mirror of
https://github.com/arc53/DocsGPT.git
synced 2025-11-30 09:03:15 +00:00
feat: answer routes re-structure for better maintainability and reuse
This commit is contained in:
Siddhant Rai
0
application/api/answer/routes/__init__.py
Normal file
0
application/api/answer/routes/__init__.py
Normal file
103
application/api/answer/routes/answer.py
Normal file
103
application/api/answer/routes/answer.py
Normal file
@@ -0,0 +1,103 @@
|
||||
import logging
|
||||
import traceback
|
||||
|
||||
from flask import make_response, request
|
||||
from flask_restx import fields, Resource
|
||||
|
||||
from application.api import api
|
||||
|
||||
from application.api.answer.routes.base import answer_ns, BaseAnswerResource
|
||||
|
||||
from application.api.answer.services.stream_processor import StreamProcessor
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
@answer_ns.route("/api/answer")
|
||||
class AnswerResource(Resource, BaseAnswerResource):
|
||||
def __init__(self, *args, **kwargs):
|
||||
Resource.__init__(self, *args, **kwargs)
|
||||
BaseAnswerResource.__init__(self)
|
||||
|
||||
answer_model = answer_ns.model(
|
||||
"AnswerModel",
|
||||
{
|
||||
"question": fields.String(
|
||||
required=True, description="Question to be asked"
|
||||
),
|
||||
"history": fields.List(
|
||||
fields.String,
|
||||
required=False,
|
||||
description="Conversation history (only for new conversations)",
|
||||
),
|
||||
"conversation_id": fields.String(
|
||||
required=False,
|
||||
description="Existing conversation ID (loads history)",
|
||||
),
|
||||
"prompt_id": fields.String(
|
||||
required=False, default="default", description="Prompt ID"
|
||||
),
|
||||
"chunks": fields.Integer(
|
||||
required=False, default=2, description="Number of chunks"
|
||||
),
|
||||
"token_limit": fields.Integer(required=False, description="Token limit"),
|
||||
"retriever": fields.String(required=False, description="Retriever type"),
|
||||
"api_key": fields.String(required=False, description="API key"),
|
||||
"active_docs": fields.String(
|
||||
required=False, description="Active documents"
|
||||
),
|
||||
"isNoneDoc": fields.Boolean(
|
||||
required=False, description="Flag indicating if no document is used"
|
||||
),
|
||||
"save_conversation": fields.Boolean(
|
||||
required=False,
|
||||
default=True,
|
||||
description="Whether to save the conversation",
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
@api.expect(answer_model)
|
||||
@api.doc(description="Provide a response based on the question and retriever")
|
||||
def post(self):
|
||||
data = request.get_json()
|
||||
if error := self.validate_request(data):
|
||||
return error
|
||||
processor = StreamProcessor(data, None)
|
||||
try:
|
||||
processor.initialize()
|
||||
if not processor.decoded_token:
|
||||
return make_response({"error": "Unauthorized"}, 401)
|
||||
agent = processor.create_agent()
|
||||
retriever = processor.create_retriever()
|
||||
|
||||
stream = self.complete_stream(
|
||||
question=data["question"],
|
||||
agent=agent,
|
||||
retriever=retriever,
|
||||
conversation_id=processor.conversation_id,
|
||||
user_api_key=processor.agent_config.get("user_api_key"),
|
||||
decoded_token=processor.decoded_token,
|
||||
isNoneDoc=data.get("isNoneDoc"),
|
||||
index=None,
|
||||
should_save_conversation=data.get("save_conversation", True),
|
||||
)
|
||||
conversation_id, response, sources, tool_calls, thought, error = (
|
||||
self.process_response_stream(stream)
|
||||
)
|
||||
if error:
|
||||
return make_response({"error": error}, 400)
|
||||
result = {
|
||||
"conversation_id": conversation_id,
|
||||
"answer": response,
|
||||
"sources": sources,
|
||||
"tool_calls": tool_calls,
|
||||
"thought": thought,
|
||||
}
|
||||
except Exception as e:
|
||||
logger.error(
|
