Spaces:
Running
Running
File size: 14,385 Bytes
81e42f7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 |
import time
import logging
import sys
from aiocache import cached
from typing import Any, Optional
import random
import json
import inspect
import uuid
import asyncio
from fastapi import Request, status
from starlette.responses import Response, StreamingResponse, JSONResponse
from open_webui.models.users import UserModel
from open_webui.socket.main import (
sio,
get_event_call,
get_event_emitter,
)
from open_webui.functions import generate_function_chat_completion
from open_webui.routers.openai import (
generate_chat_completion as generate_openai_chat_completion,
)
from open_webui.routers.ollama import (
generate_chat_completion as generate_ollama_chat_completion,
)
from open_webui.routers.pipelines import (
process_pipeline_inlet_filter,
process_pipeline_outlet_filter,
)
from open_webui.models.functions import Functions
from open_webui.models.models import Models
from open_webui.utils.plugin import load_function_module_by_id
from open_webui.utils.models import get_all_models, check_model_access
from open_webui.utils.payload import convert_payload_openai_to_ollama
from open_webui.utils.response import (
convert_response_ollama_to_openai,
convert_streaming_response_ollama_to_openai,
)
from open_webui.utils.filter import (
get_sorted_filter_ids,
process_filter_functions,
)
from open_webui.env import SRC_LOG_LEVELS, GLOBAL_LOG_LEVEL, BYPASS_MODEL_ACCESS_CONTROL
logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL)
log = logging.getLogger(__name__)
log.setLevel(SRC_LOG_LEVELS["MAIN"])
async def generate_direct_chat_completion(
request: Request,
form_data: dict,
user: Any,
models: dict,
):
log.info("generate_direct_chat_completion")
metadata = form_data.pop("metadata", {})
user_id = metadata.get("user_id")
session_id = metadata.get("session_id")
request_id = str(uuid.uuid4()) # Generate a unique request ID
event_caller = get_event_call(metadata)
channel = f"{user_id}:{session_id}:{request_id}"
if form_data.get("stream"):
q = asyncio.Queue()
async def message_listener(sid, data):
"""
Handle received socket messages and push them into the queue.
"""
await q.put(data)
# Register the listener
sio.on(channel, message_listener)
# Start processing chat completion in background
res = await event_caller(
{
"type": "request:chat:completion",
"data": {
"form_data": form_data,
"model": models[form_data["model"]],
"channel": channel,
"session_id": session_id,
},
}
)
log.info(f"res: {res}")
if res.get("status", False):
# Define a generator to stream responses
async def event_generator():
nonlocal q
try:
while True:
data = await q.get() # Wait for new messages
if isinstance(data, dict):
if "done" in data and data["done"]:
break # Stop streaming when 'done' is received
yield f"data: {json.dumps(data)}\n\n"
elif isinstance(data, str):
yield data
except Exception as e:
log.debug(f"Error in event generator: {e}")
pass
# Define a background task to run the event generator
async def background():
try:
del sio.handlers["/"][channel]
except Exception as e:
pass
# Return the streaming response
return StreamingResponse(
event_generator(), media_type="text/event-stream", background=background
)
else:
raise Exception(str(res))
else:
res = await event_caller(
{
"type": "request:chat:completion",
"data": {
"form_data": form_data,
"model": models[form_data["model"]],
"channel": channel,
"session_id": session_id,
},
}
)
if "error" in res:
raise Exception(res["error"])
return res
async def generate_chat_completion(
request: Request,
form_data: dict,
user: Any,
bypass_filter: bool = False,
):
log.debug(f"generate_chat_completion: {form_data}")
if BYPASS_MODEL_ACCESS_CONTROL:
bypass_filter = True
if hasattr(request.state, "metadata"):
if "metadata" not in form_data:
form_data["metadata"] = request.state.metadata
else:
form_data["metadata"] = {
**form_data["metadata"],
**request.state.metadata,
}
if getattr(request.state, "direct", False) and hasattr(request.state, "model"):
models = {
request.state.model["id"]: request.state.model,
}
log.debug(f"direct connection to model: {models}")
else:
models = request.app.state.MODELS
model_id = form_data["model"]
if model_id not in models:
raise Exception("Model not found")
model = models[model_id]
if getattr(request.state, "direct", False):
return await generate_direct_chat_completion(
request, form_data, user=user, models=models
)
else:
# Check if user has access to the model
if not bypass_filter and user.role == "user":
try:
check_model_access(user, model)
except Exception as e:
raise e
if model.get("owned_by") == "arena":
model_ids = model.get("info", {}).get("meta", {}).get("model_ids")
filter_mode = model.get("info", {}).get("meta", {}).get("filter_mode")
if model_ids and filter_mode == "exclude":
model_ids = [
model["id"]
for model in list(request.app.state.MODELS.values())
if model.get("owned_by") != "arena" and model["id"] not in model_ids
]
selected_model_id = None
if isinstance(model_ids, list) and model_ids:
selected_model_id = random.choice(model_ids)
else:
model_ids = [
model["id"]
for model in list(request.app.state.MODELS.values())
if model.get("owned_by") != "arena"
]
selected_model_id = random.choice(model_ids)
form_data["model"] = selected_model_id
