_complete_report_retains_source_partitions_not_import_or_final_dedup_claim(self):
|
||||
self.assertEqual(self.report['program_match_counts'],{'unique_program_title_fullauthor_agreement_not_identity_approval':1726,'unique_program_format_title_fullauthor_agreement_not_identity_approval':26,'program_author_version_requires_review':122,'program_format_author_version_requires_review':1,'no_exact_program_title_requires_review':43})
|
||||
self.assertEqual(self.report['registry_missing_abstract_count'],1918)
|
||||
self.assertEqual(self.report['abstracts_from_exact_program_correspondence'],1743)
|
||||
self.assertEqual(self.report['abstracts_from_literal_program_correspondence'],1717)
|
||||
self.assertEqual(self.report['abstracts_from_reviewed_title_formatting'],26)
|
||||
self.assertEqual(self.report['unmatched_program_count'],391)
|
||||
self.assertEqual(sum(f['count'] for f in self.report['files']),1918)
|
||||
self.assertEqual(self.report['approved_for_import'],0);self.assertFalse(self.report['database_written'])
|
||||
self.assertFalse(self.report['pdfs_downloaded']);self.assertEqual(self.report['network_requests'],0)
|
||||
self.assertIn('unknown',self.report['public_negative_availability'])
|
||||
|
||||
|
||||
def assert_native_ingestion(self, pg_service=None):
|
||||
# An isolated synthetic database exercises the writer contract, not a
|
||||
# production identity approval. All source rows/track partitions run.
|
||||
from paper_atlas.db.store import Store
|
||||
from paper_atlas.db.backend import open_store
|
||||
from paper_atlas.ingestion.records import ingest_file
|
||||
from paper_atlas.postgres.client import connect
|
||||
from paper_atlas.postgres.migration import migrate
|
||||
from paper_atlas.postgres.schema import quote
|
||||
with tempfile.TemporaryDirectory() as tmp:
|
||||
root=Path(tmp);db=root/'seed.sqlite';seed=self.rows[0]
|
||||
store=Store(db)
|
||||
paper,record=store.record('arxiv','synthetic-mm-anchor','https://example.org/arxiv',
|
||||
seed['title'],'Existing canonical abstract',seed['authors'],seed['identifiers'])
|
||||
store.decision(paper,record,'ICLR',2025,'main','synthetic-old-forum',
|
||||
'https://example.org/decision','rejected','Reject')
|
||||
before=tuple(store.conn.execute('SELECT * FROM papers WHERE id=?',(paper,)).fetchone())
|
||||
store.conn.commit();store.conn.close()
|
||||
schema='pa_mm_pub_'+uuid.uuid4().hex[:12] if pg_service else None
|
||||
try:
|
||||
with redirect_stdout(io.StringIO()):
|
||||
if pg_service:migrate(db,pg_service,root/'migration.json',schema=schema,min_free_gib=0)
|
||||
for f in self.report['files']:
|
||||
ingest_file(None if pg_service else db,f['file'],pg_service=pg_service,pg_schema=schema)
|
||||
store=open_store(None if pg_service else db,pg_service=pg_service,pg_schema=schema)
|
||||
try:
|
||||
conn=store.conn
|
||||
self.assertEqual(tuple(conn.execute('SELECT * FROM papers WHERE id=?',(paper,)).fetchone()),before)
|
||||
self.assertEqual(conn.execute('SELECT status FROM decisions WHERE submission_id=1').fetchone()[0],'rejected')
|
||||
actual={r[0]:r for r in conn.execute("SELECT sr.external_id,sr.paper_id,sr.title,sr.abstract,sr.authors_json,sr.metadata_json,sr.id FROM source_records sr JOIN sources s ON s.id=sr.source_id WHERE s.name=?",(c.SOURCE,))}
|
||||
self.assertEqual(set(actual),set(self.by_doi))
|
||||
self.assertEqual(actual[seed['external_id']][1],paper)
|
||||
for doi,row in self.by_doi.items():
|
||||
got=actual[doi];self.assertEqual((got[2],got[3],json.loads(got[4])),(row['title'],row['abstract'],row['authors']))
|
||||
meta=json.loads(got[5]);self.assertEqual(meta.pop('record_identifiers'),{'doi':doi,c.SOURCE:doi});self.assertEqual(meta,row['metadata'])
|
||||
tags=conn.execute("SELECT t.namespace,t.value,count(*) FROM paper_tags pt JOIN tags t ON t.id=pt.tag_id GROUP BY t.namespace,t.value").fetchall()
|
||||
self.assertEqual(sum(r[2] for r in tags if r[0]==c.SOURCE+':crossref:type'),1918)
|
||||
self.assertEqual(sum(r[2] for r in tags if r[0]==c.SOURCE+':mm_program:track'),1752)
|
||||
self.assertEqual({r[0] for r in tags},{c.SOURCE+':crossref:type',c.SOURCE+':mm_program:track'})
|
||||
decisions=conn.execute("SELECT d.status,d.decision_date,count(*) FROM decisions d JOIN submissions s ON s.id=d.submission_id JOIN source_records r ON r.id=s.source_record_id JOIN sources src ON src.id=r.source_id WHERE src.name=? GROUP BY d.status,d.decision_date",(c.SOURCE,)).fetchall()
|
||||
self.assertEqual([tuple(r) for r in decisions],[('accepted',None,1918)])
|
||||
self.assertEqual(conn.execute("SELECT count(*) FROM submission_identifiers WHERE scheme='publication_doi'").fetchone()[0],1918)
|
||||
coverage=conn.execute("SELECT es.status,sum(es.imported_count) FROM edition_sources es JOIN sources s ON s.id=es.source_id WHERE s.name=? GROUP BY es.status",(c.SOURCE,)).fetchall()
|
||||
