Merge pull request 'fix(ingest): D-04 bibkey heuristic + D-05a/b/c regression tests + __main__' (#9) from fix/d04-d05-ingest-fixes into main

codex/__main__.py

@@ -0,0 +1,3 @@
+from codex.cli import app
+
+app()

codex/ingest.py

@@ -27,6 +27,25 @@ class IngestResult:
     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 ingest_paper(
     paper_id: str,
     source_path: str | None = None,
@@ -61,6 +80,7 @@ def ingest_paper(
     # ---------------------------------------------------------------
     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()
@@ -78,6 +98,10 @@ def ingest_paper(
     if paper is None:
         raise ValueError(f"Paper not found: {paper_id}")
 
+    # 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)
     # ---------------------------------------------------------------
@@ -102,6 +126,7 @@ def ingest_paper(
                     %(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,
@@ -181,7 +206,10 @@ def ingest_paper(
             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]
+            api_citations = [
+                Citation(citing_id=paper.id, cited_id=c.cited_id, context=c.context)
+                for c in raw
+            ]
         else:
             api_citations = semanticscholar.fetch_references(paper_id)
 

tests/cli/test_cli.py

@@ -351,3 +351,27 @@ class TestAsk:
 
         assert result.exit_code == 1
         assert "not yet implemented" in result.stdout or "not yet implemented" in result.stderr
+
+
+# ---------------------------------------------------------------------------
+# D-05a/b — Entry-point smoke tests (python -m codex, no uv run)
+# ---------------------------------------------------------------------------
+
+
+class TestEntryPoint:
+    """Verify that `python -m codex` works without `uv run` (D-05a/b)."""
+
+    def test_python_m_codex_help(self) -> None:
+        """`python -m codex --help` exits 0 and prints usage."""
+        import subprocess
+        import sys
+
+        proc = subprocess.run(
+            [sys.executable, "-m", "codex", "--help"],
+            capture_output=True,
+            text=True,
+        )
+        assert proc.returncode == 0, f"Non-zero exit: {proc.stderr}"
+        assert "Usage" in proc.stdout or "usage" in proc.stdout.lower(), (
+            f"No usage in output: {proc.stdout[:200]}"
+        )

tests/ingest/test_ingest.py

@@ -13,7 +13,7 @@ from unittest.mock import MagicMock, patch
 import numpy as np
 import pytest
 
-from codex.ingest import IngestResult, ingest_paper
+from codex.ingest import IngestResult, _make_bibkey, ingest_paper
 from codex.models import Citation, FigureChunk, FormulaChunk, Paper
 
 # ---------------------------------------------------------------------------
@@ -71,13 +71,14 @@ def test_ingest_paper_basic() -> None:
     mock_conn.executemany = MagicMock()
     mock_conn.commit = MagicMock()
 
-    api_citations = [
+    # OpenAlex returns citing_id=openalex_id; ingest.py must rewrite to paper.id (D-05c).
+    raw_api_citations = [
         Citation(citing_id=paper.openalex_id or "", cited_id="W999"),
     ]
 
     with (
         patch("codex.ingest.openalex.fetch_paper", return_value=paper) as mock_fetch,
-        patch("codex.ingest.openalex.fetch_citations", return_value=api_citations),
+        patch("codex.ingest.openalex.fetch_citations", return_value=raw_api_citations),
         patch("codex.ingest.get_embedder", return_value=_fake_embedder()),
         patch("codex.ingest.get_conn", side_effect=_make_conn_cm(mock_conn)),
     ):
@@ -94,7 +95,15 @@ def test_ingest_paper_basic() -> None:
     assert isinstance(result, IngestResult)
     assert result.paper_id == paper.id
     assert result.chunks_upserted == 0
-    assert result.citations_upserted == len(api_citations)
+    assert result.citations_upserted == len(raw_api_citations)
+
+    # D-05c regression: citing_id must be rewritten to paper.id (DOI), not openalex_id.
+    mock_cursor = mock_conn.cursor.return_value.__enter__.return_value
+    cit_call = mock_cursor.executemany.call_args_list[0]
+    inserted_rows: list[tuple[str, str, Any]] = cit_call[0][1]
+    assert inserted_rows[0][0] == paper.id, (
+        f"citing_id must be paper.id='{paper.id}', not openalex_id='{paper.openalex_id}'"
+    )
 
