bibcite-cli 0.6.7__tar.gz → 0.6.8__tar.gz

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: bibcite-cli
-Version: 0.6.7
+Version: 0.6.8
 Summary: Resolve papers (arXiv id / DOI / title) to canonical, normalized BibTeX for agents and humans
 Project-URL: Repository, https://github.com/leo1oel/bibcite
 License-Expression: MIT

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "bibcite-cli"
-version = "0.6.7"
+version = "0.6.8"
 description = "Resolve papers (arXiv id / DOI / title) to canonical, normalized BibTeX for agents and humans"
 readme = "Readme.md"
 license = "MIT"

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/src/bibcite/cli.py

@@ -97,6 +97,27 @@ def cmd_get(args) -> int:
     return 0
 
 
+def cmd_normalize(args) -> int:
+    """Canonicalize already retrieved BibTeX without any network lookup."""
+    try:
+        db = bibfile.parse_bib(Path(args.file).read_text())
+        normalized = []
+        for entry in db.entries:
+            raw_venue = entry.get("booktitle", "") or entry.get("journal", "")
+            canonical = canonicalize(raw_venue, entry.get("year"))
+            if canonical:
+                entry.pop("booktitle", None)
+                entry.pop("journal", None)
+                entry["ENTRYTYPE"] = canonical.entry_type
+                entry[canonical.bib_field] = canonical.name
+            normalized.append(bibfile.entry_to_bibtex(entry))
+    except (OSError, ValueError) as error:
+        _log(f"[bibcite] {error}")
+        return EXIT_ISSUES
+    _emit({"bibtex": normalized})
+    return 0
+
+
 # ---------------------------------------------------------------------------
 # add
 # ---------------------------------------------------------------------------
@@ -288,7 +309,9 @@ def cmd_add(args) -> int:
 # upgrade: batch-match arXiv entries in an existing file (bibMatcher, CLI-style)
 # ---------------------------------------------------------------------------
 
-def _upgrade_entries(path: Path, dry_run: bool) -> dict:
+def _upgrade_entries(
+    path: Path, dry_run: bool, include_published_arxiv: bool = False
+) -> dict:
     """Match every preprint entry in ``path`` to its published version and
     rewrite it in place (unless dry_run). Returns the report; does NOT tidy —
     callers decide."""
@@ -301,7 +324,9 @@ def _upgrade_entries(path: Path, dry_run: bool) -> dict:
     changed = 0
     processed = 0
     for entry in db.entries:
-        if not bibfile.is_preprint(entry):
+        if not bibfile.is_preprint(entry) and not (
+            include_published_arxiv and bibfile.entry_arxiv_id(entry)
+        ):
             continue
         if entry.get("pubstate", "").strip("{}") == "preprint":
             # User-confirmed preprint-only (e.g. never-to-be-published arXiv
@@ -386,7 +411,7 @@ def cmd_upgrade(args) -> int:
     if not path.exists():
         _log(f"[bibcite] {path} does not exist")
         return EXIT_ISSUES
-    result = _upgrade_entries(path, args.dry_run)
+    result = _upgrade_entries(path, args.dry_run, args.include_published_arxiv)
     tidied = False
     if result["upgraded"] and not args.no_tidy:
         tidied = bibfile.run_tidy(path)
@@ -550,8 +575,17 @@ def main(argv=None) -> int:
     u.add_argument("--dry-run", action="store_true")
     u.add_argument("--no-tidy", action="store_true")
     u.add_argument("--no-cache", action="store_true", help="bypass the local match cache")
+    u.add_argument(
+        "--include-published-arxiv",
+        action="store_true",
+        help="also audit published-venue entries that retain an arXiv identifier",
+    )
     u.set_defaults(fn=cmd_upgrade)
 
+    n = sub.add_parser("normalize", help="offline canonicalization of a BibTeX file (prints JSON)")
+    n.add_argument("file")
+    n.set_defaults(fn=cmd_normalize)
+
     t = sub.add_parser("tidy", help="run bibtex-tidy with the canonical flags")
     t.add_argument("file")
     t.set_defaults(fn=cmd_tidy)

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/src/bibcite/sources.py

@@ -211,7 +211,7 @@ def _s2_headers() -> dict:
 
 def _s2_batch_status() -> str | None:
     status = os.environ.get("BIBCITE_S2_BATCH_STATUS")
-    return status if status in {"checked", "unavailable"} else None
+    return status if status in {"checked", "unavailable", "disabled"} else None
 
 
 def _mailto() -> str:
@@ -1048,6 +1048,11 @@ def find_published(
     elif batch_status == "unavailable":
         _log("[semanticscholar] batch unavailable")
         incomplete = True
+    elif batch_status == "disabled":
+        # The caller intentionally excluded S2. It is neither an attempted
+        # source nor a failed one, so clean misses from the enabled core
+        # sources remain authoritative.
+        _log("[semanticscholar] disabled by caller")
     # Query every still-viable source concurrently. A preprint with no published
     # version (the common case) misses everywhere, and used to pay the *sum* of
     # each source's latency; now the wall-clock is the slowest single source.

