ff03443af4
Three papers lacked abstracts and 1911.00966 was fully degraded (OpenAlex 404 -> empty `Paper(id, title="")` stub). Recovery sources verified, then wired in: - arxiv.fetch_metadata: resolve an arXiv id to a Paper via the Atom export API (authoritative for preprints). Used as an ingest fallback on OpenAlex 404, replacing the empty stub. - crossref.py (new): fetch_abstract(doi), JATS-stripped, Crossref Polite Pool. - semanticscholar.fetch_abstract: fetch just the abstract field. - ingest.py: _recover_abstract() supplements an empty abstract from S2 then Crossref *before* embedding, so the paper gets a real (non-zero) vector. Live DB backfill: 1911.00966 fully recovered from arXiv (bibkey PinkallSpringborn2019, 384-char abstract, its 10 chunks now visible to chunk search); 10.1007/s00454-019-00132-8 abstract from S2 (1186 chars). Book chapter 10.1007/978-3-642-17413-1_7 has no abstract in OpenAlex/S2/Crossref -> documented limit. Corpus now has 1 paper without an abstract. Also documented DQ-5 (not fixed): ingest_paper is not idempotent for DOI papers (openalex returns full-URL id vs the bare canonical id) -> re-ingest trips papers_openalex_id_key. Blocks roadmap R-A/R-C; needs a careful _map_paper fix. Tests: arxiv/crossref/s2 fetchers + ingest recovery paths (342 passing). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
405 lines
16 KiB
Python
405 lines
16 KiB
Python
"""End-to-end idempotent ingest pipeline for a single paper."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import logging
|
|
from dataclasses import dataclass
|
|
from pathlib import Path
|
|
|
|
import numpy as np
|
|
|
|
from codex.config import get_settings
|
|
from codex.db import get_conn
|
|
from codex.embed import get_embedder
|
|
from codex.models import Citation, Paper
|
|
from codex.quality import classify_section, filter_chunks
|
|
from codex.sources import arxiv, crossref, openalex, semanticscholar
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
@dataclass
|
|
class IngestResult:
|
|
paper_id: str
|
|
chunks_upserted: int
|
|
citations_upserted: int
|
|
formulas_upserted: int = 0
|
|
figures_upserted: int = 0
|
|
|
|
|
|
def _make_bibkey(paper: Paper) -> str | None:
|
|
"""Generate a BibTeX-style key from paper metadata.
|
|
|
|
Format: ConcatSurnames + Year (e.g. "BobenkoPinkallSpringborn2015").
|
|
Surnames are the last whitespace-separated token of each author name.
|
|
>3 authors: FirstSurname + "EtAl" + Year.
|
|
Returns None when authors or year are missing.
|
|
"""
|
|
if not paper.authors or not paper.year:
|
|
return None
|
|
surnames = [name.split()[-1] for name in paper.authors if name.strip()]
|
|
if not surnames:
|
|
return None
|
|
year = str(paper.year)
|
|
if len(surnames) <= 3:
|
|
return "".join(surnames) + year
|
|
return surnames[0] + "EtAl" + year
|
|
|
|
|
|
def _s2_id_for(pid: str) -> str | None:
|
|
"""Format a canonical paper id for the Semantic Scholar paper endpoint.
|
|
|
|
S2 needs a namespaced id (``arXiv:<id>``, ``DOI:<doi>``) or a bare 40-hex
|
|
S2 paperId — a bare arXiv id like ``2305.10988`` 404s. DOIs are detected by
|
|
the ``10.`` prefix; everything else is assumed to be an arXiv id (modern
|
|
``2305.10988`` or legacy ``math/0603097``).
|
|
"""
|
|
p = (pid or "").strip()
|
|
if not p:
|
|
return None
|
|
if p.lower().startswith(("arxiv:", "doi:")):
|
|
return p
|
|
if p.startswith("10."):
|
|
return f"DOI:{p}"
|
|
return f"arXiv:{p}"
|
|
|
|
|
|
def _norm_cited_id(cited_id: str) -> str:
|
|
"""Lowercase DOI cited-ids (DOIs are case-insensitive) so the graph join
|
|
does not fragment the same reference into distinct case-variant nodes.