self.assertEqual([tuple(r) for r in coverage],[('partial',1918)])
|
||||
finally:store.conn.close()
|
||||
finally:
|
||||
if pg_service:
|
||||
with connect(pg_service) as pg:
|
||||
for name, in pg.execute('SELECT nspname FROM pg_namespace WHERE starts_with(nspname,%s)',(schema,)).fetchall():
|
||||
pg.execute('DROP SCHEMA '+quote(name)+' CASCADE')
|
||||
|
||||
def test_all1918_native_sqlite_source_tags_and_prior_rejection_survive(self):
|
||||
self.assert_native_ingestion()
|
||||
|
||||
@unittest.skipUnless(os.environ.get('PAPER_ATLAS_TEST_PG_SERVICE'),'Requires approved remote PostgreSQL test service')
|
||||
def test_all1918_native_postgres_source_tags_and_prior_rejection_survive(self):
|
||||
self.assert_native_ingestion(os.environ['PAPER_ATLAS_TEST_PG_SERVICE'])
|
||||
@@ -0,0 +1,101 @@
|
||||
import gzip
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
from unittest.mock import patch
|
||||
|
||||
from paper_atlas.collectors import mm2026
|
||||
from paper_atlas.audits.source_identity import audit, load_inputs, load_mm_program_inputs
|
||||
|
||||
|
||||
class MMProgramIdentityInputTests(unittest.TestCase):
|
||||
@classmethod
|
||||
def setUpClass(cls):
|
||||
cls.fixture = json.loads(gzip.decompress((Path(__file__).parent/'fixtures/mm2026-source.json.gz').read_bytes()))
|
||||
|
||||
def setUp(self):
|
||||
self.tmp = tempfile.TemporaryDirectory()
|
||||
self.addCleanup(self.tmp.cleanup)
|
||||
self.root = Path(self.tmp.name)
|
||||
src = self.root/'source'
|
||||
for name,body in self.fixture['files'].items():
|
||||
path=src/name;path.parent.mkdir(parents=True,exist_ok=True);path.write_bytes(body.encode())
|
||||
mm2026.collect(self.root/'program', snapshots=src)
|
||||
self.report_path = self.root/'program/collection-report.json'
|
||||
|
||||
def change_report(self, change):
|
||||
report=json.loads(self.report_path.read_text());change(report)
|
||||
self.report_path.write_text(json.dumps(report))
|
||||
|
||||
def assert_rejected_before_database(self):
|
||||
with patch('paper_atlas.audits.source_identity.ReadConnection') as connection:
|
||||
with self.assertRaises(ValueError):
|
||||
audit(None,self.root/'audit',pg_service='unused',pg_schema='research',mm_program_report=self.report_path)
|
||||
connection.assert_not_called()
|
||||
self.assertFalse((self.root/'audit').exists())
|
||||
|
||||
def test_all2143_programs_are_unknown_readonly_inputs_not_import_records(self):
|
||||
rows,inputs=load_mm_program_inputs(self.report_path)
|
||||
self.assertEqual(len(rows),2143);self.assertEqual(len(inputs),28)
|
||||
self.assertEqual(sum(len(r['program_contexts']) for r in rows),4326)
|
||||
actual={r['program_id']:r for r in rows}
|
||||
for gold in self.fixture['independent_inventory']['inventory']:
|
||||
r=actual[gold['paper_id']]
|
||||
self.assertEqual((r['title'],r['authors'],r['track']),
|
||||
(gold['title'],gold['authors'],gold['track_literal']))
|
||||
self.assertEqual(r['external_id'],'2026/'+gold['paper_id'])
|
||||
self.assertEqual(r['status'],'unknown');self.assertEqual(r['identifiers'],{})
|
||||
self.assertEqual(r['source_observed_status'],'listed_in_official_program')
|
||||
self.assertFalse(r['publication_identity_verified']);self.assertFalse(r['final_decision_verified'])
|
||||
self.assertFalse(r['contribution_type_verified'])
|
||||
self.assertEqual([c['source_pointer'] for c in r['program_contexts']],gold['locations'])
|
||||
self.assertEqual({r['program_id'] for r in rows if not r['authors']},{'mfp4971','mfp7801'})
|
||||
self.assertTrue(all('unknown is not a source decision' in i['manifest']['scope'] for i in inputs))
|
||||
self.assertFalse(any(self.root.rglob('*.manifest.json')))
|
||||
|
||||
def test_untrusted_absolute_file_paths_are_not_followed(self):
|
||||
self.change_report(lambda r:[f.update(file='/untrusted/'+Path(f['file']).name) for f in r['files']])
|
||||
rows,inputs=load_mm_program_inputs(self.report_path)
|
||||
self.assertEqual(len(rows),2143)
|
||||
self.assertTrue(all(Path(i['file']).parent==self.report_path.parent for i in inputs))
|
||||
|
||||
def test_program_files_are_not_silently_accepted_by_regular_ingestion_input_loader(self):
|
||||
report=json.loads(self.report_path.read_text())
|
||||
with self.assertRaises(FileNotFoundError):load_inputs([report['files'][0]['file']])
|
||||
|
||||
def test_corrupt_jsonl_fails_before_connection_or_output(self):
|
||||
report=json.loads(self.report_path.read_text());path=Path(report['files'][0]['file'])
|
||||
path.write_bytes(path.read_bytes()+b' ')
|
||||
self.assert_rejected_before_database()
|
||||
|
||||
def test_rehashed_forged_identifiers_decisions_and_contexts_rejected(self):
|
||||
original=self.report_path.read_bytes();report=json.loads(original);path=Path(report['files'][0]['file']);raw=path.read_bytes()
|
||||
for field,value in [('identifiers',{'openreview':'invented'}),('status','accepted'),('contexts',[])]:
