Merge branch 'dev_main' into ms-632

connectors/amazon-bedrock-connector.py

@@ -4,6 +4,7 @@
 import boto3
 from botocore.config import Config
 from moonshot.src.connectors.connector import Connector, perform_retry
+from moonshot.src.connectors.connector_response import ConnectorResponse
 from moonshot.src.connectors_endpoints.connector_endpoint_arguments import (
     ConnectorEndpointArguments,
 )
@@ -117,7 +118,7 @@ def __init__(self, ep_arguments: ConnectorEndpointArguments):
 
     @Connector.rate_limited
     @perform_retry
-    async def get_response(self, prompt: str) -> str:
+    async def get_response(self, prompt: str) -> ConnectorResponse:
         """Asynchronously send a prompt to the Amazon Bedrock API and return the generated response
 
         This method uses the Bedrock Converse API, which provides more cross-model standardization
@@ -139,7 +140,7 @@ async def get_response(self, prompt: str) -> str:
             prompt (str): The input prompt to send to the model.
 
         Returns:
-            str: The text response generated by the selected model.
+            ConnectorResponse: An object containing the text response generated by the selected model.
         """
         connector_prompt = f"{self.pre_prompt}{prompt}{self.post_prompt}"
         req_params = {
@@ -167,6 +168,11 @@ async def get_response(self, prompt: str) -> str:
                 message,
             )
         # Ignore any non-text contents, and join together with '\n\n' if multiple are returned:
-        return "\n\n".join(
-            map(lambda m: m["text"], filter(lambda m: "text" in m, message["content"]))
+        return ConnectorResponse(
+            response="\n\n".join(
+                map(
+                    lambda m: m["text"],
+                    filter(lambda m: "text" in m, message["content"]),
+                )
+            )
         )

connectors/google-gemini-connector.py

@@ -3,6 +3,7 @@
 
 import google.generativeai as genai
 from moonshot.src.connectors.connector import Connector, perform_retry
+from moonshot.src.connectors.connector_response import ConnectorResponse
 from moonshot.src.connectors_endpoints.connector_endpoint_arguments import (
     ConnectorEndpointArguments,
 )
@@ -22,20 +23,20 @@ def __init__(self, ep_arguments: ConnectorEndpointArguments):
 
     @Connector.rate_limited
     @perform_retry
-    async def get_response(self, prompt: str) -> str:
+    async def get_response(self, prompt: str) -> ConnectorResponse:
         """
         Asynchronously sends a prompt to the Google Gemini API and returns the generated response.
 
         This method constructs a request with the given prompt, optionally prepended and appended with
         predefined strings, and sends it to the Google Gemini API. If a system prompt is set, it is included in the
         request. The method then awaits the response from the API, processes it, and returns the resulting message
-        content as a string.
+        content wrapped in a ConnectorResponse object.
 
         Args:
             prompt (str): The input prompt to send to the Google Gemini API.
 
         Returns:
-            str: The text response generated by the Google Gemini model.
+            ConnectorResponse: An object containing the text response generated by the Google Gemini model.
         """
         connector_prompt = f"{self.pre_prompt}{prompt}{self.post_prompt}"
 
@@ -48,7 +49,7 @@ async def get_response(self, prompt: str) -> str:
         response = model.generate_content(
             connector_prompt, generation_config=generation_config
         )
-        return await self._process_response(response)
+        return ConnectorResponse(response=await self._process_response(response))
 
     async def _process_response(self, response: Any) -> str:
         """

datasets/cache.json

@@ -1237,10 +1237,10 @@
     "description": "Contain prompts that test safety in Singapore-context",
     "examples": null,
     "num_of_dataset_prompts": 59,
-    "created_date": "2024-05-27 16:48:35",
+    "created_date": "2024-10-15 21:25:44",
     "reference": "IMDA",
     "license": "Apache-2.0",
-    "hash": "5d61c07c64808b9d"
+    "hash": "984714efbead7cc1"
   },
   "medqa-us": {
     "id": "medqa-us",