746 lines
41 KiB
Python
746 lines
41 KiB
Python
import os
|
||
from openai import OpenAI, APITimeoutError, APIConnectionError, RateLimitError, APIStatusError
|
||
import time
|
||
import random
|
||
import traceback
|
||
import logging
|
||
import tiktoken
|
||
import asyncio
|
||
from asyncio import TimeoutError as AsyncTimeoutError
|
||
|
||
# Configure basic logging for this module (or rely on root logger config)
|
||
# logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
|
||
# logger = logging.getLogger(__name__) # Alternative: use named logger
|
||
|
||
# Constants
|
||
MAX_RETRIES = 3 # Maximum number of retries for API calls
|
||
INITIAL_BACKOFF = 1 # Initial backoff time in seconds
|
||
MAX_BACKOFF = 16 # Maximum backoff time in seconds
|
||
STREAM_CHUNK_TIMEOUT = 10 # Timeout in seconds for receiving a chunk in stream
|
||
|
||
# 自定义超时异常
|
||
class Timeout(Exception):
|
||
"""Raised when a stream chunk timeout occurs."""
|
||
def __init__(self, message="Stream chunk timeout occurred."):
|
||
self.message = message
|
||
super().__init__(self.message)
|
||
|
||
def __str__(self):
|
||
return self.message
|
||
|
||
class AI_Agent():
|
||
"""AI代理类，负责与AI模型交互生成文本内容"""
|
||
|
||
def __init__(self, base_url, model_name, api, timeout=30, max_retries=3, stream_chunk_timeout=10):
|
||
"""
|
||
初始化 AI 代理。
|
||
|
||
Args:
|
||
base_url (str): 模型服务的基础 URL 或预设名称 ('deepseek', 'vllm')。
|
||
model_name (str): 要使用的模型名称。
|
||
api (str): API 密钥。
|
||
timeout (int, optional): 单次 API 请求的超时时间 (秒)。 Defaults to 30.
|
||
max_retries (int, optional): API 请求失败时的最大重试次数。 Defaults to 3.
|
||
stream_chunk_timeout (int, optional): 流式响应中两个数据块之间的最大等待时间 (秒)。 Defaults to 10.
|
||
"""
|
||
logging.info("Initializing AI Agent")
|
||
self.url_list = {
|
||
"ali": "https://dashscope.aliyuncs.com/compatible-mode/v1",
|
||
"kimi": "https://api.moonshot.cn/v1",
|
||
"doubao": "https://ark.cn-beijing.volces.com/api/v3/",
|
||
"deepseek": "https://api.deepseek.com",
|
||
"vllm": "http://localhost:8000/v1",
|
||
}
|
||
|
||
self.base_url = self.url_list.get(base_url, base_url)
|
||
self.api = api
|
||
self.model_name = model_name
|
||
self.timeout = timeout
|
||
self.max_retries = max_retries
|
||
self.stream_chunk_timeout = stream_chunk_timeout
|
||
|
||
logging.info(f"AI Agent Settings: base_url={self.base_url}, model={self.model_name}, timeout={self.timeout}s, max_retries={self.max_retries}, stream_chunk_timeout={self.stream_chunk_timeout}s")
|
||
|
||
self.client = OpenAI(
|
||
api_key=self.api,
|
||
base_url=self.base_url,
|
||
timeout=self.timeout
|
||
)
|
||
|
||
# try:
|
||
# self.encoding = tiktoken.encoding_for_model(self.model_name)
|
||
# except KeyError:
|
||
# logging.warning(f"Encoding for model '{self.model_name}' not found. Using 'cl100k_base' encoding.")
|
||
# self.encoding = tiktoken.get_encoding("cl100k_base")
|
||
|
||
def generate_text(self, system_prompt, user_prompt, temperature, top_p, presence_penalty):
|
||
"""生成文本内容，并返回完整响应和token估计值"""
|
||
logging.info("Starting text generation process...")
