refact glue and use base config (#424)

* refact glue and use base config

* revert manifest change

agents/ten_packages/extension/deepgram_asr_python/extension.py

@@ -11,26 +11,34 @@
 import asyncio
 
 from deepgram import AsyncListenWebSocketClient, DeepgramClientOptions, LiveTranscriptionEvents, LiveOptions
+from dataclasses import dataclass
 
-from .config import DeepgramConfig
-
-PROPERTY_API_KEY = "api_key"  # Required
-PROPERTY_LANG = "language"  # Optional
-PROPERTY_MODEL = "model"  # Optional
-PROPERTY_SAMPLE_RATE = "sample_rate"  # Optional
+from ten_ai_base import BaseConfig
 
 DATA_OUT_TEXT_DATA_PROPERTY_TEXT = "text"
 DATA_OUT_TEXT_DATA_PROPERTY_IS_FINAL = "is_final"
 DATA_OUT_TEXT_DATA_PROPERTY_STREAM_ID = "stream_id"
 DATA_OUT_TEXT_DATA_PROPERTY_END_OF_SEGMENT = "end_of_segment"
 
+@dataclass
+class DeepgramASRConfig(BaseConfig):
+    api_key: str = ""
+    language: str = "en-US"
+    model: str = "nova-2"
+    sample_rate: int = 16000
+
+    channels: int = 1
+    encoding: str = 'linear16'
+    interim_results: bool = True
+    punctuate: bool = True
+
 class DeepgramASRExtension(AsyncExtension):
     def __init__(self, name: str):
         super().__init__(name)
 
         self.stopped = False
-        self.deepgram_client : AsyncListenWebSocketClient = None
-        self.deepgram_config : DeepgramConfig = None
+        self.client : AsyncListenWebSocketClient = None
+        self.config : DeepgramASRConfig = None
         self.ten_env : AsyncTenEnv = None
 
     async def on_init(self, ten_env: AsyncTenEnv) -> None:
@@ -41,30 +49,15 @@ async def on_start(self, ten_env: AsyncTenEnv) -> None:
         self.loop = asyncio.get_event_loop()
         self.ten_env = ten_env
 
-        self.deepgram_config = DeepgramConfig.default_config()
+        self.config = DeepgramASRConfig.create(ten_env=ten_env)
+        ten_env.log_info(f"config: {self.config}")
 
-        try:
-            self.deepgram_config.api_key = ten_env.get_property_string(PROPERTY_API_KEY).strip()
-        except Exception as e:
-            ten_env.log_error(f"get property {PROPERTY_API_KEY} error: {e}")
+        if not self.config.api_key:
+            ten_env.log_error(f"get property api_key")
             return
 
-        for optional_param in [
-            PROPERTY_LANG,
-            PROPERTY_MODEL,
-            PROPERTY_SAMPLE_RATE,
-        ]:
-            try:
-                value = ten_env.get_property_string(optional_param).strip()
-                if value:
-                    self.deepgram_config.__setattr__(optional_param, value)
-            except Exception as err:
-                ten_env.log_debug(
-                    f"get property optional {optional_param} failed, err: {err}. Using default value: {self.deepgram_config.__getattribute__(optional_param)}"
-                )
-
-        self.deepgram_client = AsyncListenWebSocketClient(config=DeepgramClientOptions(
-            api_key=self.deepgram_config.api_key,
+        self.client = AsyncListenWebSocketClient(config=DeepgramClientOptions(
+            api_key=self.config.api_key,
             options={"keepalive": "true"}
         ))
 
@@ -80,12 +73,12 @@ async def on_audio_frame(self, ten_env: AsyncTenEnv, frame: AudioFrame) -> None:
             return
 
         self.stream_id = frame.get_property_int('stream_id')
-        await self.deepgram_client.send(frame_buf)
+        await self.client.send(frame_buf)
 
     async def on_stop(self, ten_env: AsyncTenEnv) -> None:
         ten_env.log_info("on_stop")
 
-        await self.deepgram_client.finish()
+        await self.client.finish()
 
     async def on_cmd(self, ten_env: AsyncTenEnv, cmd: Cmd) -> None:
         cmd_json = cmd.to_json()
@@ -118,22 +111,22 @@ async def on_message(_, result, **kwargs):
         async def on_error(_, error, **kwargs):
             self.ten_env.log_error(f"deepgram event callback on_error: {error}")
 