|
||||
path.write_bytes(raw);self.report_path.write_bytes(original)
|
||||
lines=raw.splitlines();row=json.loads(lines[0]);row[field]=value;lines[0]=json.dumps(row).encode()
|
||||
changed=b'\n'.join(lines)+b'\n';path.write_bytes(changed)
|
||||
self.change_report(lambda r:r['files'][0].update(file_sha256=hashlib.sha256(changed).hexdigest()))
|
||||
with self.subTest(field=field):self.assert_rejected_before_database()
|
||||
|
||||
def test_missing_or_repeated_track_rejected(self):
|
||||
original=self.report_path.read_bytes()
|
||||
for kind in ('missing','repeated'):
|
||||
self.report_path.write_bytes(original)
|
||||
if kind=='missing':self.change_report(lambda r:r['files'].pop())
|
||||
else:self.change_report(lambda r:r['files'].__setitem__(1,r['files'][0]))
|
||||
with self.subTest(kind=kind):self.assert_rejected_before_database()
|
||||
|
||||
def test_raw_source_corruption_or_symlink_rejected(self):
|
||||
path=self.report_path.parent/'evidence/program-data.json';raw=path.read_bytes()
|
||||
path.write_bytes(raw+b' ');self.assert_rejected_before_database()
|
||||
path.write_bytes(raw);saved=path.with_name('original.json');path.rename(saved);path.symlink_to(saved)
|
||||
self.assert_rejected_before_database()
|
||||
|
||||
def test_mixed_input_modes_rejected_before_connection(self):
|
||||
with patch('paper_atlas.audits.source_identity.ReadConnection') as connection:
|
||||
for files,program in [([],None),([Path('unused')],self.report_path)]:
|
||||
with self.assertRaisesRegex(ValueError,'Choose source files'):
|
||||
audit(files,self.root/'out',pg_service='unused',pg_schema='research',mm_program_report=program)
|
||||
connection.assert_not_called()
|
||||
@@ -15,7 +15,7 @@ from paper_atlas.db.backend import open_store, open_read_connection
|
||||
from paper_atlas.db.store import Store
|
||||
from paper_atlas.ingestion import arxiv
|
||||
from paper_atlas.ingestion.backfill import backfill
|
||||
from paper_atlas.ingestion.records import ingest_file
|
||||
from paper_atlas.ingestion.records import coverage_options, ingest_file, require_incremental_parent
|
||||
from paper_atlas.postgres.client import connect, ReadConnection
|
||||
from paper_atlas.postgres.migration import migrate
|
||||
from paper_atlas.postgres.schema import quote
|
||||
@@ -114,6 +114,36 @@ class PostgreSQLIngestionTests(unittest.TestCase):
|
||||
self.assertEqual(run[0],1)
|
||||
self.assertEqual(json.loads(run[1])['coverage']['imported_count'],3)
|
||||
|
||||
def test_incremental_parent_preflight_uses_readonly_postgres_rows_and_keeps_guards(self):
|
||||
rows = [self.row(source='dblp', external=key, status='accepted', year=2001, venue='UAI')
|
||||
for key in ('preflight1', 'preflight2')]
|
||||
path = self.file('preflight-parent', rows)
|
||||
target = path.with_suffix('.manifest.json')
|
||||
manifest = json.loads(target.read_text())
|
||||
manifest['status'] = 'partial'
|
||||
target.write_text(json.dumps(manifest))
|
||||
self.ingest(path)
|
||||
proposed = dict(manifest, coverage_mode='incremental', count=1,
|
||||
coverage_previous_count=2, coverage_previous_expected_count=2,
|
||||
coverage_expected_count=4)
|
||||
with closing(ReadConnection(self.service, self.schema)) as c:
|
||||
self.assertEqual(c.execute('SHOW transaction_read_only').fetchone()[0], 'on')
|
||||
source = c.execute("SELECT id FROM sources WHERE name='dblp'").fetchone()[0]
|
||||
before = {table: c.execute('SELECT count(*) FROM '+table).fetchone()[0]
|
||||
for table in ('papers', 'source_records', 'submissions', 'decisions', 'ingest_runs')}
|
||||
coverage = [dict(row) for row in c.execute('SELECT * FROM edition_sources ORDER BY edition_id,source_id,list_url')]
|
||||
require_incremental_parent(c, source, proposed, coverage_options(proposed))
|
||||
for changes, error in (({'coverage_previous_count': 1}, 'parent changed'),
|
||||
({'coverage_previous_expected_count': 3}, 'parent changed'),
|
||||
({'list_url': 'https://example.org/other-list'}, 'unambiguous source listing')):
|
||||
with self.subTest(changes=changes):
|
||||
stale = dict(proposed, **changes)
|
||||
with self.assertRaisesRegex(ValueError, error):
|
||||
require_incremental_parent(c, source, stale, coverage_options(stale))
|
||||
self.assertEqual({table: c.execute('SELECT count(*) FROM '+table).fetchone()[0]
|
||||
for table in before}, before)
|
||||
self.assertEqual([dict(row) for row in c.execute('SELECT * FROM edition_sources ORDER BY edition_id,source_id,list_url')], coverage)
|
||||
|
||||
def test_incremental_postgres_bound_failure_rolls_back_rows_and_summary(self):
|
||||
listing='https://example.org/uai/2001'
|
||||
rows=[self.row(source='dblp',external=key,status='accepted',year=2001,venue='UAI')
|
||||
|
||||
@@ -0,0 +1,203 @@
|
||||
"""Metadata projection never requests OCR/review column pages or promotes mirror IDs."""