bibcite_cli-0.6.8/tests/test_cli_normalize.py

@@ -0,0 +1,16 @@
+import json
+
+from bibcite.cli import main
+
+
+def test_normalize_preserves_entry_order_offline(tmp_path, capsys):
+    source = tmp_path / "s2.bib"
+    source.write_text(
+        "@inproceedings{a, title={A}, year={2024}, booktitle={CVPR}}\n"
+        "@article{b, title={B}, year={2023}, journal={Nature}}\n"
+    )
+    assert main(["normalize", str(source)]) == 0
+    payload = json.loads(capsys.readouterr().out)
+    assert len(payload["bibtex"]) == 2
+    assert "IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)" in payload["bibtex"][0]
+    assert "journal = {Nature}" in payload["bibtex"][1]

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/tests/test_cli_status.py

@@ -2,6 +2,7 @@ import json
 
 from bibcite import bibfile, cli
 from bibcite.cli import main
+from bibcite.sources import Match
 
 
 def test_check_fails_for_missing_file(tmp_path):
@@ -28,6 +29,62 @@ def test_upgrade_fails_for_missing_file(tmp_path):
     assert main(["upgrade", str(tmp_path / "missing.bib")]) == 1
 
 
+def test_upgrade_opt_in_audits_published_arxiv_entry_through_real_cli(
+    tmp_path, monkeypatch, capsys
+):
+    bib = tmp_path / "refs.bib"
+    original_venue = (
+        "IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"
+    )
+    bib.write_text(
+        "@inproceedings{ranzinger2024amradio,\n"
+        "  author={Mike Ranzinger and Greg Heinrich and Jan Kautz and Pavlo Molchanov},\n"
+        f"  booktitle={{{original_venue}}},\n"
+        "  eprint={2312.06709},\n"
+        "  title={{AM-RADIO:} Agglomerative Vision Foundation Model Reduce All Domains Into One},\n"
+        "  year={2024}\n}\n"
+    )
+    monkeypatch.setattr(bibfile, "run_tidy", lambda path: True)
+    monkeypatch.setattr(
+        cli,
+        "find_published",
+        lambda *args: (
+            Match(
+                venue=original_venue,
+                year="2024",
+                doi="10.1109/CVPR52733.2024.01219",
+                source="dblp",
+            ),
+            "found",
+        ),
+    )
+
+    assert main(["upgrade", str(bib), "--include-published-arxiv"]) == 0
+
+    result = json.loads(capsys.readouterr().out)
+    updated = bibfile.load_bib_file(bib).entries[0]
+    assert result["upgraded"] == 1
+    assert updated["booktitle"] == original_venue
+    assert updated["doi"] == "10.1109/CVPR52733.2024.01219"
+
+
+def test_upgrade_opt_in_preserves_published_venue_on_no_match(
+    tmp_path, monkeypatch, capsys
+):
+    bib = tmp_path / "refs.bib"
+    bib.write_text(
+        "@inproceedings{x, title={T}, booktitle={Existing Venue}, "
+        "eprint={2312.06709}, year={2024}}\n"
+    )
+    before = bib.read_text()
+    monkeypatch.setattr(cli, "find_published", lambda *args: (None, "not_found"))
+
+    assert main(["upgrade", str(bib), "--include-published-arxiv", "--no-tidy"]) == 0
+
+    assert json.loads(capsys.readouterr().out)["upgraded"] == 0
+    assert bib.read_text() == before
+
+
 def test_add_fails_when_automatic_tidy_fails(tmp_path, monkeypatch, capsys):
     bib = tmp_path / "refs.bib"
     monkeypatch.setattr(bibfile, "run_tidy", lambda path: False)
@@ -92,7 +149,11 @@ def test_upgrade_fails_when_automatic_tidy_fails(tmp_path, monkeypatch, capsys):
     monkeypatch.setattr(
         cli,
         "_upgrade_entries",
-        lambda path, dry_run: {"upgraded": 1, "matched": 1, "entries": []},
+        lambda path, dry_run, include_published_arxiv: {
+            "upgraded": 1,
+            "matched": 1,
+            "entries": [],
+        },
     )
     monkeypatch.setattr(bibfile, "run_tidy", lambda path: False)
 

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/tests/test_status_semantics.py

@@ -106,3 +106,28 @@ def test_batch_status_suppresses_s2_metadata_request(monkeypatch):
 
     with pytest.raises(SourceUnavailable):
         sources._s2_get(Client(), "https://api.semanticscholar.org/test", {})
+
+
+def test_disabled_s2_is_never_called_and_does_not_taint_clean_misses(monkeypatch):
+    calls = []
+
+    def source(name):
+        def run(*args):
+            calls.append(name)
+            return None
+
+        return run
+
+    monkeypatch.setenv("BIBCITE_S2_BATCH_STATUS", "disabled")
+    monkeypatch.setenv("S2_API_KEY", "must-not-be-used")
+    monkeypatch.setenv("BIBCITE_PUBLIC_SERVICE_URL", "https://example.test")
+    monkeypatch.setattr(
+        sources,
+        "CASCADE",
+        (("semanticscholar", source("s2")), ("dblp", source("dblp"))),
+    )
+
+    match, status = find_published("Clean miss")
+
+    assert (match, status) == (None, "not_found")
+    assert calls == ["dblp"]

{bibcite_cli-0.6.7 → bibcite_cli-0.6.8}/uv.lock

@@ -18,7 +18,7 @@ wheels = [
 
 [[package]]
 name = "bibcite-cli"
-version = "0.6.7"
+version = "0.6.8"
 source = { editable = "." }
 dependencies = [
     { name = "bibtexparser" },