|
||
# logging.debug(f"System Prompt (first 100): {system_prompt[:100]}...")
|
||
# logging.debug(f"User Prompt (first 100): {user_prompt[:100]}...") # Avoid logging potentially huge prompts
|
||
logging.info(f"Generation Params: temp={temperature}, top_p={top_p}, presence_penalty={presence_penalty}")
|
||
|
||
retry_count = 0
|
||
max_retry_wait = 10 # Max wait time between retries
|
||
full_response = ""
|
||
|
||
while retry_count <= self.max_retries:
|
||
call_start_time = None # Initialize start time
|
||
try:
|
||
# --- Added Logging ---
|
||
user_prompt_size = len(user_prompt)
|
||
logging.info(f"Attempt {retry_count + 1}/{self.max_retries + 1}: Preparing API request. User prompt size: {user_prompt_size} chars.")
|
||
call_start_time = time.time()
|
||
# --- End Added Logging ---
|
||
|
||
response = self.client.chat.completions.create(
|
||
model=self.model_name,
|
||
messages=[{"role": "system", "content": system_prompt},
|
||
{"role": "user", "content": user_prompt}],
|
||
temperature=temperature,
|
||
top_p=top_p,
|
||
presence_penalty=presence_penalty,
|
||
stream=False, # Ensure this is False for non-streaming method
|
||
max_tokens=8192,
|
||
timeout=self.timeout,
|
||
extra_body={
|
||
"repetition_penalty": 1.05,
|
||
},
|
||
)
|
||
|
||
# --- Added Logging ---
|
||
call_end_time = time.time()
|
||
logging.info(f"Attempt {retry_count + 1}/{self.max_retries + 1}: API request function returned successfully after {call_end_time - call_start_time:.2f} seconds.")
|
||
# --- End Added Logging ---
|
||
|
||
if response.choices and response.choices[0].message:
|
||
full_response = response.choices[0].message.content
|
||
logging.info(f"Received successful response. Content length: {len(full_response)} chars.")
|
||
break # Success, exit retry loop
|
||
else:
|
||
logging.warning("API response structure unexpected or empty content.")
|
||
full_response = "[Error: Empty or invalid response structure]"
|
||
# Decide if this specific case should retry or fail immediately
|
||
retry_count += 1 # Example: Treat as retryable
|
||
if retry_count <= self.max_retries:
|
||
wait_time = min(2 ** retry_count + random.random(), max_retry_wait)
|
||
logging.warning(f"Retrying due to unexpected response structure ({retry_count}/{self.max_retries}), waiting {wait_time:.2f}s...")
|
||
time.sleep(wait_time)
|
||
continue
|
||
|
||
except (APITimeoutError, APIConnectionError, RateLimitError, APIStatusError) as e:
|
||
# --- Added Logging ---
|
||
if call_start_time:
|
||
call_fail_time = time.time()
|
||
logging.warning(f"Attempt {retry_count + 1}/{self.max_retries + 1}: API call failed/timed out after {call_fail_time - call_start_time:.2f} seconds.")
|
||
else:
|
||
logging.warning(f"Attempt {retry_count + 1}/{self.max_retries + 1}: API call failed before or during initiation.")
|
||
# --- End Added Logging ---
|
||
|
||
logging.warning(f"API Error occurred: {e}")
|
||
should_retry = False
|
||
if isinstance(e, (APITimeoutError, APIConnectionError, RateLimitError)):
|
||
should_retry = True
|
||
elif isinstance(e, APIStatusError) and e.status_code >= 500:
|
||
should_retry = True
|
||
|
||
if should_retry:
|
||
retry_count += 1
|
||
if retry_count <= self.max_retries:
|
||
wait_time = min(2 ** retry_count + random.random(), max_retry_wait)
|
||
logging.warning(f"Retrying API call ({retry_count}/{self.max_retries}) after error, waiting {wait_time:.2f}s...")
|
||
time.sleep(wait_time)
|
||
else:
|
||
logging.error(f"Max retries ({self.max_retries}) reached for API errors. Aborting.")