|
||||
from collections import Counter
|
||||
import hashlib
|
||||
import importlib.util
|
||||
import io
|
||||
import json
|
||||
from pathlib import Path
|
||||
import re
|
||||
import struct
|
||||
import tempfile
|
||||
import unittest
|
||||
from unittest.mock import patch
|
||||
import urllib.error
|
||||
|
||||
from paper_atlas.collectors.reviewarena import COLUMNS, COMMIT, collect, decision_claim, decision_track_claim, inventory, record
|
||||
from paper_atlas.io.parquet_metadata import RangeSource, SparseFile, projection
|
||||
|
||||
|
||||
class MemorySource:
|
||||
def __init__(self, raw):
|
||||
self.raw, self.size, self.bytes_read, self.receipts = raw, len(raw), 0, []
|
||||
|
||||
def read(self, start, length):
|
||||
raw = self.raw[start:start+length]
|
||||
self.bytes_read += len(raw)
|
||||
self.receipts.append({'start': start, 'length': length, 'sha256': hashlib.sha256(raw).hexdigest()})
|
||||
return raw
|
||||
|
||||
|
||||
def fixture():
|
||||
import pyarrow as pa
|
||||
import pyarrow.parquet as pq
|
||||
rows = []
|
||||
for i, decision in enumerate(['Accept-Findings', 'Accept-Main', 'Reject', 'Withdrawn', None]):
|
||||
row = {'forum_id': 'unverified-source-'+str(i)+'.xml', 'conference': 'emnlp', 'year': 2023,
|
||||
'track': 'main', 'venue_id': 'source-venue-claim',
|
||||
'paper_number': i, 'title': 'Metadata title '+str(i), 'abstract': 'Metadata abstract\x00 '+str(i),
|
||||
'authors': ['Anonymous'] if i == 2 else ['Ada Author', 'Yi Researcher'],
|
||||
'keywords': ['metadata'], 'venue': 'Source venue label', 'decision': decision,
|
||||
'markdown': 'OCR_FULLTEXT_SENTINEL '+('private-ish fixture body '*10000),
|
||||
'reviews_json': 'REVIEW_FULLTEXT_SENTINEL '+('fixture review '*10000),
|
||||
'decision_comment': 'META_REVIEW_SENTINEL '+('fixture meta-review '*10000)}
|
||||
rows.append(row)
|
||||
sink = io.BytesIO(); pq.write_table(pa.Table.from_pylist(rows), sink, row_group_size=2, compression='snappy')
|
||||
return sink.getvalue(), [{k: v for k, v in r.items() if k in COLUMNS} for r in rows]
|
||||
|
||||
|
||||
class RangeHTTPTests(unittest.TestCase):
|
||||
def response(self, data, status=206, start=2, size=10, etag='"stable"'):
|
||||
response = io.BytesIO(data); response.status = status
|
||||
response.headers = {'Content-Range': f'bytes {start}-{start+len(data)-1}/{size}', 'ETag': etag}
|
||||
return response
|
||||
|
||||
def test_strict_range_and_changed_object_detection(self):
|
||||
opener = unittest.mock.Mock(); opener.open.side_effect = [self.response(b'ab'), self.response(b'cd', start=4, etag='"changed"')]
|
||||
source = RangeSource('https://example.org/public', 10, opener=opener)
|
||||
self.assertEqual(source.read(2, 2), b'ab')
|
||||
self.assertEqual(opener.open.call_args.args[0].get_header('Range'), 'bytes=2-3')
|
||||
with self.assertRaisesRegex(ValueError, 'ETag'):
|
||||
source.read(4, 2)
|
||||
self.assertEqual(source.bytes_read, 2)
|
||||
|
||||
def test_no_full_file_fallback_or_denied_source_retry(self):
|
||||
opener = unittest.mock.Mock(); response = self.response(b'ab', status=200)
|
||||
response.read = unittest.mock.Mock(side_effect=AssertionError('Must not consume full body'))
|
||||
opener.open.return_value = response
|
||||
with self.assertRaisesRegex(ValueError, 'full download forbidden'):
|
||||
RangeSource('https://example.org/public', 10, opener=opener).read(2, 2)
|
||||
response.read.assert_not_called()
|
||||
opener = unittest.mock.Mock(); opener.open.side_effect = urllib.error.HTTPError('https://cdn.example/signed?SECRET=hidden', 403, 'Forbidden', {}, None)
|
||||
with self.assertRaisesRegex(RuntimeError, '^Metadata source HTTP 403; no retry or full-file fallback$'):
|
||||
RangeSource('https://example.org/public', 10, opener=opener).read(2, 2)
|
||||
self.assertEqual(opener.open.call_count, 1)
|
||||
|
||||
def test_sparse_footer_windows_are_synthetic_not_network_reads(self):
|
||||
source = SparseFile(100, [(0, b'PAR1'), (95, b'footer')])
|
||||
source.seek(90); self.assertEqual(source.read(10), b'\x00'*5+b'foote')
|
||||
source.seek(-4, 2); self.assertEqual(source.tell(), 96)
|
||||
with self.assertRaises(ValueError):
|
||||
source.seek(-101, 2)
|
||||
|
||||
|
||||
@unittest.skipUnless(importlib.util.find_spec('pyarrow'), 'Install mirrors extra on the approved remote runner')
|
||||
class ParquetProjectionTests(unittest.TestCase):
|
||||
def test_complete_selected_rows_and_no_excluded_page_requests(self):
|
||||
import pyarrow.parquet as pq
|
||||
raw, expected = fixture(); source = MemorySource(raw); rows, proof = projection(source, COLUMNS)
|
||||
self.assertEqual(rows, expected)
|
||||
metadata = pq.ParquetFile(io.BytesIO(raw)).metadata
|
||||
excluded = []
|
||||
for i in range(metadata.num_row_groups):
|
||||
group = metadata.row_group(i)
|
||||
for j in range(group.num_columns):
|
||||
col = group.column(j)
|
||||
if col.path_in_schema.split('.')[0] not in COLUMNS:
|
||||
start = min(x for x in [col.data_page_offset, col.dictionary_page_offset] if x is not None and x > 0)
|
||||
excluded.append((start, start+col.total_compressed_size))
|
||||
for request in source.receipts:
|
||||
a, b = request['start'], request['start']+request['length']
|
||||