if form_data.get("stream") == True:
async def stream_wrapper(stream):
yield f"data: {json.dumps({'selected_model_id': selected_model_id})}\n\n"
async for chunk in stream:
yield chunk
response = await generate_chat_completion(
request, form_data, user, bypass_filter=True
)
return StreamingResponse(
stream_wrapper(response.body_iterator),
media_type="text/event-stream",
background=response.background,
)
else:
return {
**(
await generate_chat_completion(
request, form_data, user, bypass_filter=True
)
),
"selected_model_id": selected_model_id,
}
if model.get("pipe"):
# Below does not require bypass_filter because this is the only route the uses this function and it is already bypassing the filter
return await generate_function_chat_completion(
request, form_data, user=user, models=models
)
if model.get("owned_by") == "ollama":
# Using /ollama/api/chat endpoint
form_data = convert_payload_openai_to_ollama(form_data)
response = await generate_ollama_chat_completion(
request=request,
form_data=form_data,
user=user,
bypass_filter=bypass_filter,
)
if form_data.get("stream"):
response.headers["content-type"] = "text/event-stream"
return StreamingResponse(
convert_streaming_response_ollama_to_openai(response),
headers=dict(response.headers),
background=response.background,
)
else:
return convert_response_ollama_to_openai(response)
else:
return await generate_openai_chat_completion(
request=request,
form_data=form_data,
user=user,
bypass_filter=bypass_filter,
)
chat_completion = generate_chat_completion
async def chat_completed(request: Request, form_data: dict, user: Any):
if not request.app.state.MODELS:
await get_all_models(request, user=user)
if getattr(request.state, "direct", False) and hasattr(request.state, "model"):
models = {
request.state.model["id"]: request.state.model,
}
else:
models = request.app.state.MODELS
data = form_data
model_id = data["model"]
if model_id not in models:
raise Exception("Model not found")
model = models[model_id]
try:
data = await process_pipeline_outlet_filter(request, data, user, models)
except Exception as e:
return Exception(f"Error: {e}")
metadata = {
"chat_id": data["chat_id"],
"message_id": data["id"],
"session_id": data["session_id"],
"user_id": user.id,
}
extra_params = {
"__event_emitter__": get_event_emitter(metadata),
"__event_call__": get_event_call(metadata),
"__user__": {
"id": user.id,
"email": user.email,
"name": user.name,
"role": user.role,
},
"__metadata__": metadata,
"__request__": request,
"__model__": model,
}
try:
result, _ = await process_filter_functions(
request=request,
filter_ids=get_sorted_filter_ids(model),
filter_type="outlet",
form_data=data,
extra_params=extra_params,
)
return result
except Exception as e:
return Exception(f"Error: {e}")
async def chat_action(request: Request, action_id: str, form_data: dict, user: Any):
if "." in action_id:
action_id, sub_action_id = action_id.split(".")
else:
sub_action_id = None
action = Functions.get_function_by_id(action_id)
if not action:
raise Exception(f"Action not found: {action_id}")
if not request.app.state.MODELS:
await get_all_models(request, user=user)
if getattr(request.state, "direct", False) and hasattr(request.state, "model"):
models = {
request.state.model["id"]: request.state.model,
}
else:
models = request.app.state.MODELS
data = form_data
model_id = data["model"]
if model_id not in models:
raise Exception("Model not found")
model = models[model_id]
__event_emitter__ = get_event_emitter(
{
"chat_id": data["chat_id"],
"message_id": data["id"],
"session_id": data["session_id"],
"user_id": user.id,
}
)
__event_call__ = get_event_call(
{
"chat_id": data["chat_id"],
"message_id": data["id"],
"session_id": data["session_id"],
"user_id": user.id,
}
)
if action_id in request.app.state.FUNCTIONS:
function_module = request.app.state.FUNCTIONS[action_id]
else:
function_module, _, _ = load_function_module_by_id(action_id)
request.app.state.FUNCTIONS[action_id] = function_module
if hasattr(function_module, "valves") and hasattr(function_module, "Valves"):
valves = Functions.get_function_valves_by_id(action_id)
function_module.valves = function_module.Valves(**(valves if valves else {}))
if hasattr(function_module, "action"):
try:
action = function_module.action
# Get the signature of the function
sig = inspect.signature(action)
params = {"body": data}
# Extra parameters to be passed to the function
extra_params = {
"__model__": model,
"__id__": sub_action_id if sub_action_id is not None else action_id,
"__event_emitter__": __event_emitter__,
"__event_call__": __event_call__,
"__request__": request,
}
# Add extra params in contained in function signature
for key, value in extra_params.items():
if key in sig.parameters:
params[key] = value
if "__user__" in sig.parameters:
__user__ = {
"id": user.id,
"email": user.email,
"name": user.name,
"role": user.role,
}
try:
if hasattr(function_module, "UserValves"):
__user__["valves"] = function_module.UserValves(
**Functions.get_user_valves_by_id_and_user_id(
action_id, user.id
)
)
except Exception as e:
log.exception(f"Failed to get user values: {e}")
params = {**params, "__user__": __user__}
if inspect.iscoroutinefunction(action):
data = await action(**params)
else:
data = action(**params)
except Exception as e:
return Exception(f"Error: {e}")
return data
|