|
|
arXiv ids and S2 paperIds are left untouched."""
|
|
return cited_id.lower() if cited_id.startswith("10.") else cited_id
|
|
|
|
|
|
def _recover_abstract(paper: Paper) -> str | None:
|
|
"""Recover a missing abstract from S2, then Crossref (DQ-2 supplement).
|
|
|
|
Tried in order of coverage for this corpus: Semantic Scholar (indexes most
|
|
arXiv + journal abstracts), then Crossref (publisher-deposited, DOI only).
|
|
Network/parse failures degrade to None rather than aborting the ingest.
|
|
"""
|
|
s2_id = _s2_id_for(paper.id)
|
|
if s2_id is not None:
|
|
try:
|
|
abstract = semanticscholar.fetch_abstract(s2_id)
|
|
if abstract and abstract.strip():
|
|
return abstract.strip()
|
|
except Exception:
|
|
logger.warning("S2 abstract recovery failed for %s", paper.id, exc_info=True)
|
|
if paper.id.startswith("10."):
|
|
try:
|
|
abstract = crossref.fetch_abstract(paper.id)
|
|
if abstract and abstract.strip():
|
|
return abstract.strip()
|
|
except Exception:
|
|
logger.warning("Crossref abstract recovery failed for %s", paper.id, exc_info=True)
|
|
return None
|
|
|
|
|
|
def _s2_reference_supplement(paper: Paper) -> list[Citation]:
|
|
"""Fetch a paper's references from Semantic Scholar (DQ-1 supplement).
|
|
|
|
Used when OpenAlex returns an empty ``referenced_works`` list — common for
|
|
arXiv preprints and institutional theses that OpenAlex indexes without a
|
|
parsed bibliography. ``citing_id`` is rewritten to the canonical ``paper.id``
|
|
so the FK holds, and cited DOIs are case-normalised. Network/parse failures
|
|
degrade to an empty list rather than aborting the ingest.
|
|
"""
|
|
s2_id = _s2_id_for(paper.id)
|
|
if s2_id is None:
|
|
return []
|
|
try:
|
|
raw = semanticscholar.fetch_references(s2_id)
|
|
except Exception:
|
|
logger.warning("S2 reference supplement failed for %s", paper.id, exc_info=True)
|
|
return []
|
|
return [
|
|
Citation(citing_id=paper.id, cited_id=_norm_cited_id(c.cited_id), context=c.context)
|
|
for c in raw
|
|
if c.cited_id
|
|
]
|
|
|
|
|
|
def ingest_paper(
|
|
paper_id: str,
|
|
source_path: str | None = None,
|
|
rich: bool = False,
|
|
) -> IngestResult:
|
|
"""Idempotent ingest of one paper.
|
|
|
|
Parameters
|
|
----------
|
|
paper_id:
|
|
An arXiv ID (``"2301.07041"``), DOI (``"10.1145/…"``), or
|
|
OpenAlex W-ID (``"W2741809807"``).
|
|
source_path:
|
|
Optional path to a local ``.tex`` or ``.pdf`` file.
|
|
If given, the file is parsed into text chunks and stored.
|
|
Supports:
|
|
- ``.tex`` → :func:`codex.parsing.tex.latex_to_text` + chunk
|
|
- ``.pdf`` → :func:`codex.parsing.nougat.pdf_to_markdown` + chunk
|
|
+ :func:`codex.parsing.grobid.extract_references` for refs
|
|
rich:
|
|
When True and *source_path* is a PDF, also extract formulas via
|
|
:func:`codex.parsing.mathpix.extract_formulas` and figures via
|
|
:func:`codex.parsing.figures.extract_figures` (F-09).