self.assertFalse(any(max(a, c) < min(b, d) for c, d in excluded))
|
||||
self.assertFalse(proof['raw_parquet_saved']); self.assertFalse(proof['full_file_checksum_verified'])
|
||||
self.assertLess(source.bytes_read, len(raw))
|
||||
encoded = json.dumps(rows)+json.dumps(proof)
|
||||
self.assertNotIn('OCR_FULLTEXT_SENTINEL', encoded); self.assertNotIn('REVIEW_FULLTEXT_SENTINEL', encoded)
|
||||
self.assertNotIn('META_REVIEW_SENTINEL', encoded)
|
||||
|
||||
def test_wrong_schema_and_footer_fail_before_any_payload_download(self):
|
||||
raw, _ = fixture(); source = MemorySource(raw)
|
||||
with self.assertRaisesRegex(ValueError, 'Missing'):
|
||||
projection(source, {'missing_column': 'string'})
|
||||
self.assertEqual(len(source.receipts), 3)
|
||||
source = MemorySource(raw)
|
||||
with self.assertRaisesRegex(ValueError, 'type differs'):
|
||||
projection(source, {**COLUMNS, 'year': 'string'})
|
||||
self.assertEqual(len(source.receipts), 3)
|
||||
bad = bytearray(raw); bad[-8:-4] = struct.pack('<I', len(raw))
|
||||
source = MemorySource(bytes(bad))
|
||||
with self.assertRaisesRegex(ValueError, 'footer'):
|
||||
projection(source, COLUMNS)
|
||||
self.assertEqual(len(source.receipts), 2)
|
||||
|
||||
def test_collector_output_preserves_claims_without_original_identity(self):
|
||||
raw, _ = fixture(); item = {'path': 'data/emnlp-00000-of-00001.parquet', 'size': len(raw), 'lfs': {'oid': 'a'*64}}
|
||||
with tempfile.TemporaryDirectory() as temp:
|
||||
out = Path(temp)/'projected'
|
||||
with patch('paper_atlas.collectors.reviewarena.inventory', return_value=[item]):
|
||||
report = collect('emnlp', out, source_factory=lambda url, size: MemorySource(raw))
|
||||
self.assertEqual(report['mirror_decision_counts'], {'accepted': 2, 'rejected': 1, 'withdrawn': 1, 'unknown': 1})
|
||||
self.assertEqual(report['records'], 5); self.assertFalse(report['database_written'])
|
||||
projected = [json.loads(line) for line in next(out.glob('*.jsonl')).read_text().splitlines()]
|
||||
for row in projected:
|
||||
self.assertEqual(row['identifiers'], {})
|
||||
self.assertFalse(row['metadata']['original_forum_id_verified'])
|
||||
self.assertFalse(row['metadata']['original_decision_signature_verified'])
|
||||
self.assertNotIn('status', row)
|
||||
self.assertIn(COMMIT, row['external_id'])
|
||||
self.assertEqual(projected[2]['authors'], ['Anonymous'])
|
||||
self.assertEqual(projected[2]['metadata']['forum_id_claim'], 'unverified-source-2.xml')
|
||||
self.assertEqual(projected[0]['track'], 'main')
|
||||
self.assertEqual(projected[0]['mirror_decision_track_claim'], 'findings')
|
||||
all_saved=''.join(p.read_text() for p in out.iterdir())
|
||||
self.assertNotIn('FULLTEXT_SENTINEL', all_saved); self.assertNotIn('META_REVIEW_SENTINEL', all_saved)
|
||||
with self.assertRaisesRegex(ValueError, 'New output'):
|
||||
collect('emnlp', out)
|
||||
|
||||
def test_offline_replay_reclassifies_literals_preserving_all_raw_fields_without_network(self):
|
||||
raw, _ = fixture(); item = {'path': 'data/emnlp-00000-of-00001.parquet', 'size': len(raw), 'lfs': {'oid': 'a'*64}}
|
||||
with tempfile.TemporaryDirectory() as temp, patch('paper_atlas.collectors.reviewarena.inventory', return_value=[item]):
|
||||
out = Path(temp)/'v1'; collect('emnlp', out, source_factory=lambda url, size: MemorySource(raw))
|
||||
file, = out.glob('*.jsonl'); before = [json.loads(x) for x in file.read_text().splitlines()]
|
||||
# Mimic the first reviewed source version: new literal spellings were unknown.
|
||||
for row in before[:2]:
|
||||
row['mirror_decision_category'] = 'unknown'; row.pop('mirror_decision_track_claim')
|
||||
file.write_text(''.join(json.dumps(x)+'\n' for x in before))
|
||||
report_path = out/'collection-report.json'; report = json.loads(report_path.read_text())
|
||||
report['files'][0]['file_sha256'] = hashlib.sha256(file.read_bytes()).hexdigest()
|
||||
report_path.write_text(json.dumps(report)); report_hash = hashlib.sha256(report_path.read_bytes()).hexdigest()
|
||||
no_network = unittest.mock.Mock(side_effect=AssertionError('Offline replay must not contact source'))
|
||||
result = collect('emnlp', Path(temp)/'v2', from_projection=report_path, projection_sha256=report_hash, source_factory=no_network)
|
||||
no_network.assert_not_called(); self.assertEqual(result['metadata_bytes_read'], 0)
|
||||
self.assertEqual(result['source_projection_bytes'], report['metadata_bytes_read'])
|
||||
self.assertEqual(result['mirror_decision_counts']['accepted'], 2)
|
||||
self.assertEqual(result['mirror_decision_track_claim_counts'], {'findings': 1, 'main': 1, 'unspecified': 3})
|
||||
new = [json.loads(x) for x in next((Path(temp)/'v2').glob('*.jsonl')).read_text().splitlines()]
|
||||
self.assertEqual([r['raw_fields'] for r in new], [r['raw_fields'] for r in before])
|
||||
self.assertEqual([r['source_locator'] for r in new], [r['source_locator'] for r in before])
|
||||
with self.assertRaisesRegex(ValueError, 'hash differs'):
|
||||
collect('emnlp', Path(temp)/'bad', from_projection=report_path, projection_sha256='0'*64, source_factory=no_network)
|
||||
no_network.assert_not_called()
|
||||
|
||||
def test_failure_is_retained_without_source_payload_or_retries(self):
|
||||
item = {'path': 'data/emnlp-00000-of-00001.parquet', 'size': 50, 'lfs': {'oid': 'a'*64}}
