Change function call type to object (#815)

# Description

**Issue:**
The function_call can be of type 'str' or 'dict'. However, when
function_call is set to 'str' type in the char API, it becomes unsafe to
pass the parameter. Exceptions could occur when trying to convert a
'dict' input to a 'str' input in the process of passing this parameter.

**Solution:**
The type of function_call was changed from 'str' to 'object' to enhance
safety. The 'object' type is more versatile and can handle a wider range
of data types, thus reducing the risk of exceptions.

**Tests:** 
Additional unit and end-to-end tests were added to ensure the
functionality and safety of the changes. All tests were successfully
passed, indicating that the solution is effective and doesn't introduce
new issues.

# All Promptflow Contribution checklist:
- [X] **The pull request does not introduce [breaking changes].**
- [ ] **CHANGELOG is updated for new features, bug fixes or other
significant changes.**
- [X] **I have read the [contribution guidelines](../CONTRIBUTING.md).**
- [ ] **Create an issue and link to the pull request to get dedicated
review from promptflow team. Learn more: [suggested
workflow](../CONTRIBUTING.md#suggested-workflow).**

## General Guidelines and Best Practices
- [X] Title of the pull request is clear and informative.
- [X] There are a small number of commits, each of which have an
informative message. This means that previously merged commits do not
appear in the history of the PR. For more information on cleaning up the
commits in your PR, [see this
page](https://github.com/Azure/azure-powershell/blob/master/documentation/development-docs/cleaning-up-commits.md).

### Testing Guidelines
- [x] Pull request includes test coverage for the included changes.

---------

Co-authored-by: cs_lucky <[email protected]>

src/promptflow-tools/promptflow/tools/aoai.py

@@ -112,7 +112,8 @@ def chat(
         frequency_penalty: float = 0,
         logit_bias: dict = {},
         user: str = "",
-        function_call: str = None,
+        # function_call can be of type str or dict.
+        function_call: object = None,
         functions: list = None,
         **kwargs,
     ) -> [str, dict]:
@@ -218,7 +219,7 @@ def chat(
     frequency_penalty: float = 0,
     logit_bias: dict = {},
     user: str = "",
-    function_call: str = None,
+    function_call: object = None,
     functions: list = None,
     **kwargs,
 ) -> str:

src/promptflow-tools/promptflow/tools/common.py

@@ -218,24 +218,15 @@ def process_function_call(function_call):
         common_tsg = f"Here is a valid example: {function_call_example}. See the guide at " \
                      "https://platform.openai.com/docs/api-reference/chat/create#chat/create-function_call " \
                      "or view sample 'How to call functions with chat models' in our gallery."
-        try:
-            param = json.loads(function_call)
-        except json.JSONDecodeError:
-            raise ChatAPIInvalidFunctions(
-                message=f"function_call parameter '{function_call}' is an invalid json. {common_tsg}")
-        except TypeError:
-            raise ChatAPIInvalidFunctions(
-                message=f"function_call parameter '{function_call}' must be str, bytes or bytearray"
-                        f", but not {type(function_call)}. {common_tsg}"
-                )
+        param = function_call
         if not isinstance(param, dict):
             raise ChatAPIInvalidFunctions(
-                message=f"function_call parameter '{function_call}' must be a dict, but not {type(param)}. {common_tsg}"
+                message=f"function_call parameter '{param}' must be a dict, but not {type(function_call)}. {common_tsg}"
             )
         else:
-            if "name" not in param:
+            if "name" not in function_call:
                 raise ChatAPIInvalidFunctions(
-                    message=f'function_call parameter {function_call} must contain "name" field. {common_tsg}'
+                    message=f'function_call parameter {json.dumps(param)} must contain "name" field. {common_tsg}'
                 )
     return param
 

src/promptflow-tools/promptflow/tools/openai.py

@@ -107,7 +107,8 @@ def chat(
             frequency_penalty: float = 0,
             logit_bias: dict = {},
             user: str = "",
-            function_call: str = None,
+            # function_call can be of type str or dict.
+            function_call: object = None,
             functions: list = None,
             **kwargs,
     ) -> [str, dict]:
@@ -211,7 +212,7 @@ def chat(
         frequency_penalty: float = 0,
         logit_bias: dict = {},
         user: str = "",
-        function_call: str = None,
+        function_call: object = None,
         functions: list = None,
         **kwargs,
 ) -> [str, dict]:

src/promptflow-tools/tests/test_aoai.py

@@ -48,8 +48,15 @@ def test_aoai_chat_api(self, azure_open_ai_connection, example_prompt_template,
         )
         assert "Product X".lower() in result.lower()
 
+    @pytest.mark.parametrize(
+        "function_call",
+        [
+            "auto",
+            {"name": "get_current_weather"},
+        ],
+    )
     def test_aoai_chat_with_function(
-            self, azure_open_ai_connection, example_prompt_template, chat_history, functions):
+            self, azure_open_ai_connection, example_prompt_template, chat_history, functions, function_call):
         result = chat(
             connection=azure_open_ai_connection,
             prompt=example_prompt_template,
@@ -59,7 +66,7 @@ def test_aoai_chat_with_function(
             user_input="What is the weather in Boston?",
             chat_history=chat_history,
             functions=functions,
-            function_call="auto"
+            function_call=function_call
         )
         assert "function_call" in result
         assert result["function_call"]["name"] == "get_current_weather"

