Workaround for older model

Author: pamelafox

src/fastapi_app/query_rewriter.py

@@ -18,7 +18,7 @@ def build_search_function() -> list[ChatCompletionToolParam]:
                     "properties": {
                         "search_query": {
                             "type": "string",
-                            "description": "Query string to use for full text search, e.g. 'red shoes'",
+                            "description": "Query string to use for full text search, e.g. 'red shoes'. This required parameter must always be specified.",
                         },
                         "price_filter": {
                             "type": "object",
@@ -56,7 +56,7 @@ def build_search_function() -> list[ChatCompletionToolParam]:
     ]
 
 
-def extract_search_arguments(chat_completion: ChatCompletion):
+def extract_search_arguments(original_user_query: str, chat_completion: ChatCompletion):
     response_message = chat_completion.choices[0].message
     search_query = None
     filters = []
@@ -67,7 +67,8 @@ def extract_search_arguments(chat_completion: ChatCompletion):
             function = tool.function
             if function.name == "search_database":
                 arg = json.loads(function.arguments)
-                search_query = arg.get("search_query")
+                # Even though its required, search_query is not always specified
+                search_query = arg.get("search_query", original_user_query)
                 if "price_filter" in arg and arg["price_filter"]:
                     price_filter = arg["price_filter"]
                     filters.append(

src/fastapi_app/rag_advanced.py

@@ -65,7 +65,7 @@ async def run(
             tool_choice="auto",
         )
 
-        query_text, filters = extract_search_arguments(chat_completion)
+        query_text, filters = extract_search_arguments(original_user_query, chat_completion)
 
         # Retrieve relevant items from the database with the GPT optimized query
         results = await self.searcher.search_and_embed(