|
||||
with tempfile.TemporaryDirectory() as temp:
|
||||
out = Path(temp)/'failed'
|
||||
with patch('paper_atlas.collectors.reviewarena.inventory', return_value=[item]), patch('paper_atlas.collectors.reviewarena.projection', side_effect=RuntimeError('private body example')):
|
||||
with self.assertRaises(RuntimeError):
|
||||
collect('emnlp', out, source_factory=lambda url, size: None)
|
||||
report = json.loads((out/'failure.json').read_text())
|
||||
self.assertEqual(report['status'], 'incomplete_do_not_import')
|
||||
self.assertTrue(report['no_automatic_retry']); self.assertNotIn('private body', (out/'failure.json').read_text())
|
||||
|
||||
|
||||
class MirrorSemanticsTests(unittest.TestCase):
|
||||
def test_pinned_inventory_covers_all_seven_families_and_fifteen_shards(self):
|
||||
splits = ['colm', 'corl', 'emnlp', 'iclr', 'icml', 'neurips', 'tmlr']
|
||||
result = {s: inventory(s) for s in splits}
|
||||
self.assertEqual(sum(len(v) for v in result.values()), 15)
|
||||
self.assertEqual([len(result[s]) for s in splits], [1, 1, 1, 5, 1, 5, 1])
|
||||
self.assertEqual(result['emnlp'][0]['size'], 90794691)
|
||||
self.assertEqual(result['emnlp'][0]['lfs']['oid'], 'e9e3aab42d28ca32a02287ed0bd0f1f9e1f3af5cb8ed8d6a9bf28d4f630773da')
|
||||
with self.assertRaises(ValueError):
|
||||
inventory('uai')
|
||||
|
||||
def test_missing_or_ambiguous_decision_is_never_inferred_from_venue(self):
|
||||
for literal in ['', None, 'Under review', 'Final venue pending', 'Accept maybe', 'Reject?']:
|
||||
self.assertEqual(decision_claim(literal), 'unknown')
|
||||
self.assertEqual(decision_claim('Desk Rejected'), 'desk_rejected')
|
||||
self.assertEqual(decision_claim('Withdrawn'), 'withdrawn')
|
||||
self.assertEqual(decision_claim('Accept (Findings)'), 'accepted')
|
||||
self.assertEqual(decision_claim('Accept-Main'), 'accepted')
|
||||
self.assertEqual(decision_claim('Accept-Findings'), 'accepted')
|
||||
self.assertEqual(decision_track_claim('Accept-Findings'), 'findings')
|
||||
self.assertIsNone(decision_track_claim('Accept (Poster)'))
|
||||
@@ -0,0 +1,209 @@
|
||||
"""Whole-source scope correction, stale refresh, full history and atomicity on both engines."""
|
||||
from contextlib import closing, redirect_stdout
|
||||
import copy
|
||||
import hashlib
|
||||
import io
|
||||
import json
|
||||
import os
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
from unittest.mock import patch
|
||||
import uuid
|
||||
|
||||
from paper_atlas.db.backend import open_store
|
||||
from paper_atlas.db.edition_scopes import KEY
|
||||
from paper_atlas.db.source_revisions import available, require_row, require_manifest
|
||||
from paper_atlas.db.store import Store
|
||||
from paper_atlas.ingestion.source_revisions import proposal, apply_plan, run, validate
|
||||
from paper_atlas.ingestion.event_identity import database_identity, digest, values
|
||||
from paper_atlas.ingestion.records import coverage_options, ingest_file
|
||||
from paper_atlas.postgres.client import connect
|
||||
from paper_atlas.postgres.migration import migrate
|
||||
from paper_atlas.postgres.schema import quote
|
||||
|
||||
URL = 'https://example.org/volume'
|
||||
|
||||
|
||||
def seed(path):
|
||||
old = []
|
||||
for i in range(4):
|
||||
doi = '10.1234/publication'+str(i)
|
||||
old.append({'source': 'fixture', 'venue': 'MM', 'year': 2026, 'track': 'main' if i == 0 else 'unclassified',
|
||||
'external_id': doi, 'url': 'https://doi.org/'+doi, 'title': 'Distinct publication '+str(i),
|
||||
'authors': ['Alice Author', 'Bob Writer'], 'abstract': None if i == 1 else 'Source abstract '+str(i),
|
||||
'identifiers': {'doi': doi}, 'submission_identifiers': {'publication_doi': doi},
|
||||
'status': 'accepted', 'raw_decision': 'Publication observation', 'evidence_url': URL,
|
||||
'observed_at': '2026-06-01T00:00:00Z', 'tags': [{'namespace': 'fixture:track', 'value': 'main' if i == 0 else 'unclassified'}],
|
||||
'metadata': {'original': i, 'program': None, 'scientific_work_identity_approved': False}})
|
||||
current = copy.deepcopy(old)
|
||||
current[1].update(track='main', abstract='Reviewed program abstract with NUL\x00 and SOH\x01')
|
||||
current[1]['metadata']['program'] = {'id': 'p1', 'evidence_url': URL}
|
||||
current[1]['tags'][0]['value'] = 'main'
|
||||
store = Store(path)
|
||||
for row in old[:3]:
|
||||
p, r = store.record(row['source'], row['external_id'], row['url'], row['title'], row['abstract'], row['authors'], row['identifiers'], row['metadata'])
|
||||
store.decision(p, r, 'MM', 2026, row['track'], row['external_id'], URL, 'accepted', row['raw_decision'],
|
||||
identities=[('publication_doi', row['external_id'])], observed_at=row['observed_at'])
|
||||
store.record_tags(p, r, 'fixture', row['tags'])
|
||||
if row == old[1]:
|
||||
store.conn.execute('UPDATE papers SET abstract=? WHERE id=?', ('Canonical preprint abstract', p))
|
||||
p, other = store.record('prior', 'old-forum', 'https://example.org/prior', row['title'], 'Canonical preprint abstract', row['authors'], row['identifiers'])
|
||||
store.decision(p, other, 'PRIOR', 2025, 'main', 'old-forum', 'https://example.org/prior', 'rejected', 'Reject', observed_at='2025-01-01T00:00:00Z')
