bibcite-cli 0.1.0__tar.gz → 0.2.0__tar.gz

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bibcite-cli
-Version: 0.1.0
+Version: 0.2.0
 Summary: Resolve papers (arXiv id / DOI / title) to canonical, normalized BibTeX for agents and humans
 License-Expression: MIT
 License-File: LICENSE
@@ -57,10 +57,13 @@ bibcite add refs.bib 2103.14030 --json
 # Add a raw BibTeX entry you already have (venue still canonicalized, file still tidied)
 bibcite add refs.bib --bibtex "$(pbpaste)"
 
+# One-shot cleanup: upgrade preprints → tidy → lint
+bibcite fix refs.bib
+
 # Upgrade every arXiv entry in a file to its published version (bibMatcher, CLI-style)
 bibcite upgrade refs.bib --dry-run
 
-# Just format, or just lint
+# Just format, or just lint (check is read-only)
 bibcite tidy refs.bib
 bibcite check refs.bib
 ```

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/Readme.md

@@ -45,10 +45,13 @@ bibcite add refs.bib 2103.14030 --json
 # Add a raw BibTeX entry you already have (venue still canonicalized, file still tidied)
 bibcite add refs.bib --bibtex "$(pbpaste)"
 
+# One-shot cleanup: upgrade preprints → tidy → lint
+bibcite fix refs.bib
+
 # Upgrade every arXiv entry in a file to its published version (bibMatcher, CLI-style)
 bibcite upgrade refs.bib --dry-run
 
-# Just format, or just lint
+# Just format, or just lint (check is read-only)
 bibcite tidy refs.bib
 bibcite check refs.bib
 ```

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "bibcite-cli"
-version = "0.1.0"
+version = "0.2.0"
 description = "Resolve papers (arXiv id / DOI / title) to canonical, normalized BibTeX for agents and humans"
 readme = "Readme.md"
 license = "MIT"

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/src/bibcite/__init__.py

@@ -1,3 +1,3 @@
 """bibcite: canonical BibTeX resolution for papers (arXiv id / DOI / title)."""
 
-__version__ = "0.1.0"
+__version__ = "0.2.0"

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/src/bibcite/cli.py

@@ -8,6 +8,7 @@ citation key it prints.
 import argparse
 import json
 import sys
+import time
 from pathlib import Path
 
 from . import bibfile
@@ -123,21 +124,27 @@ def cmd_add(args) -> int:
 # upgrade: batch-match arXiv entries in an existing file (bibMatcher, CLI-style)
 # ---------------------------------------------------------------------------
 
-def cmd_upgrade(args) -> int:
-    path = Path(args.file)
+def _upgrade_entries(path: Path, dry_run: bool) -> 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."""
     db = bibfile.load_bib_file(path)
     if db is None or not db.entries:
         _log(f"[bibcite] nothing to do in {path}")
-        return 0
+        return {"upgraded": 0, "matched": 0, "entries": []}
 
     report = []
     changed = 0
+    processed = 0
     for entry in db.entries:
         if not bibfile.is_preprint(entry):
             continue
         title = entry.get("title", "").replace("{", "").replace("}", "")
         if not title:
             continue
+        if processed:
+            time.sleep(1)  # be polite to the APIs on batch runs
+        processed += 1
         _log(f"[upgrade] matching: {title[:80]}")
         aid = bibfile.entry_arxiv_id(entry)
         hint = (
@@ -149,7 +156,7 @@ def cmd_upgrade(args) -> int:
             continue
         canonical = canonicalize(match.venue, match.year or entry.get("year"))
         venue_name = canonical.name if canonical else match.venue
-        if not args.dry_run:
+        if not dry_run:
             entry.pop("journal", None)
             entry.pop("booktitle", None)
             entry.pop("howpublished", None)
@@ -179,17 +186,23 @@ def cmd_upgrade(args) -> int:
             }
         )
 
-    if changed and not args.dry_run:
+    if changed and not dry_run:
         bibfile._write_db(path, db)
-        if not args.no_tidy:
-            bibfile.run_tidy(path)
 
     matched = sum(1 for r in report if r["matched"])
     for r in report:
         mark = "✓" if r["matched"] else "✗"
         _log(f"{mark} {r['key']}: {r.get('venue', 'no match')}")
-    _log(f"[bibcite] {matched} matched, {changed} upgraded{' (dry-run)' if args.dry_run else ''}")
-    _emit({"upgraded": changed, "matched": matched, "dry_run": args.dry_run, "entries": report})
+    _log(f"[bibcite] {matched} matched, {changed} upgraded{' (dry-run)' if dry_run else ''}")
+    return {"upgraded": changed, "matched": matched, "entries": report}
+
+
+def cmd_upgrade(args) -> int:
+    path = Path(args.file)
+    result = _upgrade_entries(path, args.dry_run)
+    if result["upgraded"] and not args.no_tidy:
+        bibfile.run_tidy(path)
+    _emit({**result, "dry_run": args.dry_run})
     return 0
 
 
@@ -201,12 +214,11 @@ def cmd_tidy(args) -> int:
     return 0 if bibfile.run_tidy(Path(args.file)) else 1
 
