Serve DataSet summaries from VFBquery term info

src/vfb_connect/cached_query.py

@@ -277,6 +277,39 @@ def cached_queries_enabled():
     return os.getenv('VFB_USE_CACHED_QUERIES', 'true').lower() not in ('false', '0', 'no', 'off')
 
 
+def dataset_summary_from_term_info(term_info):
+    """Map a VFBquery ``get_term_info`` payload for a DataSet onto the summary
+    dict that ``QueryWrapper.get_DataSet_TermInfo`` has always returned
+    (label, symbol, id, tags, description, miniref, FlyBase, PMID, DOI).
+
+    Returns ``None`` if the payload is not a DataSet term.
+    """
+    if not isinstance(term_info, dict) or 'DataSet' not in (term_info.get('SuperTypes') or []):
+        return None
+    meta = term_info.get('Meta') or {}
+    description = meta.get('Description') or ''
+    d = {
+        'label': term_info.get('Name', ''),
+        'symbol': '',
+        'id': term_info.get('Id', ''),
+        'tags': list(term_info.get('SuperTypes') or []),
+        'description': [description] if description else [],
+        'miniref': [], 'FlyBase': [], 'PMID': [], 'DOI': [],
+    }
+    for pub in term_info.get('Publications') or []:
+        d['miniref'].append(str(pub.get('title', '')))
+        if str(pub.get('short_form', '')).startswith('FBrf'):
+            d['FlyBase'].append(str(pub['short_form']))
+        for ref in pub.get('refs') or []:
+            m = re.search(r'pubmed/\?term=(\d+)', ref)
+            if m:
+                d['PMID'].append(m.group(1))
+            m = re.search(r'doi\.org/(.+)$', ref)
+            if m:
+                d['DOI'].append(m.group(1))
+    return d
+
+
 class CachedQueryClient:
     """Fetch pre-computed VFBquery results, or ``None`` if they cannot be trusted.
 

src/vfb_connect/neo/query_wrapper.py

@@ -791,7 +791,25 @@ def get_DataSet_TermInfo(self, short_forms, summary=True, return_dataframe=True)
         :param summary: Optional.  Returns summary reports if `True`. Default `True`
         :rtype: list of VFB_json or summary_report_json
         """
-        return self._get_TermInfo(short_forms, typ='Get JSON for DataSet', summary=summary, return_dataframe=return_dataframe)
+        short_forms = list(short_forms)
+        if not summary:
+            return self._get_TermInfo(short_forms, typ='Get JSON for DataSet', summary=False, return_dataframe=return_dataframe)
+        # Summaries come from VFBquery's cached term info: the equivalent
+        # Neo4j query walks has_source|SUBCLASSOF|INSTANCEOF* unbounded and can
+        # run for many minutes on large datasets. Anything VFBquery cannot
+        # serve falls back to Neo4j.
+        from vfb_connect.cached_query import get_default_client, dataset_summary_from_term_info
+        client = get_default_client()
+        found = {}
+        for sf in short_forms:
+            r = dataset_summary_from_term_info(client.get_term_info(sf))
+            if r is not None:
+                found[sf] = r
+        missing = [sf for sf in short_forms if sf not in found]
+        if missing:
+            for r in self._get_TermInfo(missing, typ='Get JSON for DataSet', summary=True, return_dataframe=return_dataframe):
+                found[r['id']] = r
+        return [found[sf] for sf in short_forms if sf in found]
 
     def get_license_TermInfo(self, short_forms, summary=True, return_dataframe=True):
         """

src/vfb_connect/test/cached_query_test.py

@@ -182,3 +182,32 @@ def test_column_contract_and_pass_through(self):
 
 if __name__ == '__main__':
     unittest.main()
+
+
+class DatasetSummaryFromTermInfoTest(unittest.TestCase):
+    """VFBquery DataSet term info maps onto the legacy dataset summary dict."""
+
+    def test_mapping(self):
+        from vfb_connect.cached_query import dataset_summary_from_term_info
+        ti = {
+            'Name': 'Ito lab adult brain lineage clone image set', 'Id': 'Ito2013',
+            'SuperTypes': ['Entity', 'Individual', 'DataSet', 'has_image'],
+            'Meta': {'Description': 'A set of clones.'},
+            'Publications': [{
+                'title': 'Ito et al., 2013, Curr. Biol. 23(8): 644--655', 'short_form': 'FBrf0221438',
+                'refs': ['http://www.ncbi.nlm.nih.gov/pubmed/?term=23541729',
+                         'http://flybase.org/reports/FBrf0221438',
+                         'https://doi.org/10.1016/j.cub.2013.03.015']}],
+        }
+        d = dataset_summary_from_term_info(ti)
+        self.assertEqual(d['id'], 'Ito2013')
+        self.assertEqual(d['description'], ['A set of clones.'])
+        self.assertEqual(d['miniref'], ['Ito et al., 2013, Curr. Biol. 23(8): 644--655'])
+        self.assertEqual(d['FlyBase'], ['FBrf0221438'])
+        self.assertEqual(d['PMID'], ['23541729'])
+        self.assertEqual(d['DOI'], ['10.1016/j.cub.2013.03.015'])
+
+    def test_non_dataset_returns_none(self):
+        from vfb_connect.cached_query import dataset_summary_from_term_info
+        self.assertIsNone(dataset_summary_from_term_info({'Id': 'FBbt_1', 'SuperTypes': ['Class']}))
+        self.assertIsNone(dataset_summary_from_term_info(None))