|
||||
# An unrelated source already has an attempt in the populated target edition.
|
||||
p, r = store.record('other', 'other', 'https://example.org/other', 'Unrelated target paper', 'Other abstract', ['Someone Else'])
|
||||
store.decision(p, r, 'MM', 2026, 'main', 'other', 'https://example.org/other', 'accepted', 'Published')
|
||||
sid = store.conn.execute("SELECT id FROM sources WHERE name='fixture'").fetchone()[0]
|
||||
for track, expected, imported in [('main', 1, 1), ('unclassified', 3, 2)]:
|
||||
eid = store.conn.execute('SELECT id FROM editions WHERE year=2026 AND track=?', (track,)).fetchone()[0]
|
||||
store.conn.execute("INSERT INTO edition_sources VALUES(?,?,?,'partial',?,?,?,'2026-06-01',NULL)", (eid, sid, URL, expected, imported, 'checkpoint'))
|
||||
store.conn.commit(); store.conn.close()
|
||||
return {'format': 1, 'source': 'fixture', 'venue': 'MM', 'year': 2026, 'event_key': '', 'list_url': URL,
|
||||
'previous_rows': old, 'rows': current, 'evidence': [{'url': URL, 'sha256': 'a'*64}]}
|
||||
|
||||
|
||||
class RevisionCases:
|
||||
def store(self):
|
||||
return open_store(**self.target)
|
||||
|
||||
def plan(self, c):
|
||||
return proposal(c, self.inventory, database_identity(c, self.target.get('db')))
|
||||
|
||||
def test_move_to_populated_target_preserves_full_history_and_before_images(self):
|
||||
with closing(self.store().conn) as c:
|
||||
plan = self.plan(c); self.assertEqual(len(plan['changes']), 1)
|
||||
self.assertEqual(len(plan['before']['packet']['submissions']), 5)
|
||||
before_papers = plan['before']['packet']['papers']
|
||||
before_tags = plan['before']['packet']['paper_tags']
|
||||
result = apply_plan(c, plan); c.commit()
|
||||
self.assertEqual((result['events_moved'], result['records_changed']), (1, 1))
|
||||
self.assertEqual(result['protected_decisions'], 5)
|
||||
self.assertEqual(values(c.execute('SELECT * FROM papers ORDER BY id')), before_papers)
|
||||
self.assertEqual(values(c.execute('SELECT * FROM paper_tags ORDER BY paper_id,tag_id,evidence_key')), before_tags)
|
||||
self.assertEqual(c.execute("SELECT count(*) FROM decisions WHERE status='rejected'").fetchone()[0], 1)
|
||||
self.assertEqual(values(c.execute('SELECT status,expected_count,imported_count,cursor,last_success_at,last_error FROM edition_sources ORDER BY edition_id')),
|
||||
[['partial', 2, 2, 'checkpoint', '2026-06-01', None], ['partial', 2, 1, 'checkpoint', '2026-06-01', None]])
|
||||
saved = json.loads(c.execute('SELECT plan_json FROM source_scope_revisions').fetchone()[0])
|
||||
self.assertEqual(saved, plan)
|
||||
self.assertEqual(apply_plan(c, plan)['status'], 'already_applied')
|
||||
self.assertEqual(c.execute('SELECT count(*) FROM source_scope_revisions').fetchone()[0], 1)
|
||||
|
||||
def test_stale_direct_store_and_jsonl_refresh_fail_but_current_refresh_succeeds(self):
|
||||
with closing(self.store().conn) as c:
|
||||
plan = self.plan(c); apply_plan(c, plan); c.commit()
|
||||
with closing(self.store().conn) as c:
|
||||
store = object.__new__(Store); store.conn = c; store.event_identity = True
|
||||
old = self.inventory['previous_rows'][1]
|
||||
with self.assertRaisesRegex(ValueError, 'Stale source'):
|
||||
store.record('fixture', old['external_id'], old['url'], old['title'], old['abstract'], old['authors'], old['identifiers'], old['metadata'])
|
||||
c.rollback()
|
||||
row = self.inventory['rows'][1]
|
||||
p, r = store.record('fixture', row['external_id'], row['url'], row['title'], row['abstract'], row['authors'], row['identifiers'], row['metadata'])
|
||||
self.assertEqual(c.execute('SELECT abstract FROM papers WHERE id=?', (p,)).fetchone()[0], 'Canonical preprint abstract')
|
||||
self.assertEqual(json.loads(c.execute('SELECT metadata_json FROM source_records WHERE id=?', (r,)).fetchone()[0])[KEY][0]['target']['track'], 'main')
|
||||
c.rollback()
|
||||
path = self.root/'old.jsonl'; raw = (json.dumps(old)+'\n').encode(); path.write_bytes(raw)
|
||||
path.with_suffix('.manifest.json').write_text(json.dumps({'source': 'fixture', 'venue': 'MM', 'year': 2026,
|
||||
'track': 'unclassified', 'list_url': URL, 'status': 'partial', 'count': 1, 'file_sha256': hashlib.sha256(raw).hexdigest()}))
|
||||
with self.assertRaisesRegex(ValueError, 'source snapshot'):
|
||||
ingest_file(self.target.get('db'), path, **{k: v for k, v in self.target.items() if k != 'db'})
|
||||
|
||||
def test_unregistered_member_and_changed_program_metadata_are_rejected(self):
|
||||
with closing(self.store().conn) as c:
|
||||
apply_plan(c, self.plan(c)); c.commit()
|
||||
sid = c.execute("SELECT id FROM sources WHERE name='fixture'").fetchone()[0]
|
||||
cache = {}; unknown = copy.deepcopy(self.inventory['rows'][0]); unknown['external_id'] = '10.1234/unlisted'
|
||||
with self.assertRaisesRegex(ValueError, 'absent from current'):
|
||||
require_row(c, sid, unknown, {'list_url': URL}, cache)
|
||||
bad = copy.deepcopy(self.inventory['rows'][1]); bad['metadata']['program'] = None
|
||||
with self.assertRaisesRegex(ValueError, 'Stale source'):
|
||||
require_row(c, sid, bad, {'list_url': URL}, cache)
|
||||
good = copy.deepcopy(self.inventory['rows'][1]); good['metadata']['verified_work_anchor'] = {'reason': 'extra evidence'}
|
||||
require_row(c, sid, good, {'list_url': URL}, cache)
|
||||
with self.assertRaisesRegex(ValueError, 'coverage manifest'):
