File size: 9,059 Bytes
1f7ef29 5cecd79 e06c1a9 1f7ef29 bda3109 1f7ef29 5cecd79 1f7ef29 e1b2c3b 5cecd79 1f7ef29 a93ac0e 1f7ef29 944e367 01607ce 1a5834c 01607ce a93ac0e 01607ce 1f7ef29 f7ad6cb e06c1a9 5cecd79 f7ad6cb 1f7ef29 f7ad6cb 1f7ef29 a93ac0e f7ad6cb 1f7ef29 f7ad6cb 1f7ef29 944e367 1f7ef29 f7ad6cb 1f7ef29 a93ac0e 1a5834c a93ac0e 1a5834c f7ad6cb 1a5834c 6fc1306 1f7ef29 f7ad6cb e1b2c3b 1f7ef29 f7ad6cb 1f7ef29 a660cc0 944e367 1f7ef29 c257815 944e367 c257815 1f7ef29 c257815 1f7ef29 c257815 1f7ef29 5cecd79 1f7ef29 a93ac0e 1f7ef29 944e367 1f7ef29 944e367 1f7ef29 f7ad6cb 1f7ef29 a93ac0e 944e367 a93ac0e 944e367 a93ac0e 1a5834c f7ad6cb e1b2c3b a93ac0e 1f7ef29 f7ad6cb a660cc0 944e367 a660cc0 1f7ef29 c257815 944e367 c257815 1f7ef29 c257815 1f7ef29 c257815 1f7ef29 5cecd79 1f7ef29 5cecd79 1f7ef29 944e367 1f7ef29 944e367 1f7ef29 a93ac0e 1f7ef29 f7ad6cb 1f7ef29 41c59b2 |
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 |
from datetime import datetime
import json
import uuid
import asyncio
import random
import string
from typing import Any, Dict, Optional
import httpx
from fastapi import HTTPException
from api.config import (
MODEL_MAPPING,
get_headers_api_chat,
get_headers_chat,
BASE_URL,
AGENT_MODE,
TRENDING_AGENT_MODE,
MODEL_PREFIXES,
MODEL_REFERERS,
)
from api.models import ChatRequest
from api.logger import (
log_generated_chat_id_with_referer,
log_model_delay,
log_http_error,
log_request_error,
log_strip_prefix,
)
# Helper function to generate a random alphanumeric chat ID
def generate_chat_id(length: int = 7) -> str:
characters = string.ascii_letters + string.digits
return ''.join(random.choices(characters, k=length))
# Helper function to create chat completion data
def create_chat_completion_data(
content: str, model: str, timestamp: int, finish_reason: Optional[str] = None
) -> Dict[str, Any]:
return {
"id": f"chatcmpl-{uuid.uuid4()}",
"object": "chat.completion.chunk",
"created": timestamp,
"model": model,
"choices": [
{
"index": 0,
"delta": {"content": content, "role": "assistant"},
"finish_reason": finish_reason,
}
],
"usage": None,
}
# Convert message to dictionary format with optional model prefix and image data
def message_to_dict(message, model_prefix: Optional[str] = None):
content = message.content
if model_prefix:
content = f"{model_prefix} {content}"
if isinstance(message.content, str):
return {"role": message.role, "content": content}
elif isinstance(message.content, list) and len(message.content) == 2:
return {
"role": message.role,
"content": content,
"data": {
"imageBase64": message.content[1]["image_url"]["url"],
"fileText": "",
"title": "snapshot",
},
}
else:
return {"role": message.role, "content": content}
# Function to strip model prefix from content if present
def strip_model_prefix(content: str, model_prefix: Optional[str] = None) -> str:
"""Remove the model prefix from the response content if present."""