-        self.deepgram_client.on(LiveTranscriptionEvents.Open, on_open)
-        self.deepgram_client.on(LiveTranscriptionEvents.Close, on_close)
-        self.deepgram_client.on(LiveTranscriptionEvents.Transcript, on_message)
-        self.deepgram_client.on(LiveTranscriptionEvents.Error, on_error)
-
-        options = LiveOptions(language=self.deepgram_config.language,
-                              model=self.deepgram_config.model,
-                              sample_rate=self.deepgram_config.sample_rate,
-                              channels=self.deepgram_config.channels,
-                              encoding=self.deepgram_config.encoding,
-                              interim_results=self.deepgram_config.interim_results,
-                              punctuate=self.deepgram_config.punctuate)
+        self.client.on(LiveTranscriptionEvents.Open, on_open)
+        self.client.on(LiveTranscriptionEvents.Close, on_close)
+        self.client.on(LiveTranscriptionEvents.Transcript, on_message)
+        self.client.on(LiveTranscriptionEvents.Error, on_error)
+
+        options = LiveOptions(language=self.config.language,
+                              model=self.config.model,
+                              sample_rate=self.config.sample_rate,
+                              channels=self.config.channels,
+                              encoding=self.config.encoding,
+                              interim_results=self.config.interim_results,
+                              punctuate=self.config.punctuate)
         # connect to websocket
-        result = await self.deepgram_client.start(options)
+        result = await self.client.start(options)
         if result is False:
-            if self.deepgram_client.status_code == 402:
+            if self.client.status_code == 402:
                 self.ten_env.log_error("Failed to connect to Deepgram - your account has run out of credits.")
             else:
                 self.ten_env.log_error("Failed to connect to Deepgram")

agents/ten_packages/extension/glue_python_async/__init__.py

@@ -4,6 +4,4 @@
 # See the LICENSE file for more information.
 #
 from . import addon
-from .log import logger
 
-logger.info("glue_python_async extension loaded")

agents/ten_packages/extension/glue_python_async/addon.py

@@ -15,6 +15,5 @@ class AsyncGlueExtensionAddon(Addon):
 
     def on_create_instance(self, ten_env: TenEnv, name: str, context) -> None:
         from .extension import AsyncGlueExtension
-        from .log import logger
-        logger.info("AsyncGlueExtensionAddon on_create_instance")
+        ten_env.log_info("AsyncGlueExtensionAddon on_create_instance")
         ten_env.on_create_instance_done(AsyncGlueExtension(name), context)

agents/ten_packages/extension/glue_python_async/examples/openai_wrapper.py

@@ -1,16 +1,51 @@
 import os
 import openai
+import json
 from openai import AsyncOpenAI
-import traceback  # Add this import
+import traceback
+import logging
+import logging.config
 
-from typing import List, Union
-from pydantic import BaseModel, HttpUrl
+from typing import List, Union, Dict, Optional
+from pydantic import BaseModel, HttpUrl, ValidationError
 
 from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
-from fastapi.responses import StreamingResponse
+from fastapi.responses import StreamingResponse, JSONResponse
 from fastapi import Depends, FastAPI, HTTPException, Request
 import asyncio
 
+# Enable Pydantic debug mode
+from pydantic import BaseConfig
+
+BaseConfig.debug = True
+
+# Set up logging
+logging.config.dictConfig({
+    "version": 1,
+    "disable_existing_loggers": False,
+    "formatters": {
+        "default": {
+            "format": "%(asctime)s - %(name)s - %(levelname)s - %(message)s",
+        },
+    },
+    "handlers": {
+        "file": {
+            "level": "DEBUG",
+            "formatter": "default",
+            "class": "logging.FileHandler",
+            "filename": "example.log",
+        },
+    },
+    "loggers": {
+        "": {
+            "handlers": ["file"],
+            "level": "DEBUG",
+            "propagate": True,
+        },
+    },
+})
+logger = logging.getLogger(__name__)
+
 app = FastAPI(title="Chat Completion API",
               description="API for streaming chat completions with support for text, image, and audio content",
               version="1.0.0")
@@ -27,79 +62,109 @@ class ImageContent(BaseModel):
     image_url: HttpUrl
 
 class AudioContent(BaseModel):
-    type: str = "audio"
-    audio_url: HttpUrl
-
-class Message(BaseModel):
-    role: str
-    content: Union[TextContent, ImageContent, AudioContent, List[Union[TextContent, ImageContent, AudioContent]]]
+    type: str = "input_audio"
+    input_audio: Dict[str, str]
+
+class ToolFunction(BaseModel):
+    name: str
+    description: Optional[str]
+    parameters: Optional[Dict]
+    strict: bool = False
+
+class Tool(BaseModel):
+    type: str = "function"
+    function: ToolFunction
+
+class ToolChoice(BaseModel):
+    type: str = "function"
+    function: Optional[Dict]
+
+class ResponseFormat(BaseModel):
+    type: str = "json_schema"
+    json_schema: Optional[Dict[str, str]]
+
+class SystemMessage(BaseModel):
+    role: str = "system"
+    content: Union[str, List[str]]
+
+class UserMessage(BaseModel):
+    role: str = "user"
+    content: Union[str, List[Union[TextContent, ImageContent, AudioContent]]]
+
+class AssistantMessage(BaseModel):
+    role: str = "assistant"
+    content: Union[str, List[TextContent]] = None
+    audio: Optional[Dict[str, str]] = None
+    tool_calls: Optional[List[Dict]] = None
+
+class ToolMessage(BaseModel):
+    role: str = "tool"
+    content: Union[str, List[str]]
+    tool_call_id: str
 