|
||||
require_manifest(c, sid, {'venue': 'MM', 'year': 2026, 'track': 'main', 'list_url': URL, 'count': 1, 'status': 'partial'}, cache)
|
||||
require_manifest(c, sid, {'venue': 'MM', 'year': 2026, 'track': 'main', 'list_url': URL, 'count': 2, 'status': 'partial'}, cache)
|
||||
|
||||
def test_stale_plan_and_target_identifier_collision_prevent_writes(self):
|
||||
with closing(self.store().conn) as c:
|
||||
plan = self.plan(c)
|
||||
c.execute("UPDATE decisions SET raw_value='Additional evidence' WHERE status='rejected'")
|
||||
with self.assertRaisesRegex(ValueError, 'Stale'):
|
||||
apply_plan(c, plan)
|
||||
c.rollback()
|
||||
move = plan['changes'][0]
|
||||
target_sub = c.execute('SELECT id FROM submissions WHERE edition_id=? LIMIT 1', (move['target_edition_id'],)).fetchone()[0]
|
||||
c.execute('INSERT INTO submission_identifiers VALUES(?,?,?,?)', (move['target_edition_id'], 'publication_doi', move['external_id'], target_sub))
|
||||
with self.assertRaisesRegex(ValueError, 'collision'):
|
||||
self.plan(c)
|
||||
c.rollback()
|
||||
self.assertFalse(available(c))
|
||||
|
||||
def test_failed_postcheck_rolls_back_rows_extension_and_coverage(self):
|
||||
with closing(self.store().conn) as c:
|
||||
plan = self.plan(c)
|
||||
with patch('paper_atlas.ingestion.source_revisions.verify', side_effect=ValueError('Injected verification failure')):
|
||||
with self.assertRaisesRegex(ValueError, 'Injected'):
|
||||
apply_plan(c, plan)
|
||||
c.rollback()
|
||||
self.assertFalse(available(c))
|
||||
self.assertEqual(self.plan(c), plan)
|
||||
|
||||
def test_cli_binds_database_and_file_hash_without_writing_during_preflight(self):
|
||||
inventory = self.root/'inventory.json'; inventory.write_text(json.dumps(self.inventory))
|
||||
path = self.root/'plan.json'
|
||||
args = {k: v for k, v in self.target.items() if k != 'db'}
|
||||
run(self.target.get('db'), path, inventory=inventory, **args)
|
||||
with closing(self.store().conn) as c:
|
||||
self.assertFalse(available(c))
|
||||
with self.assertRaisesRegex(ValueError, 'hash'):
|
||||
run(self.target.get('db'), self.root/'bad.json', plan=path, plan_sha256='0'*64, **args)
|
||||
report = run(self.target.get('db'), self.root/'applied.json', plan=path, plan_sha256=hashlib.sha256(path.read_bytes()).hexdigest(), **args)
|
||||
self.assertEqual(report['events_moved'], 1)
|
||||
|
||||
def test_membership_identity_changes_and_incremental_shrink_remain_forbidden(self):
|
||||
for mutate in [lambda x: x['rows'].pop(), lambda x: x['rows'][0].update(title='Different work'),
|
||||
lambda x: x['rows'][0].update(status='rejected')]:
|
||||
bad = copy.deepcopy(self.inventory); mutate(bad)
|
||||
with self.assertRaises(ValueError):
|
||||
validate(bad)
|
||||
with self.assertRaisesRegex(ValueError, 'incremental'):
|
||||
coverage_options({'coverage_mode': 'incremental', 'venue': 'MM', 'year': 2026, 'track': 'unclassified',
|
||||
'status': 'partial', 'count': 1, 'coverage_previous_count': 2,
|
||||
'coverage_previous_expected_count': 3, 'coverage_expected_count': 2})
|
||||
|
||||
|
||||
class SQLiteSourceRevisionTests(RevisionCases, unittest.TestCase):
|
||||
def setUp(self):
|
||||
temp = tempfile.TemporaryDirectory(); self.addCleanup(temp.cleanup); self.root = Path(temp.name)
|
||||
db = self.root/'source.sqlite'; self.inventory = seed(db); self.target = {'db': db}
|
||||
|
||||
@unittest.skipUnless(os.environ.get('PAPER_ATLAS_TEST_PG_SERVICE'), 'Requires approved remote PostgreSQL service')
|
||||
def test_optional_revision_table_migrates_with_full_history(self):
|
||||
with closing(self.store().conn) as c:
|
||||
plan = self.plan(c); apply_plan(c, plan); c.commit()
|
||||
service = os.environ['PAPER_ATLAS_TEST_PG_SERVICE']; schema = 'pa_rev_copy_'+uuid.uuid4().hex[:10]
|
||||
self.addCleanup(drop_schema, service, schema)
|
||||
with redirect_stdout(io.StringIO()):
|
||||
migrate(self.target['db'], service, self.root/'copied.json', schema=schema, min_free_gib=0)
|
||||
with closing(open_store(pg_service=service, pg_schema=schema).conn) as c:
|
||||
self.assertEqual(json.loads(c.execute('SELECT plan_json FROM source_scope_revisions').fetchone()[0]), plan)
|
||||
self.assertEqual(c.execute('SELECT count(*) FROM source_scope_revisions').fetchone()[0], 1)
|
||||
|
||||
|
||||
def drop_schema(service, schema):
|
||||
with connect(service) as c:
|
||||
for name, in c.execute('SELECT nspname FROM pg_namespace WHERE starts_with(nspname,%s)', (schema,)):
|
||||
c.execute('DROP SCHEMA '+quote(name)+' CASCADE')
|
||||
|
||||
|
||||
@unittest.skipUnless(os.environ.get('PAPER_ATLAS_TEST_PG_SERVICE'), 'Requires approved remote PostgreSQL service')
|
||||
class PostgreSQLSourceRevisionTests(RevisionCases, unittest.TestCase):
|
||||
def setUp(self):
|
||||
temp = tempfile.TemporaryDirectory(); self.addCleanup(temp.cleanup); self.root = Path(temp.name)
|
||||
db = self.root/'source.sqlite'; self.inventory = seed(db)
|
||||
service = os.environ['PAPER_ATLAS_TEST_PG_SERVICE']; schema = 'pa_rev_'+uuid.uuid4().hex[:12]
|
||||
self.addCleanup(drop_schema, service, schema)
|
||||
with redirect_stdout(io.StringIO()):
|
||||
migrate(db, service, self.root/'migration.json', schema=schema, min_free_gib=0)
|
||||
self.target = {'pg_service': service, 'pg_schema': schema}
|
||||
Reference in New Issue
Block a user