||||
f"/api/answer - error: {str(e)} - traceback: {traceback.format_exc()}",
|
||||
extra={"error": str(e), "traceback": traceback.format_exc()},
|
||||
)
|
||||
return make_response({"error": str(e)}, 500)
|
||||
return make_response(result, 200)
|
||||
226
application/api/answer/routes/base.py
Normal file
226
application/api/answer/routes/base.py
Normal file
@@ -0,0 +1,226 @@
|
||||
import datetime
|
||||
import json
|
||||
import logging
|
||||
from typing import Any, Dict, Generator, List, Optional
|
||||
|
||||
from flask import Response
|
||||
from flask_restx import Namespace
|
||||
|
||||
from application.api.answer.services.conversation_service import ConversationService
|
||||
|
||||
from application.core.mongo_db import MongoDB
|
||||
from application.core.settings import settings
|
||||
from application.llm.llm_creator import LLMCreator
|
||||
from application.utils import check_required_fields, get_gpt_model
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
answer_ns = Namespace("answer", description="Answer related operations", path="/")
|
||||
|
||||
|
||||
class BaseAnswerResource:
|
||||
"""Shared base class for answer endpoints"""
|
||||
|
||||
def __init__(self):
|
||||
mongo = MongoDB.get_client()
|
||||
db = mongo[settings.MONGO_DB_NAME]
|
||||
self.user_logs_collection = db["user_logs"]
|
||||
self.gpt_model = get_gpt_model()
|
||||
self.conversation_service = ConversationService()
|
||||
|
||||
def validate_request(
|
||||
self, data: Dict[str, Any], require_conversation_id: bool = False
|
||||
) -> Optional[Response]:
|
||||
"""Common request validation"""
|
||||
required_fields = ["question"]
|
||||
if require_conversation_id:
|
||||
required_fields.append("conversation_id")
|
||||
if missing_fields := check_required_fields(data, required_fields):
|
||||
return missing_fields
|
||||
return None
|
||||
|
||||
def complete_stream(
|
||||
self,
|
||||
question: str,
|
||||
agent: Any,
|
||||
retriever: Any,
|
||||
conversation_id: Optional[str],
|
||||
user_api_key: Optional[str],
|
||||
decoded_token: Dict[str, Any],
|
||||
isNoneDoc: bool = False,
|
||||
index: Optional[int] = None,
|
||||
should_save_conversation: bool = True,
|
||||
attachment_ids: Optional[List[str]] = None,
|
||||
agent_id: Optional[str] = None,
|
||||
is_shared_usage: bool = False,
|
||||
shared_token: Optional[str] = None,
|
||||
) -> Generator[str, None, None]:
|
||||
"""
|
||||
Generator function that streams the complete conversation response.
|
||||
|
||||
Args:
|
||||
question: The user's question
|
||||
agent: The agent instance
|
||||
retriever: The retriever instance
|
||||
conversation_id: Existing conversation ID
|
||||
user_api_key: User's API key if any
|
||||
decoded_token: Decoded JWT token
|
||||
isNoneDoc: Flag for document-less responses
|
||||
index: Index of message to update
|
||||
should_save_conversation: Whether to persist the conversation
|
||||
attachment_ids: List of attachment IDs
|
||||
agent_id: ID of agent used
|
||||
is_shared_usage: Flag for shared agent usage
|
||||
shared_token: Token for shared agent
|
||||
|
||||
Yields:
|
||||
Server-sent event strings
|
||||
"""
|
||||
try:
|
||||
response_full, thought, source_log_docs, tool_calls = "", "", [], []
|
||||
|
||||
for line in agent.gen(query=question, retriever=retriever):
|
||||
if "answer" in line:
|
||||
response_full += str(line["answer"])
|
||||
data = json.dumps({"type": "answer", "answer": line["answer"]})
|
||||
yield f"data: {data}\n\n"
|
||||
elif "sources" in line:
|
||||
truncated_sources = []
|
||||
source_log_docs = line["sources"]
|
||||
for source in line["sources"]:
|
||||
truncated_source = source.copy()
|
||||
if "text" in truncated_source:
|
||||
truncated_source["text"] = (
|
||||
truncated_source["text"][:100].strip() + "..."