if model_prefix and content.startswith(model_prefix):
log_strip_prefix(model_prefix, content)
return content[len(model_prefix):].strip()
return content
# Streaming response processing with headers from config.py
async def process_streaming_response(request: ChatRequest):
chat_id = generate_chat_id() if request.model in MODEL_REFERERS else None
referer_path = MODEL_REFERERS.get(request.model, "")
referer_url = f"{BASE_URL}/chat/{chat_id}?model={request.model}" if chat_id else BASE_URL
# Log with chat ID, model, and referer URL if applicable
log_generated_chat_id_with_referer(chat_id, request.model, referer_url)
agent_mode = AGENT_MODE.get(request.model, {})
trending_agent_mode = TRENDING_AGENT_MODE.get(request.model, {})
model_prefix = MODEL_PREFIXES.get(request.model, "")
# Generate headers for API chat request with dynamic Referer
headers_api_chat = get_headers_api_chat(referer_url)
# Introduce delay for 'o1-preview' model
if request.model == 'o1-preview':
delay_seconds = random.randint(20, 60)
log_model_delay(delay_seconds, request.model, chat_id)
await asyncio.sleep(delay_seconds)
json_data = {
"agentMode": agent_mode,
"clickedAnswer2": False,
"clickedAnswer3": False,
"clickedForceWebSearch": False,
"codeModelMode": True,
"githubToken": None,
"id": chat_id,
"isChromeExt": False,
"isMicMode": False,
"maxTokens": request.max_tokens,
"messages": [message_to_dict(msg, model_prefix=model_prefix) for msg in request.messages],
"mobileClient": False,
"playgroundTemperature": request.temperature,
"playgroundTopP": request.top_p,
"previewToken": None,
"trendingAgentMode": trending_agent_mode,
"userId": None,
"userSelectedModel": MODEL_MAPPING.get(request.model, request.model),
"userSystemPrompt": None,
"validated": "69783381-2ce4-4dbd-ac78-35e9063feabc",
"visitFromDelta": False,
}
async with httpx.AsyncClient() as client:
try:
async with client.stream(
"POST",
f"{BASE_URL}/api/chat",
headers=headers_api_chat,
json=json_data,
timeout=100,
) as response:
response.raise_for_status()
async for line in response.aiter_lines():
timestamp = int(datetime.now().timestamp())
if line:
content = line.lstrip("$@$v=undefined-rv1$@$")
cleaned_content = strip_model_prefix(content, model_prefix)
yield f"data: {json.dumps(create_chat_completion_data(cleaned_content, request.model, timestamp))}\n\n"
yield f"data: {json.dumps(create_chat_completion_data('', request.model, timestamp, 'stop'))}\n\n"
yield "data: [DONE]\n\n"
except httpx.HTTPStatusError as e:
log_http_error(e, chat_id)
raise HTTPException(status_code=e.response.status_code, detail=str(e))
except httpx.RequestError as e:
log_request_error(e, chat_id)
raise HTTPException(status_code=500, detail=str(e))
# Non-streaming response processing with headers from config.py
async def process_non_streaming_response(request: ChatRequest):
chat_id = generate_chat_id() if request.model in MODEL_REFERERS else None
referer_path = MODEL_REFERERS.get(request.model, "")
referer_url = f"{BASE_URL}/chat/{chat_id}?model={request.model}" if chat_id else BASE_URL
# Log with chat ID, model, and referer URL if applicable
log_generated_chat_id_with_referer(chat_id, request.model, referer_url)
agent_mode = AGENT_MODE.get(request.model, {})
trending_agent_mode = TRENDING_AGENT_MODE.get(request.model, {})
model_prefix = MODEL_PREFIXES.get(request.model, "")
# Generate headers for API chat request and secondary chat request with dynamic Referer
headers_api_chat = get_headers_api_chat(referer_url)
headers_chat = get_headers_chat(referer_url, next_action=str(uuid.uuid4()), next_router_state_tree=json.dumps([""]))
# Introduce delay for 'o1-preview' model
if request.model == 'o1-preview':
delay_seconds = random.randint(20, 60)
log_model_delay(delay_seconds, request.model, chat_id)
await asyncio.sleep(delay_seconds)
json_data = {
"agentMode": agent_mode,
"clickedAnswer2": False,
"clickedAnswer3": False,
"clickedForceWebSearch": False,
"codeModelMode": True,
"githubToken": None,
"id": chat_id,
"isChromeExt": False,
"isMicMode": False,
"maxTokens": request.max_tokens,
"messages": [message_to_dict(msg, model_prefix=model_prefix) for msg in request.messages],
"mobileClient": False,
"playgroundTemperature": request.temperature,
"playgroundTopP": request.top_p,
"previewToken": None,
"trendingAgentMode": trending_agent_mode,
"userId": None,
"userSelectedModel": MODEL_MAPPING.get(request.model, request.model),
"userSystemPrompt": None,
"validated": "69783381-2ce4-4dbd-ac78-35e9063feabc",
"visitFromDelta": False,
}
full_response = ""
async with httpx.AsyncClient() as client:
try:
async with client.stream(
method="POST", url=f"{BASE_URL}/api/chat", headers=headers_api_chat, json=json_data
) as response:
response.raise_for_status()
async for chunk in response.aiter_text():
full_response += chunk
except httpx.HTTPStatusError as e:
log_http_error(e, chat_id)
raise HTTPException(status_code=e.response.status_code, detail=str(e))
except httpx.RequestError as e:
log_request_error(e, chat_id)
raise HTTPException(status_code=500, detail=str(e))
if full_response.startswith("$@$v=undefined-rv1$@$"):
full_response = full_response[21:]
# Strip the model prefix from the full response
cleaned_full_response = strip_model_prefix(full_response, model_prefix)
return {
"id": f"chatcmpl-{uuid.uuid4()}",
"object": "chat.completion",
"created": int(datetime.now().timestamp()),
"model": request.model,
"choices": [
{
"index": 0,
"message": {"role": "assistant", "content": cleaned_full_response},
"finish_reason": "stop",
}
],
"usage": None,
}
|