Compare commits

..

1 Commits

Author SHA1 Message Date
codex-bot 4d97822d9e Preserve confirmed single-name Springer authors
Metadata integrity checks / metadata (push) Successful in 8m28s
2026-11-18 20:30:11 +08:00
5 changed files with 59 additions and 29 deletions
+2 -2
View File
@@ -9,8 +9,8 @@ paper-atlas collect eccv2026-publications \
--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。数字上标的两种数学定界符可按同一标题核对,章节标题和目录原文分别保存。已有输出目录需另查回执或换新空目录后运行。
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。
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。单名署名须由同页的结构化作者列表按顺序确认,确认后按原名保存为字符串,引用元数据仍保留原文。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。
@@ -102,20 +102,10 @@ def publisher_inventory(root):
return entries
def title_for_comparison(title):
# Springer may render a numeric superscript with inline or display math delimiters.
return clean(re.sub(r'\\\((\^[0-9]+)\\\)|\$\$(\^[0-9]+)\$\$',
lambda match: match[1] or match[2], title))
def publisher_row(chapter, toc, evidence):
doi = chapter['doi']
literal_title_differs = clean(chapter['title']) != toc['title']
if literal_title_differs and title_for_comparison(chapter['title']) != title_for_comparison(toc['title']):
if clean(chapter['title']) != toc['title']:
raise ValueError('Chapter title differs from Springer contents, review the version')
contents_evidence = {'url': toc['evidence_url'], 'sha256': toc['evidence_sha256']}
if literal_title_differs:
contents_evidence['title_literal'] = toc['title']
url = ORIGIN + '/chapter/' + doi
return {
'source': SOURCE, 'external_id': doi, 'url': url,
@@ -130,7 +120,7 @@ def publisher_row(chapter, toc, evidence):
'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': contents_evidence,
'publisher_contents_evidence': {'url': toc['evidence_url'], 'sha256': toc['evidence_sha256']},
'scope': 'Refereed conference proceedings, LNCS 17001–17078.',
'original_decision_signature_verified': False,
},
+31
View File
@@ -54,6 +54,27 @@ def read_page(index, url):
return raw, evidence
def require_display_authors(nodes, authors):
matching_article_lists = []
for node in nodes:
if node.tag != 'script' or node.attrs.get('type') != 'application/ld+json':
continue
try:
article = json.loads(''.join(node.children))
except json.JSONDecodeError:
continue
if not isinstance(article, dict) or article.get('@type') != 'ScholarlyArticle':
continue
entries = article.get('author')
if not isinstance(entries, list) or not all(
isinstance(entry, dict) and isinstance(entry.get('name'), str) for entry in entries
):
raise ValueError('Unreviewed Springer article author list')
matching_article_lists.append([clean(entry['name']) for entry in entries])
if matching_article_lists != [authors]:
raise ValueError('Single-name author requires matching Springer display author list')
def parse_chapter(html, doi, *, book_title, publication_year):
nodes = list(Document(html).root.walk())
meta = defaultdict(list)
@@ -73,12 +94,22 @@ def parse_chapter(html, doi, *, book_title, publication_year):
raise ValueError('Missing named chapter authors')
authors = []
for raw in raw_authors:
if ',' not in raw:
name = clean(raw)
if len(name.split()) != 1 or not any(character.isalpha() for character in name):
raise ValueError('Unreviewed citation author format')
authors.append(name)
continue
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})
if any(isinstance(author, str) for author in authors):
display_names = [author if isinstance(author, str) else author['given']+' '+author['family']
for author in authors]
require_display_authors(nodes, display_names)
abstract_nodes = [n for n in nodes if n.attrs.get('id') == 'Abs1-content']
if len(abstract_nodes) > 1:
raise ValueError('Ambiguous abstract section')
+1 -14
View File
@@ -6,7 +6,7 @@ import sqlite3
import tempfile
import unittest
from paper_atlas.collectors.eccv2026_publications import BOOK_TITLE, SOURCE, collect, publisher_row
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
@@ -114,19 +114,6 @@ class ECCV2026PublicationsTests(unittest.TestCase):
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)
+23 -1
View File
@@ -9,7 +9,7 @@ import tempfile
import unittest
from paper_atlas.collectors.springer_selected import (
BOOK, SOURCE, collect, parse_chapter, parse_toc, read_page, snapshot_index,
BOOK, SCOPE, SOURCE, collect, parse_chapter, parse_toc, read_page, snapshot_index,
)
from paper_atlas.db.store import Store
from paper_atlas.ingestion.records import ingest_file
@@ -153,6 +153,28 @@ class SpringerSelectedTests(unittest.TestCase):
html = self.fixture['chapters']['chapter-01.html'].replace('id="Abs1-content"','id="not-an-abstract"')
self.assertIsNone(parse_chapter(html, BOOK+'_1')['abstract'])
def test_single_name_preserves_literal_and_requires_ordered_display_confirmation(self):
html = f"""
<meta name="citation_doi" content="{BOOK}_1">
<meta name="citation_inbook_title" content="{SCOPE['book_title']}">
<meta name="citation_publication_date" content="{SCOPE['publication_year']}">
<meta name="citation_title" content="Example">
<meta name="citation_author" content="Rajkumar">
<meta name="citation_author" content="Doe, Jane">
<meta name="citation_firstpage" content="1">
<meta name="citation_lastpage" content="5">
"""
article = {'@type':'ScholarlyArticle', 'author':[{'name':'Rajkumar'}, {'name':'Jane Doe'}]}
script = '<script type="application/ld+json">{}</script>'
chapter = parse_chapter(html+script.format(json.dumps(article)), BOOK+'_1')
self.assertEqual(chapter['authors'], ['Rajkumar', {'family':'Doe', 'given':'Jane'}])
self.assertEqual(chapter['authors_citation_raw'], ['Rajkumar','Doe, Jane'])
with self.assertRaisesRegex(ValueError, 'matching Springer display author list'):
parse_chapter(html, BOOK+'_1')
article['author'].reverse()
with self.assertRaisesRegex(ValueError, 'matching Springer display author list'):
parse_chapter(html+script.format(json.dumps(article)), BOOK+'_1')
def test_toc_ignores_frontmatter_but_rejects_duplicate_publications(self):
html = self.fixture['toc']['springer-toc1.html']
self.assertEqual(len(parse_toc(html)), 17)