Files
paper-atlas/tests/test_eccv2026_publications.py
T
codex-bot bc01f90e70
Metadata integrity checks / metadata (push) Successful in 12m44s
Preserve Springer title literals across superscript formatting
2026-11-18 19:25:00 +08:00

138 lines
7.3 KiB
Python

"""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, publisher_row
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_numeric_superscript_wrappers_preserve_both_title_literals(self):
chapter = {'doi': FIRST_DOI, 'title': 'Example $$^2$$ : Formal title',
'authors': [], 'abstract': 'Published abstract.', 'authors_citation_raw': [],
'first_page': '1', 'last_page': '10'}
toc = {'title': r'Example \(^2\) : Formal title', 'book_doi': FIRST_BOOK, 'lncs_volume': 17001,
'evidence_url': 'https://link.springer.com/book/' + FIRST_BOOK, 'evidence_sha256': 'a' * 64}
evidence = {'observed_at': '2026-11-18T00:00:00Z'}
row = publisher_row(chapter, toc, evidence)
self.assertEqual(row['title'], chapter['title'])
self.assertEqual(row['metadata']['publisher_contents_evidence']['title_literal'], toc['title'])
with self.assertRaisesRegex(ValueError, 'title differs'):
publisher_row(chapter, {**toc, 'title': r'Example \(^3\) : Formal title'}, evidence)
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())