Unify source key, fix ADS auth, add parse() bool return convention

- Rename db_ref/source_item → source in all connector JSON configs
- Registry now gates on source key instead of properties
- Fix SourceItem.extract() to pass Request object directly (preserving headers like Authorization)
- Change parse() signature to bool: True=return self, False=return None from extract()
- Add tracing log in extract() for non-200 responses
- Add extract() docstring describing None/empty-patch/non-empty-patch convention
- Add ADS unit tests covering real lookups and not-found cases
- Add CLAUDE.md noting unittest framework

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: ghuron

CLAUDE.md

@@ -0,0 +1,9 @@
+# Project Notes
+
+## Testing
+
+Unit tests use Python's built-in `unittest` framework. Run with:
+
+```
+python -m unittest discover tests/
+```

src/wdpy/connectors/__init__.py

@@ -1,3 +1,4 @@
 from .astro import *
 from .exoplanet import *
-from .simbad import *
+from .simbad import *
+from .ads import *

src/wdpy/connectors/ads.json

@@ -1,5 +1,5 @@
 {
-  "db_ref": "Q752099",
+  "source": "Q752099",
   "properties": {
     "P819": "https://api.adsabs.harvard.edu/v1/search/query?q=bibcode:\"{}\"",
     "P356": "https://api.adsabs.harvard.edu/v1/search/query?q=doi:\"{}\"",

src/wdpy/connectors/ads.py

@@ -15,13 +15,14 @@ def make_request(cls, ident: Statement) -> Optional[Request]:
                 headers={'Authorization': 'Bearer ogOoi0uDxIebyeseB3tAbf5mBTJxXQQWQqE5TW40'}
             )
 
-    def parse(self, text: str) -> None:
+    def parse(self, text: str) -> bool:
         docs = json.loads(text).get('response', {}).get('docs', [])
         if len(docs) != 1:
             if self.patch and self.patch[0].mainsnak.property == 'P819':
                 logging.warning('Got %d results for %s', len(docs), self.patch[0].mainsnak.value)
-            self.patch = []
-            return
+                self.patch = []
+                return True
+            return False
         d = docs[0]
         if 'page' in d:
             p = d['page'][0] if isinstance(d['page'], list) else d['page']
@@ -41,6 +42,7 @@ def parse(self, text: str) -> None:
             elif ident.startswith('10.') and 'ARXIV' not in ident.upper():
                 self.add_claim('P356', ident)
         self.add_claim('P31', 'Q13442814')
+        return True
 
 
 def _get_orcid(d: dict, idx: int) -> Optional[str]:

src/wdpy/connectors/arxiv.json

@@ -1,5 +1,5 @@
 {
-    "db_ref": "Q118398",
+    "source": "Q118398",
     "properties": {
         "P818": "https://export.arxiv.org/api/query?id_list={}",
         "P356": "https://export.arxiv.org/api/query?search_query={}"

src/wdpy/connectors/arxiv.py

@@ -2,11 +2,11 @@
 from wdpy import SourceItem
 
 class ArxivItem(SourceItem):
-    def parse(self, text: str) -> None:
+    def parse(self, text: str) -> bool:
         ns = self._config['namespaces']
         if (entry := ElementTree.fromstring(text).find('w3:entry', ns)) is None:
             self.patch = []
-            return
+            return True
         id_text = getattr(entry.find('w3:id', ns), 'text', '') or ''
         arxiv_id = id_text.split('/')[-1].split('v')[0]
         self.add_claim('P31', 'Q13442814')
@@ -20,6 +20,7 @@ def parse(self, text: str) -> None:
                 self.add_claim('P2093', name).set_qualifier('P1545', str(i := i + 1))
         if len(doi := entry.findall('arxiv:doi', ns)) == 1 and doi[0].text:
             self.add_claim('P356', doi[0].text.upper())
+        return True
 
 # r = ArxivItem.extract(Statement(Snak('P818', ('1309.0951',))))
 # r = ArxivItem.extract(Statement(Snak('P356', ('10.4171/161',))))

src/wdpy/connectors/astro.py

@@ -1,6 +1,5 @@
 from typing import Any, Dict, List, Optional, Union
-from wdpy.core import build_request, parse_csv
-from wdpy.source_item import SourceItem
+from wdpy import build_request, parse_csv, SourceItem, Statement
 
 class AstroItem(SourceItem):
     @classmethod
@@ -36,4 +35,8 @@ def get_canonical_name(name: str) -> Optional[str]:
         )
         if (rows := SimbadItem.exec(query)) and (main_id := list(rows.keys())[0]):
             return main_id
-        return None
+        return None
+
+    @classmethod
+    def extract(cls, ident: Statement) -> Optional[SourceItem]:
+        return None # Temporary until ADQL extractor is implemented

src/wdpy/connectors/crossref.json

@@ -1,5 +1,5 @@
 {
-  "db_ref": "Q5188229",
+  "source": "Q5188229",
   "extract": [404],
   "properties": {
     "P356": "https://api.crossref.org/works/{}"

src/wdpy/connectors/crossref.py

@@ -3,5 +3,6 @@
 
 
 class Crossref(SourceItem):
-    def parse(self, text: str) -> None:
+    def parse(self, text: str) -> bool:
         self.obtain(json.loads(text), self._config.get('fields', {}), self._config.get('translate', {}))
+        return True

src/wdpy/connectors/europepmc.json

@@ -1,5 +1,5 @@
 {
-  "db_ref": "Q5412157",
+  "source": "Q5412157",
   "properties": {
     "P698": "https://www.ebi.ac.uk/europepmc/webservices/rest/search?resulttype=core&format=json&query=EXT_ID:{}",
     "P356": "https://www.ebi.ac.uk/europepmc/webservices/rest/search?resulttype=core&format=json&query=DOI:{}"