|
|
|
|
Returns
|
|
-------
|
|
IngestResult
|
|
Counts of upserted chunks, citations, formulas, and figures.
|
|
"""
|
|
# ---------------------------------------------------------------
|
|
# 1. Fetch metadata (OpenAlex primary, SemanticScholar fallback)
|
|
# ---------------------------------------------------------------
|
|
paper: Paper | None = openalex.fetch_paper(paper_id)
|
|
|
|
if paper is None:
|
|
# Detect arXiv IDs: numeric pattern like "2301.07041" or "arxiv:..." prefix
|
|
pid_lower = paper_id.lower()
|
|
looks_like_arxiv = pid_lower.startswith("arxiv:") or (
|
|
len(paper_id) > 4 and paper_id[4:5] == "." and paper_id[:4].isdigit()
|
|
)
|
|
if looks_like_arxiv:
|
|
# OpenAlex 404 on an arXiv id → recover authoritative metadata from
|
|
# the arXiv API (DQ-2); fall back to a minimal stub if arXiv also
|
|
# has nothing. Citation recovery from S2 happens in the citation
|
|
# block below (DQ-1 supplement).
|
|
paper = arxiv.fetch_metadata(paper_id) or Paper(id=paper_id, title="")
|
|
else:
|
|
raise ValueError(f"Paper not found: {paper_id}")
|
|
|
|
if paper is None:
|
|
raise ValueError(f"Paper not found: {paper_id}")
|
|
|
|
# DQ-2: OpenAlex often has the paper but no abstract (it cannot redistribute
|
|
# some publishers' abstracts). Supplement from S2, then Crossref, so the
|
|
# paper gets a real (non-zero) abstract embedding instead of a zero vector.
|
|
if not (paper.abstract or "").strip():
|
|
recovered = _recover_abstract(paper)
|
|
if recovered:
|
|
paper.abstract = recovered
|
|
|
|
# Auto-generate bibkey when source has none (D-04).
|
|
if paper.bibkey is None:
|
|
paper.bibkey = _make_bibkey(paper)
|
|
|
|
# ---------------------------------------------------------------
|
|
# 2. Embed abstract (dense only — schema has one vector column)
|
|
# ---------------------------------------------------------------
|
|
embedder = get_embedder()
|
|
abstract_text = paper.abstract or ""
|
|
if abstract_text:
|
|
abstract_emb_arr: np.ndarray = embedder.encode_dense([abstract_text])
|
|
abstract_emb: list[float] = abstract_emb_arr[0].tolist()
|
|
else:
|
|
dim = embedder.dim
|
|
abstract_emb = [0.0] * dim
|
|
|
|
# ---------------------------------------------------------------
|
|
# 3. Upsert paper (INSERT … ON CONFLICT (id) DO UPDATE SET …)
|
|
# ---------------------------------------------------------------
|
|
with get_conn() as conn:
|
|
conn.execute(
|
|
"""
|
|
INSERT INTO papers (id, openalex_id, bibkey, title, authors, year, abstract,
|
|
source_path, abstract_emb)
|
|
VALUES (%(id)s, %(openalex_id)s, %(bibkey)s, %(title)s, %(authors)s, %(year)s,
|
|
%(abstract)s, %(source_path)s, %(abstract_emb)s)
|
|
ON CONFLICT (id) DO UPDATE SET
|
|
openalex_id = EXCLUDED.openalex_id,
|
|
bibkey = COALESCE(papers.bibkey, EXCLUDED.bibkey),
|
|
title = EXCLUDED.title,
|
|
authors = EXCLUDED.authors,
|
|
year = EXCLUDED.year,
|
|
abstract = EXCLUDED.abstract,
|
|
source_path = COALESCE(EXCLUDED.source_path, papers.source_path),
|
|
abstract_emb = EXCLUDED.abstract_emb
|
|
""",
|
|
{
|
|
"id": paper.id,
|
|
"openalex_id": paper.openalex_id,
|
|
"bibkey": paper.bibkey,
|
|
"title": paper.title,
|
|
"authors": paper.authors,
|
|
"year": paper.year,
|
|
"abstract": paper.abstract,
|
|
"source_path": source_path,
|
|
"abstract_emb": abstract_emb,
|
|
},
|
|
)