 class ChatCompletionRequest(BaseModel):
-    messages: List[Message]
-    model: str
-    temperature: float = 1.0
+    context: Optional[Dict] = None
+    model: Optional[str] = None
+    messages: List[Union[SystemMessage, UserMessage, AssistantMessage, ToolMessage]]
+    response_format: Optional[ResponseFormat] = None
+    modalities: List[str] = ["text"]
+    audio: Optional[Dict[str, str]] = None
+    tools: Optional[List[Tool]] = None
+    tool_choice: Optional[Union[str, ToolChoice]] = "auto"
+    parallel_tool_calls: bool = True
     stream: bool = True
+    stream_options: Optional[Dict] = None
 
-def format_openai_messages(messages):
-    formatted_messages = []
-    for msg in messages:
-        if isinstance(msg.content, list):
-            content = []
-            for item in msg.content:
-                if item.type == "text":
-                    content.append({"type": "text", "text": item.text})
-                elif item.type == "image":
-                    content.append({"type": "image_url", "image_url": str(item.image_url)})
-                elif item.type == "audio":
-                    content.append({"type": "audio_url", "audio_url": str(item.audio_url)})
-        else:
-            if msg.content.type == "text":
-                content = [{"type": "text", "text": msg.content.text}]
-            elif msg.content.type == "image":
-                content = [{"type": "image_url", "image_url": {"url": str(msg.content.image_url)}}]
-            elif msg.content.type == "audio":
-                content = [{"type": "audio_url", "audio_url": {"url": str(msg.content.audio_url)}}]
-
-        formatted_messages.append({"role": msg.role, "content": content})
-    return formatted_messages
+'''
+{'messages': [{'role': 'user', 'content': 'Hello. Hello. Hello.'}, {'role': 'user', 'content': 'Unprocessedable.'}], 'tools': [], 'tools_choice': 'none', 'model': 'gpt-3.5-turbo', 'stream': True}
+'''
 
 security = HTTPBearer()
 
 def verify_token(credentials: HTTPAuthorizationCredentials = Depends(security)):
     token = credentials.credentials
     if token != os.getenv("API_TOKEN"):
+        logger.warning("Invalid or missing token")
         raise HTTPException(status_code=403, detail="Invalid or missing token")
 
 @app.post("/chat/completions", dependencies=[Depends(verify_token)])
 async def create_chat_completion(request: ChatCompletionRequest, req: Request):
     try:
-        messages = format_openai_messages(request.messages)
+        logger.debug(f"Received request: {request.json()}")
         client = AsyncOpenAI(api_key=os.getenv("OPENAI_API_KEY"))
         response = await client.chat.completions.create(
             model=request.model,
-            messages=messages,
-            temperature=request.temperature,
-            stream=request.stream
+            messages=request.messages,  # Directly use request messages
+            tool_choice=request.tool_choice if request.tools and request.tool_choice else None,
+            tools=request.tools if request.tools else None,
+            # modalities=request.modalities,
+            response_format=request.response_format,
+            stream=request.stream,
+            stream_options=request.stream_options
         )
 
-        async def generate():
-            try:
-                async for chunk in response:
-                    if chunk.choices[0].delta.content is not None:
-                        yield f"data: {chunk.choices[0].delta.content}\n\n"
-                yield "data: [DONE]\n\n"
-            except asyncio.CancelledError:
-                print("Request was cancelled")
-                raise
-
-        return StreamingResponse(generate(), media_type="text/event-stream")
+        if request.stream:
+            async def generate():
+                try:
+                    async for chunk in response:
+                        logger.info(f"Received chunk: {chunk}")
+                        yield f"data: {json.dumps(chunk.to_dict())}\n\n"
+                    yield "data: [DONE]\n\n"
+                except asyncio.CancelledError:
+                    logger.info("Request was cancelled")
+                    raise
+
+            return StreamingResponse(generate(), media_type="text/event-stream")
+        else:
+            result = await response
+            return result
     except asyncio.CancelledError:
-        print("Request was cancelled")
+        logger.info("Request was cancelled")
         raise HTTPException(status_code=499, detail="Request was cancelled")
     except Exception as e:
         traceback_str = ''.join(traceback.format_tb(e.__traceback__))
         error_message = f"{str(e)}\n{traceback_str}"
-        print(error_message)
+        logger.error(error_message)
         raise HTTPException(status_code=500, detail=error_message)
 
 if __name__ == "__main__":