src/promptflow-tools/tests/test_common.py

@@ -30,12 +30,10 @@ def test_chat_api_invalid_functions(self, functions, error_message):
     @pytest.mark.parametrize(
         "function_call, error_message",
         [
-            ({"name": "get_current_weather"}, "must be str, bytes or bytearray"),
-            ("{'name': 'get_current_weather'}", "is an invalid json"),
-            ("get_current_weather", "is an invalid json"),
             ("123", "function_call parameter '123' must be a dict"),
-            ('{"name1": "get_current_weather"}', 'function_call parameter {"name1": "get_current_weather"} must '
-                                                 'contain "name" field'),
+            ({"name1": "get_current_weather"},
+             'function_call parameter {"name1": "get_current_weather"} must '
+             'contain "name" field'),
         ],
     )
     def test_chat_api_invalid_function_call(self, function_call, error_message):

src/promptflow/tests/executor/e2etests/test_executor_happypath.py

@@ -105,6 +105,7 @@ def test_executor_storage(self, dev_connections):
             "prompt_tools",
             "script_with___file__",
             "connection_as_input",
+            "sample_flow_with_functions"
         ],
     )
     def test_executor_exec_bulk(self, flow_folder, dev_connections):

src/promptflow/tests/test_configs/flows/sample_flow_with_functions/flow.dag.yaml

@@ -0,0 +1,104 @@
+id: use_functions_with_chat_models
+name: Use Functions with Chat Models
+inputs:
+  chat_history:
+    type: list
+    default:
+    - inputs:
+        question: What is the weather like in Boston?
+      outputs:
+        answer: '{"forecast":["sunny","windy"],"location":"Boston","temperature":"72","unit":"fahrenheit"}'
+        llm_output:
+          content: null
+          function_call:
+            name: get_current_weather
+            arguments: |-
+              {
+                "location": "Boston"
+              }
+          role: assistant
+    is_chat_input: false
+  question:
+    type: string
+    default: How about London next week?
+    is_chat_input: true
+outputs:
+  answer:
+    type: string
+    reference: ${run_function.output}
+    is_chat_output: true
+  llm_output:
+    type: object
+    reference: ${use_functions_with_chat_models.output}
+nodes:
+- name: run_function
+  type: python
+  source:
+    type: code
+    path: run_function.py
+  inputs:
+    response_message: ${use_functions_with_chat_models.output}
+  use_variants: false
+- name: use_functions_with_chat_models
+  type: llm
+  source:
+    type: code
+    path: use_functions_with_chat_models.jinja2
+  inputs:
+    deployment_name: gpt-35-turbo
+    temperature: 0.7
+    top_p: 1
+    stop: ""
+    max_tokens: 256
+    presence_penalty: 0
+    frequency_penalty: 0
+    logit_bias: ""
+    functions:
+    - name: get_current_weather
+      description: Get the current weather in a given location
+      parameters:
+        type: object
+        properties:
+          location:
+            type: string
+            description: The city and state, e.g. San Francisco, CA
+          unit:
+            type: string
+            enum:
+            - celsius
+            - fahrenheit
+        required:
+        - location
+    - name: get_n_day_weather_forecast
+      description: Get an N-day weather forecast
+      parameters:
+        type: object
+        properties:
+          location:
+            type: string
+            description: The city and state, e.g. San Francisco, CA
+          format:
+            type: string
+            enum:
+            - celsius
+            - fahrenheit
+            description: The temperature unit to use. Infer this from the users location.
+          num_days:
+            type: integer
+            description: The number of days to forecast
+        required:
+        - location
+        - format
+        - num_days
+    function_call:
+      name: get_current_weather
+    chat_history: ${inputs.chat_history}
+    question: ${inputs.question}
+  provider: AzureOpenAI
+  connection: azure_open_ai_connection
+  api: chat
+  module: promptflow.tools.aoai
+  use_variants: false
+node_variants: {}
+environment:
+  python_requirements_txt: requirements.txt

src/promptflow/tests/test_configs/flows/sample_flow_with_functions/run_function.py

@@ -0,0 +1,41 @@
+from promptflow import tool
+import json
+
+
+def get_current_weather(location, unit="fahrenheit"):
+    """Get the current weather in a given location"""
+    weather_info = {
+        "location": location,
+        "temperature": "72",
+        "unit": unit,
+        "forecast": ["sunny", "windy"],
+    }
+    return weather_info
+
+
+def get_n_day_weather_forecast(location, format, num_days):
+    """Get next num_days weather in a given location"""
+    weather_info = {
+        "location": location,
+        "temperature": "60",
+        "format": format,
+        "forecast": ["rainy"],
+        "num_days": num_days,
+    }
+    return weather_info
+
+
+@tool
+def run_function(response_message: dict) -> str:
+    if "function_call" in response_message:
+        function_name = response_message["function_call"]["name"]
+        function_args = json.loads(response_message["function_call"]["arguments"])
+        print(function_args)
+        result = globals()[function_name](**function_args)
+    else:
+        print("No function call")
+        if isinstance(response_message, dict):
+            result = response_message["content"]
+        else:
+            result = response_message
+    return result

src/promptflow/tests/test_configs/flows/sample_flow_with_functions/use_functions_with_chat_models.jinja2

@@ -0,0 +1,27 @@
+system:
+Don't make assumptions about what values to plug into functions. Ask for clarification if a user request is ambiguous.
+
+{% for item in chat_history %}
+user:
+{{item.inputs.question}}
+
+{% if 'function_call' in item.outputs.llm_output %}
+assistant:
+Function generation requested, function = {{item.outputs.llm_output.function_call.name}}, args = {{item.outputs.llm_output.function_call.arguments}}
+
+function:
+name:
+{{item.outputs.llm_output.function_call.name}}
+content:
+{{item.outputs.answer}}
+
+{% else %}
+assistant:
+{{item.outputs.llm_output}}}}
+
+{% endif %}}
+
+{% endfor %}
+
+user:
+{{question}}