devdocs/todo.txt

@@ -39,12 +39,11 @@ Current unbounded search for "et*" in huge dataset takes avg 21 seconds to proce
 	- Search and search text indexing		
 	- https://gist.github.com/ruckus/5718112			
 		
-		- Can it be sped up with the huge dataset test
-		(The actual slowness is directly related to namefetch so that's where I am concentrating effort)
-		
-		- One final test *WITH* the index again just to confirm the earlier results, if it's confirmed slower then remove the index 
-		 but keep the code in place in case it becomes faster again with it later
-		- Re-run the integration tests for search performance and set a new benchmark minimum for that shit as it should be way faster now				
+		//new issue came up during testing under heavy load:
+		- 23505: duplicate key value violates unique constraint "asearchdictionary_word_idx"
+			- Between the read of finding matching words and the adding a not found word another run added it independently
+
+
 		- Change return object from a search to include a count of results plus the restricted list of results	
 			- this way the Client can know there are tons more than just weren't shown so they can narrow their terms
 		- see if any other callers to name fetcher are in tight loops and could benefit from using the new Direct version