|
||||
)
|
||||
truncated_sources.append(truncated_source)
|
||||
if truncated_sources:
|
||||
data = json.dumps(
|
||||
{"type": "source", "source": truncated_sources}
|
||||
)
|
||||
yield f"data: {data}\n\n"
|
||||
elif "tool_calls" in line:
|
||||
tool_calls = line["tool_calls"]
|
||||
elif "thought" in line:
|
||||
thought += line["thought"]
|
||||
data = json.dumps({"type": "thought", "thought": line["thought"]})
|
||||
yield f"data: {data}\n\n"
|
||||
elif "type" in line:
|
||||
data = json.dumps(line)
|
||||
yield f"data: {data}\n\n"
|
||||
if isNoneDoc:
|
||||
for doc in source_log_docs:
|
||||
doc["source"] = "None"
|
||||
llm = LLMCreator.create_llm(
|
||||
settings.LLM_PROVIDER,
|
||||
api_key=settings.API_KEY,
|
||||
user_api_key=user_api_key,
|
||||
decoded_token=decoded_token,
|
||||
)
|
||||
|
||||
if should_save_conversation:
|
||||
conversation_id = self.conversation_service.save_conversation(
|
||||
conversation_id,
|
||||
question,
|
||||
response_full,
|
||||
thought,
|
||||
source_log_docs,
|
||||
tool_calls,
|
||||
llm,
|
||||
self.gpt_model,
|
||||
decoded_token,
|
||||
index=index,
|
||||
api_key=user_api_key,
|
||||
agent_id=agent_id,
|
||||
is_shared_usage=is_shared_usage,
|
||||
shared_token=shared_token,
|
||||
attachment_ids=attachment_ids,
|
||||
)
|
||||
else:
|
||||
conversation_id = None
|
||||
# Send conversation ID
|
||||
|
||||
data = json.dumps({"type": "id", "id": str(conversation_id)})
|
||||
yield f"data: {data}\n\n"
|
||||
|
||||
# Log the interaction
|
||||
|
||||
retriever_params = retriever.get_params()
|
||||
self.user_logs_collection.insert_one(
|
||||
{
|
||||
"action": "stream_answer",
|
||||
"level": "info",
|
||||
"user": decoded_token.get("sub"),
|
||||
"api_key": user_api_key,
|
||||
"question": question,
|
||||
"response": response_full,
|
||||
"sources": source_log_docs,
|
||||
"retriever_params": retriever_params,
|
||||
"attachments": attachment_ids,
|
||||
"timestamp": datetime.datetime.now(datetime.timezone.utc),
|
||||
}
|
||||
)
|
||||
|
||||
# End of stream
|
||||
|
||||
data = json.dumps({"type": "end"})
|
||||
yield f"data: {data}\n\n"
|
||||
except Exception as e:
|
||||
logger.error(f"Error in stream: {str(e)}", exc_info=True)
|
||||
data = json.dumps(
|
||||
{
|
||||
"type": "error",
|
||||
"error": "Please try again later. We apologize for any inconvenience.",
|
||||
}
|
||||
)
|
||||
yield f"data: {data}\n\n"
|
||||
return
|
||||
|
||||
def process_response_stream(self, stream):
|
||||
"""Process the stream response for non-streaming endpoint"""
|
||||
conversation_id = ""
|
||||
response_full = ""
|
||||
source_log_docs = []
|
||||
tool_calls = []
|
||||
thought = ""
|
||||
stream_ended = False
|
||||
|
||||
for line in stream:
|
||||
try:
|
||||
event_data = line.replace("data: ", "").strip()
|
||||
event = json.loads(event_data)
|
||||
|
||||
if event["type"] == "id":
|
||||
conversation_id = event["id"]
|
||||
elif event["type"] == "answer":
|
||||
response_full += event["answer"]
|
||||
elif event["type"] == "source":
|
||||
source_log_docs = event["source"]
|
||||
elif event["type"] == "tool_calls":
|
||||
tool_calls = event["tool_calls"]
|
||||
elif event["type"] == "thought":
|
||||
thought = event["thought"]
|
||||
elif event["type"] == "error":
|
||||
logger.error(f"Error from stream: {event['error']}")
|
||||
return None, None, None, None, event["error"]
|
||||
elif event["type"] == "end":
|
||||
stream_ended = True
|
||||
except (json.JSONDecodeError, KeyError) as e:
|
||||
logger.warning(f"Error parsing stream event: {e}, line: {line}")
|
||||
continue
|
||||
if not stream_ended:
|
||||
logger.error("Stream ended unexpectedly without an 'end' event.")