 
 # ---------------------------------------------------------------------------
@@ -404,6 +413,76 @@ def test_ingest_paper_rich_requires_pdf_source(tmp_path: Any) -> None:
     assert result.figures_upserted == 0
 
 
+# ---------------------------------------------------------------------------
+# D-04 — bibkey auto-generation heuristic
+# ---------------------------------------------------------------------------
+
+
+def test_make_bibkey_single_author() -> None:
+    paper = Paper(id="doi:10.1/x", title="T", authors=["John Smith"], year=2023)
+    assert _make_bibkey(paper) == "Smith2023"
+
+
+def test_make_bibkey_two_authors() -> None:
+    paper = Paper(id="doi:10.1/x", title="T", authors=["Alice Foo", "Bob Bar"], year=2015)
+    assert _make_bibkey(paper) == "FooBar2015"
+
+
+def test_make_bibkey_three_authors() -> None:
+    paper = Paper(
+        id="doi:10.1/x",
+        title="T",
+        authors=["Alexander Bobenko", "Ulrich Pinkall", "Boris Springborn"],
+        year=2015,
+    )
+    assert _make_bibkey(paper) == "BobenkoPinkallSpringborn2015"
+
+
+def test_make_bibkey_many_authors_uses_etal() -> None:
+    paper = Paper(
+        id="doi:10.1/x",
+        title="T",
+        authors=["A One", "B Two", "C Three", "D Four"],
+        year=2020,
+    )
+    assert _make_bibkey(paper) == "OneEtAl2020"
+
+
+def test_make_bibkey_no_authors_returns_none() -> None:
+    paper = Paper(id="doi:10.1/x", title="T", authors=[], year=2023)
+    assert _make_bibkey(paper) is None
+
+
+def test_make_bibkey_no_year_returns_none() -> None:
+    paper = Paper(id="doi:10.1/x", title="T", authors=["John Smith"], year=None)
+    assert _make_bibkey(paper) is None
+
+
+def test_ingest_sets_bibkey_when_none(tmp_path: Any) -> None:
+    """ingest_paper auto-generates bibkey when paper has none (D-04)."""
+    paper = _make_paper()  # bibkey=None (default from _make_paper)
+    mock_conn = MagicMock()
+    mock_conn.execute = MagicMock()
+    mock_conn.executemany = MagicMock()
+    mock_conn.commit = MagicMock()
+
+    with (
+        patch("codex.ingest.openalex.fetch_paper", return_value=paper),
+        patch("codex.ingest.openalex.fetch_citations", return_value=[]),
+        patch("codex.ingest.get_embedder", return_value=_fake_embedder()),
+        patch("codex.ingest.get_conn", side_effect=_make_conn_cm(mock_conn)),
+    ):
+        ingest_paper(paper.id)
+
+    upsert_params: dict[str, Any] = mock_conn.execute.call_args_list[0][0][1]
+    assert upsert_params["bibkey"] is not None, "bibkey must be auto-generated when paper has none"
+    # authors=["Alice", "Bob"], year=2023 → 2 authors ≤ 3 → concat last tokens
+    assert upsert_params["bibkey"] == "AliceBob2023"
+
+
+# ---------------------------------------------------------------------------
+
+
 def test_ingest_result_has_formula_figure_counts() -> None:
     """IngestResult has formulas_upserted and figures_upserted fields with default 0."""
     result = IngestResult(paper_id="test", chunks_upserted=0, citations_upserted=0)