Merge pull request #89 from TranslatorSRI/try-sort-by-curie-suffix

This PR adds support for sorting results by the CURIE suffix, which requires:
1. Loading the CURIE suffix from the JSON files generated by Babel in data-loading (in PR TranslatorSRI/Babel#179).
2. Indexing it as a long numerical field in data-loading (so we can do a numerical sort instead of an alphabetical sort).
3. Updating our search sort to sort by score first, followed by sorting by the CURIE suffix for entries that have identical scores.

api/server.py

@@ -250,6 +250,7 @@ async def lookup(string: str,
                 "pf": "preferred_name_exactish^20 preferred_name^3 names^2"
             },
         },
+        "sort": "score DESC, curie_suffix ASC",
         "limit": limit,
         "offset": offset,
         "filter": filters

data-loading/setup-and-load-solr.sh

@@ -95,7 +95,15 @@ curl -X POST -H 'Content-type:application/json' --data-binary '{
             "name":"shortest_name_length",
             "type":"pint",
             "stored":true
-        }
+    	},
+	{
+	    "name":"curie_suffix",
+	    "type":"plong",
+	    "docValues":true,
+	    "stored":true,
+	    "required":false,
+	    "sortMissingLast":true
+	}
     ] }' 'http://localhost:8983/solr/name_lookup/schema'
 
 # Add a copy field to copy preferred_name into preferred_name_exactish.