|
||||
return None, None, None, None, "Stream ended unexpectedly"
|
||||
return (
|
||||
conversation_id,
|
||||
response_full,
|
||||
source_log_docs,
|
||||
tool_calls,
|
||||
thought,
|
||||
None,
|
||||
)
|
||||
|
||||
def error_stream_generate(self, err_response):
|
||||
data = json.dumps({"type": "error", "error": err_response})
|
||||
yield f"data: {data}\n\n"
|
||||
116
application/api/answer/routes/stream.py
Normal file
116
application/api/answer/routes/stream.py
Normal file
@@ -0,0 +1,116 @@
|
||||
import logging
|
||||
import traceback
|
||||
|
||||
from flask import make_response, request, Response
|
||||
from flask_restx import fields, Resource
|
||||
|
||||
from application.api import api
|
||||
|
||||
from application.api.answer.routes.base import answer_ns, BaseAnswerResource
|
||||
|
||||
from application.api.answer.services.stream_processor import StreamProcessor
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
@answer_ns.route("/stream")
|
||||
class StreamResource(Resource, BaseAnswerResource):
|
||||
def __init__(self, *args, **kwargs):
|
||||
Resource.__init__(self, *args, **kwargs)
|
||||
BaseAnswerResource.__init__(self)
|
||||
|
||||
stream_model = answer_ns.model(
|
||||
"StreamModel",
|
||||
{
|
||||
"question": fields.String(
|
||||
required=True, description="Question to be asked"
|
||||
),
|
||||
"history": fields.List(
|
||||
fields.String,
|
||||
required=False,
|
||||
description="Conversation history (only for new conversations)",
|
||||
),
|
||||
"conversation_id": fields.String(
|
||||
required=False,
|
||||
description="Existing conversation ID (loads history)",
|
||||
),
|
||||
"prompt_id": fields.String(
|
||||
required=False, default="default", description="Prompt ID"
|
||||
),
|
||||
"chunks": fields.Integer(
|
||||
required=False, default=2, description="Number of chunks"
|
||||
),
|
||||
"token_limit": fields.Integer(required=False, description="Token limit"),
|
||||
"retriever": fields.String(required=False, description="Retriever type"),
|
||||
"api_key": fields.String(required=False, description="API key"),
|
||||
"active_docs": fields.String(
|
||||
required=False, description="Active documents"
|
||||
),
|
||||
"isNoneDoc": fields.Boolean(
|
||||
required=False, description="Flag indicating if no document is used"
|
||||
),
|
||||
"index": fields.Integer(
|
||||
required=False, description="Index of the query to update"
|
||||
),
|
||||
"save_conversation": fields.Boolean(
|
||||
required=False,
|
||||
default=True,
|
||||
description="Whether to save the conversation",
|
||||
),
|
||||
"attachments": fields.List(
|
||||
fields.String, required=False, description="List of attachment IDs"
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
@api.expect(stream_model)
|
||||
@api.doc(description="Stream a response based on the question and retriever")
|
||||
def post(self):
|
||||
data = request.get_json()
|
||||
if error := self.validate_request(data, "index" in data):
|
||||
return error
|
||||
decoded_token = getattr(request, "decoded_token", None)
|
||||
if not decoded_token:
|
||||
return make_response({"error": "Unauthorized"}, 401)
|
||||
processor = StreamProcessor(data, decoded_token)
|
||||
try:
|
||||
processor.initialize()
|
||||
agent = processor.create_agent()
|
||||
retriever = processor.create_retriever()
|
||||
|
||||
return Response(
|
||||
self.complete_stream(
|
||||
question=data["question"],
|
||||
agent=agent,
|
||||
retriever=retriever,
|
||||
conversation_id=processor.conversation_id,
|
||||
user_api_key=processor.agent_config.get("user_api_key"),
|
||||
decoded_token=processor.decoded_token,
|
||||
isNoneDoc=data.get("isNoneDoc"),
|
||||
index=data.get("index"),
|
||||
should_save_conversation=data.get("save_conversation", True),
|
||||
attachment_ids=data.get("attachments", []),
|
||||
agent_id=data.get("agent_id"),
|
||||
is_shared_usage=processor.is_shared_usage,
|
||||
shared_token=processor.shared_token,
|
||||
),
|
||||
mimetype="text/event-stream",
|
||||
)
|
||||
except ValueError:
|
||||
message = "Malformed request body"
|
||||
logger.error(f"/stream - error: {message}")
|
||||
return Response(
|
||||
self.error_stream_generate(message),
|
||||
status=400,
|
||||
mimetype="text/event-stream",
|
||||
)
|
||||
except Exception as e:
|
||||
logger.error(
|
||||
f"/stream - error: {str(e)} - traceback: {traceback.format_exc()}",
|
||||
extra={"error": str(e), "traceback": traceback.format_exc()},
|
||||
)
|
||||
return Response(
|
||||
self.error_stream_generate("Unknown error occurred"),
|
||||
status=400,
|
||||
mimetype="text/event-stream",
|
||||
)
|
||||
Reference in New Issue
Block a user