 
-def cmd_check(args) -> int:
-    path = Path(args.file)
+def _check_problems(path: Path) -> tuple[int, list] | None:
+    """(entry count, problem list) for a .bib file, or None if unparseable."""
     db = bibfile.load_bib_file(path)
     if db is None:
-        _log(f"[bibcite] {path} could not be parsed")
-        return 1
+        return None
     problems = []
     seen_titles: dict[str, str] = {}
     for entry in db.entries:
@@ -223,7 +235,37 @@ def cmd_check(args) -> int:
     for p in problems:
         _log(f"{p['key']}: {p['issue']}")
     _log(f"[bibcite] {len(db.entries)} entries, {len(problems)} issues")
-    _emit({"entries": len(db.entries), "problems": problems})
+    return len(db.entries), problems
+
+
+def cmd_check(args) -> int:
+    checked = _check_problems(Path(args.file))
+    if checked is None:
+        _log(f"[bibcite] {args.file} could not be parsed")
+        return 1
+    entries, problems = checked
+    _emit({"entries": entries, "problems": problems})
+    return 0
+
+
+def cmd_fix(args) -> int:
+    """One-shot cleanup: upgrade preprints, always tidy, then re-lint."""
+    path = Path(args.file)
+    if not path.exists():
+        _log(f"[bibcite] {path} does not exist")
+        return 1
+    result = _upgrade_entries(path, dry_run=False)
+    tidied = bibfile.run_tidy(path)
+    checked = _check_problems(path)
+    entries, problems = checked if checked else (0, [])
+    _emit(
+        {
+            **result,
+            "tidied": tidied,
+            "entries_total": entries,
+            "remaining_problems": problems,
+        }
+    )
     return 0
 
 
@@ -260,10 +302,17 @@ def main(argv=None) -> int:
     t.add_argument("file")
     t.set_defaults(fn=cmd_tidy)
 
-    c = sub.add_parser("check", help="offline sanity check of a .bib file (prints JSON)")
+    c = sub.add_parser("check", help="offline read-only lint of a .bib file (prints JSON)")
     c.add_argument("file")
     c.set_defaults(fn=cmd_check)
 
+    f = sub.add_parser(
+        "fix",
+        help="one-shot cleanup: upgrade preprints to published versions, tidy, then lint (prints JSON)",
+    )
+    f.add_argument("file")
+    f.set_defaults(fn=cmd_fix)
+
     args = p.parse_args(argv)
     return args.fn(args)
 

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/src/bibcite/sources.py

@@ -7,6 +7,7 @@ preprint venues (arXiv / CoRR / bioRxiv / ...).
 """
 
 import html
+import os
 import re
 import sys
 import time
@@ -44,6 +45,14 @@ def _client(browser: bool = False) -> httpx.Client:
     )
 
 
+def _s2_headers() -> dict:
+    """Semantic Scholar's unauthenticated pool is shared globally and 429s
+    often; a free API key (https://api.semanticscholar.org) gets a private
+    quota. Set S2_API_KEY (or SEMANTIC_SCHOLAR_API_KEY)."""
+    key = os.environ.get("S2_API_KEY") or os.environ.get("SEMANTIC_SCHOLAR_API_KEY")
+    return {"x-api-key": key} if key else {}
+
+
 @dataclass
 class Match:
     source: str
@@ -260,6 +269,7 @@ def s2_arxiv_metadata(arxiv_id: str) -> ArxivMeta | None:
         r = c.get(
             f"https://api.semanticscholar.org/graph/v1/paper/arXiv:{arxiv_id}",
             params={"fields": "title,year,authors"},
+            headers=_s2_headers(),
         )
         if r.status_code != 200:
             return None
@@ -284,6 +294,7 @@ def try_semantic_scholar(
             r = c.get(
                 f"https://api.semanticscholar.org/graph/v1/paper/arXiv:{arxiv_id}",
                 params={"fields": S2_FIELDS},
+                headers=_s2_headers(),
             )
             if r.status_code == 429:
                 raise SourceUnavailable("Semantic Scholar rate-limited (429)")
@@ -294,6 +305,7 @@ def try_semantic_scholar(
         r = c.get(
             "https://api.semanticscholar.org/graph/v1/paper/search",
             params={"query": title, "fields": S2_FIELDS, "limit": 5},
+            headers=_s2_headers(),
         )
         if r.status_code == 429:
             raise SourceUnavailable("Semantic Scholar rate-limited (429)")
@@ -575,19 +587,27 @@ CASCADE = (
     ("openalex", lambda t, y, a, au: try_openalex(t)),
 )
 
+# Sources that rate-limited/blocked us this process: skip them for the rest of
+# the run instead of hammering them once per entry during batch `upgrade`
+# (PaperMemory's DISABLE_MATCH, ported).
+_DISABLED: dict[str, str] = {}
+
 
 def find_published(
     title: str, year: str = "", arxiv_id: str = "", author_hint: str = ""
 ) -> Match | None:
     """Try each source in order; first verified hit wins."""
     for name, fn in CASCADE:
+        if name in _DISABLED:
+            continue
         try:
             m = fn(title, year, arxiv_id, author_hint)
             if m:
                 return m
             _log(f"[{name}] no publication found")
         except SourceUnavailable as e:
-            _log(f"[{name}] skipped: {e}")
+            _DISABLED[name] = str(e)
+            _log(f"[{name}] disabled for the rest of this run: {e}")
         except Exception as e:  # network hiccup on one source must not kill the run
             _log(f"[{name}] error: {type(e).__name__}: {e}")
     return None

{bibcite_cli-0.1.0 → bibcite_cli-0.2.0}/uv.lock

@@ -18,7 +18,7 @@ wheels = [
 
 [[package]]
 name = "bibcite-cli"
-version = "0.1.0"
+version = "0.2.0"
 source = { editable = "." }
 dependencies = [
     { name = "bibtexparser" },