|
|
|
|
# Register openalex_id in paper_identifiers (alias table).
|
|
# Every ingest records the primary openalex_id so the graph JOIN
|
|
# resolves it even if papers.openalex_id is later updated.
|
|
if paper.openalex_id:
|
|
conn.execute(
|
|
"""
|
|
INSERT INTO paper_identifiers (paper_id, openalex_id)
|
|
VALUES (%s, %s)
|
|
ON CONFLICT DO NOTHING
|
|
""",
|
|
(paper.id, paper.openalex_id),
|
|
)
|
|
|
|
# ---------------------------------------------------------------
|
|
# 4. Parse + chunk + embed source file (if provided)
|
|
# DELETE existing chunks for this paper first, then bulk INSERT
|
|
# ---------------------------------------------------------------
|
|
chunks_upserted = 0
|
|
pdf_citations: list[Citation] = []
|
|
|
|
if source_path is not None:
|
|
from codex.parsing.grobid import extract_references
|
|
from codex.parsing.nougat import pdf_to_markdown
|
|
from codex.parsing.tex import chunk_text, latex_to_text
|
|
|
|
suffix = Path(source_path).suffix.lower()
|
|
text = ""
|
|
|
|
if suffix == ".tex":
|
|
text = latex_to_text(Path(source_path).read_text())
|
|
elif suffix == ".txt":
|
|
text = Path(source_path).read_text(encoding="utf-8", errors="replace")
|
|
elif suffix == ".pdf":
|
|
text = pdf_to_markdown(source_path)
|
|
# Also extract GROBID refs
|
|
grobid_refs = extract_references(source_path)
|
|
for ref in grobid_refs:
|
|
cited_id = ref.get("doi") or ref.get("arxiv_id")
|
|
if cited_id:
|
|
pdf_citations.append(Citation(citing_id=paper.id, cited_id=cited_id))
|
|
else:
|
|
logger.warning("Unbekannter Dateityp: %s — kein Text-Parsing", source_path)
|
|
|
|
raw_chunks = chunk_text(text) if text else []
|
|
chunks_text = filter_chunks(raw_chunks, settings=get_settings())
|
|
|
|
# Delete existing chunks for this paper
|
|
conn.execute("DELETE FROM chunks WHERE paper_id = %s", (paper.id,))
|
|
|
|
if chunks_text:
|
|
# Embed all chunks in one batch
|
|
chunk_embeddings = embedder.encode_dense(chunks_text)
|
|
chunk_rows = [
|
|
(
|
|
paper.id,
|
|
ord_idx,
|
|
content,
|
|
chunk_embeddings[ord_idx].tolist(),
|
|
classify_section(content),
|
|
)
|
|
for ord_idx, content in enumerate(chunks_text)
|
|
]
|
|
with conn.cursor() as cur:
|
|
cur.executemany(
|
|
"INSERT INTO chunks (paper_id, ord, content, embedding, section)"
|
|
" VALUES (%s, %s, %s, %s, %s)",
|
|
chunk_rows,
|
|
)
|
|
chunks_upserted = len(chunk_rows)
|
|
|
|
# ---------------------------------------------------------------
|
|
# 5. Fetch citations (OpenAlex if openalex_id, else S2)
|
|
# Insert with ON CONFLICT (citing_id, cited_id) DO NOTHING
|
|
# ---------------------------------------------------------------
|
|
api_citations: list[Citation]
|
|
if paper.openalex_id:
|
|
raw = openalex.fetch_citations(paper.openalex_id)
