Merge pull request #143 from pagreene/save-the-curations

Convert curation API to return JSON.

indra_db/client/principal/curation.py

@@ -13,8 +13,9 @@
 logger = logging.getLogger(__name__)
 
 
-def submit_curation(hash_val, tag, curator, ip, text=None,
-                    ev_hash=None, source='direct_client', db=None):
+def submit_curation(hash_val, tag, curator, ip, text=None, ev_hash=None,
+                    source='direct_client', stmt_json=None, ev_json=None,
+                    db=None):
     """Submit a curation for a given preassembled or raw extraction.
 
     Parameters
@@ -36,14 +37,19 @@ def submit_curation(hash_val, tag, curator, ip, text=None,
         The name of the access point through which the curation was performed.
         The default is 'direct_client', meaning this function was used
         directly. Any higher-level application should identify itself here.
+    stmt_json : Optional[dict]
+        The JSON of a preassembled or raw statement that was curated.
+    ev_json : Optional[dict]
+        The JSON of the evidence that was curated.
     db : DatabaseManager
         A database manager object used to access the database.
     """
     if db is None:
         db = get_db('primary')
 
     inp = {'tag': tag, 'text': text, 'curator': curator, 'ip': ip,
-           'source': source, 'pa_hash': hash_val, 'source_hash': ev_hash}
+           'source': source, 'pa_hash': hash_val, 'source_hash': ev_hash,
+           'pa_json': stmt_json, 'ev_json': ev_json}
 
     logger.info("Adding curation: %s" % str(inp))
 
@@ -85,7 +91,7 @@ def get_curations(db=None, **params):
         else:
             constraints.append(getattr(cur, key) == val)
 
-    return db.select_all(cur, *constraints)
+    return [c.to_json() for c in db.select_all(cur, *constraints)]
 
 
 def get_grounding_curations(db=None):
@@ -108,15 +114,15 @@ def get_grounding_curations(db=None):
     groundings = {}
     for cur in curs:
         # If there is no curation given, we skip it
-        if not cur.text:
+        if not cur['text']:
             continue
         # We now try to match the standard pattern for grounding curation
-        cur_text = cur.text.strip()
+        cur_text = cur['text'].strip()
         match = re.match('^\[(.*)\] -> ([^ ]+)$', cur_text)
         # We log any instances of curations that don't match the pattern
         if not match:
             logger.info('"%s" by %s does not match the grounding curation '
-                        'pattern.' % (cur_text, cur.curator))
+                        'pattern.' % (cur_text, cur['curator']))
             continue
         txt, dbid_str = match.groups()
         # We now get a dict of curated mappings to return

indra_db/schemas/principal_schema.py

@@ -594,11 +594,14 @@ class Curation(Base, IndraDBTable):
         source = Column(String)
         ip = Column(INET)
         date = Column(DateTime, default=func.now())
+        pa_json = Column(JSONB)
+        ev_json = Column(JSONB)
 
         def to_json(self):
             return {attr: getattr(self, attr)
                     for attr in ['id', 'pa_hash', 'source_hash', 'tag', 'text',
-                                 'date', 'curator', 'source']}
+                                 'date', 'curator', 'source', 'pa_json',
+                                 'ev_json']}
 
     table_dict[Curation.__tablename__] = Curation
 

rest_api/api.py

@@ -297,7 +297,7 @@ def expand_meta_row():
     if w_cur_counts:
         curations = get_curations(pa_hash=set(entry_hash_lookup.keys()))
         for cur in curations:
-            for entry in entry_hash_lookup[cur.pa_hash]:
+            for entry in entry_hash_lookup[cur['pa_hash']]:
                 entry['cur_count'] += 1
 
     res_json = result.json()
@@ -362,8 +362,7 @@ def submit_curation_endpoint(hash_val, **kwargs):
 @app.route('/curation/list/<stmt_hash>/<src_hash>', methods=['GET'])
 def list_curations(stmt_hash, src_hash):
     curations = get_curations(pa_hash=stmt_hash, source_hash=src_hash)
-    curation_json = [cur.to_json() for cur in curations]
-    return jsonify(curation_json)
+    return jsonify(curations)
 
 
 def main():

rest_api/call_handlers.py

@@ -284,13 +284,13 @@ def get_curation_counts(result):
         cur_counts = {}
         for curation in curations:
             # Update the overall counts.
-            if curation.pa_hash not in cur_counts:
-                cur_counts[curation.pa_hash] = 0
-            cur_counts[curation.pa_hash] += 1
+            if curation['pa_hash'] not in cur_counts:
+                cur_counts[curation['pa_hash']] = 0
+            cur_counts[curation['pa_hash']] += 1
 
             # Work these counts into the evidence dict structure.
-            for ev_json in result.results[curation.pa_hash]['evidence']:
-                if str(ev_json['source_hash']) == str(curation.source_hash):
+            for ev_json in result.results[curation['pa_hash']]['evidence']:
+                if str(ev_json['source_hash']) == str(curation['source_hash']):
                     ev_json['num_curations'] = \
                         ev_json.get('num_curations', 0) + 1
                     break
@@ -372,7 +372,7 @@ def produce_response(self, result):
                             rel['hashes'] = None
                 curations = get_curations(pa_hash=set(rel_hash_lookup.keys()))
                 for cur in curations:
-                    for rel in rel_hash_lookup[cur.pa_hash]:
+                    for rel in rel_hash_lookup[cur['pa_hash']]:
                         rel['cur_count'] += 1
 
             logger.info("Returning with %s results after %.2f seconds."