|
|
# OpenAlex returns citing_id=openalex_id, but papers.id uses the DOI.
|
|
# Rewrite citing_id to paper.id so the FK constraint is satisfied.
|
|
api_citations = [
|
|
Citation(citing_id=paper.id, cited_id=c.cited_id, context=c.context) for c in raw
|
|
]
|
|
# DQ-1: OpenAlex indexes many arXiv preprints / theses WITHOUT a
|
|
# parsed reference list (referenced_works == []). Fall back to the
|
|
# Semantic Scholar references, which often has them.
|
|
if not api_citations:
|
|
api_citations = _s2_reference_supplement(paper)
|
|
else:
|
|
api_citations = _s2_reference_supplement(paper)
|
|
|
|
# Merge API citations and GROBID PDF citations; dedup via set
|
|
all_citations_set: set[tuple[str, str]] = set()
|
|
merged_citations: list[Citation] = []
|
|
for cit in api_citations + pdf_citations:
|
|
key = (cit.citing_id, cit.cited_id)
|
|
if key not in all_citations_set:
|
|
all_citations_set.add(key)
|
|
merged_citations.append(cit)
|
|
|
|
if merged_citations:
|
|
with conn.cursor() as cur:
|
|
cur.executemany(
|
|
"""
|
|
INSERT INTO citations (citing_id, cited_id, context)
|
|
VALUES (%s, %s, %s)
|
|
ON CONFLICT (citing_id, cited_id) DO NOTHING
|
|
""",
|
|
[(c.citing_id, c.cited_id, c.context) for c in merged_citations],
|
|
)
|
|
|
|
citations_upserted = len(merged_citations)
|
|
conn.commit()
|
|
|
|
# ---------------------------------------------------------------
|
|
# 6. F-09 Rich Parsing: formulas + figures (PDF only)
|
|
# ---------------------------------------------------------------
|
|
formulas_upserted = 0
|
|
figures_upserted = 0
|
|
|
|
if rich and source_path is not None and Path(source_path).suffix.lower() == ".pdf":
|
|
from codex.parsing.figures import extract_figures
|
|
from codex.parsing.mathpix import extract_formulas
|
|
|
|
settings = get_settings()
|
|
|
|
formulas = extract_formulas(source_path)
|
|
figures = extract_figures(source_path, output_dir=settings.figures_dir)
|
|
|
|
if formulas or figures:
|
|
with get_conn() as conn2:
|
|
with conn2.cursor() as cur:
|
|
# Delete-before-insert keeps re-ingest idempotent (BIGSERIAL has no
|
|
# natural UNIQUE key, so ON CONFLICT DO NOTHING never fires).
|
|
cur.execute("DELETE FROM formulas WHERE paper_id = %s", (paper.id,))
|
|
cur.execute("DELETE FROM figures WHERE paper_id = %s", (paper.id,))
|
|
|
|
if formulas:
|
|
with conn2.cursor() as cur:
|
|
cur.executemany(
|
|
"""
|
|
INSERT INTO formulas (paper_id, page, raw_latex, context, eq_label)
|
|
VALUES (%s, %s, %s, %s, %s)
|
|
""",
|
|
[
|
|
(f.paper_id, f.page, f.raw_latex, f.context, f.eq_label)
|
|
for f in formulas
|
|
],
|
|
)
|
|
formulas_upserted = len(formulas)
|
|
|
|
if figures:
|
|
with conn2.cursor() as cur:
|
|
cur.executemany(
|
|
"""
|
|
INSERT INTO figures (paper_id, page, caption, image_path)
|
|
VALUES (%s, %s, %s, %s)
|
|
""",
|
|
[
|
|
(fig.paper_id, fig.page, fig.caption, fig.image_path)
|
|
for fig in figures
|
|
],
|
|
)
|
|
figures_upserted = len(figures)
|
|
|
|
conn2.commit()
|
|
|
|
return IngestResult(
|
|
paper_id=paper.id,
|
|
chunks_upserted=chunks_upserted,
|
|
citations_upserted=citations_upserted,
|
|
formulas_upserted=formulas_upserted,
|
|
figures_upserted=figures_upserted,
|
|
)
|