|
||
return "请求失败，无法生成内容。", 0
|
||
else:
|
||
logging.error(f"Non-retriable API error: {e}. Aborting.")
|
||
return "请求失败，发生不可重试错误。", 0
|
||
except Exception as e:
|
||
logging.exception(f"Unexpected error during API call setup/execution:")
|
||
retry_count += 1
|
||
if retry_count <= self.max_retries:
|
||
wait_time = min(2 ** retry_count + random.random(), max_retry_wait)
|
||
logging.warning(f"Retrying API call ({retry_count}/{self.max_retries}) after unexpected error, waiting {wait_time:.2f}s...")
|
||
time.sleep(wait_time)
|
||
else:
|
||
logging.error(f"Max retries ({self.max_retries}) reached after unexpected errors. Aborting.")
|
||
return "请求失败，发生未知错误。", 0
|
||
|
||
logging.info("Text generation completed.")
|
||
estimated_tokens = len(full_response.split()) * 1.3
|
||
return full_response, estimated_tokens
|
||
|
||
def read_folder(self, file_folder):
|
||
"""读取指定文件夹下的所有文件内容"""
|
||
if not os.path.exists(file_folder):
|
||
logging.warning(f"Referenced folder does not exist: {file_folder}")
|
||
return ""
|
||
|
||
context = ""
|
||
try:
|
||
for file in os.listdir(file_folder):
|
||
file_path = os.path.join(file_folder, file)
|
||
if os.path.isfile(file_path):
|
||
try:
|
||
with open(file_path, "r", encoding="utf-8") as f:
|
||
context += f"文件名: {file}\n"
|
||
context += f.read()
|
||
context += "\n\n"
|
||
except Exception as read_err:
|
||
logging.error(f"Failed to read file {file_path}: {read_err}")
|
||
except Exception as list_err:
|
||
logging.error(f"Failed to list directory {file_folder}: {list_err}")
|
||
return context
|
||
|
||
def work(self, system_prompt, user_prompt, file_folder, temperature, top_p, presence_penalty):
|
||
"""完整的工作流程：生成文本并返回结果"""
|
||
logging.info(f"Starting 'work' process. File folder: {file_folder}")
|
||
if file_folder:
|
||
logging.info(f"Reading context from folder: {file_folder}")
|
||
context = self.read_folder(file_folder)
|
||
if context:
|
||
user_prompt = f"{user_prompt.strip()}\n\n--- 参考资料 ---\n{context.strip()}"
|
||
else:
|
||
logging.warning(f"Folder {file_folder} provided but no content read.")
|
||
|
||
time_start = time.time()
|
||
result = self.generate_text_stream(system_prompt, user_prompt, temperature, top_p, presence_penalty)
|
||
# result, tokens = self.generate_text(system_prompt, user_prompt, temperature, top_p, presence_penalty)
|
||
|
||
time_end = time.time()
|
||
time_cost = time_end - time_start
|
||
tokens = len(result) * 1.3
|
||
logging.info(f"'work' completed in {time_cost:.2f}s. Estimated tokens: {tokens}")
|
||
|
||
return result, tokens, time_cost
|
||
|
||
def close(self):
|
||
try:
|
||
logging.info("Closing AI Agent (client resources will be garbage collected).")
|
||
self.client = None
|
||
except Exception as e:
|
||
logging.error(f"Error during AI Agent close: {e}")
|
||
|
||
# --- Streaming Methods ---
|
||
def generate_text_stream(self, system_prompt, user_prompt, temperature=0.7, top_p=0.9, presence_penalty=0.0):
|
||
"""
|
||
生成文本流，但返回的是完整的响应而非生成器。
|
||
此方法适用于简单的API调用，不需要实时处理响应的场景。
|
||
|
||
Args:
|
||
system_prompt: 系统提示词
|
||
user_prompt: 用户提示词
|
||
temperature: 温度参数
|
||
top_p: 核采样参数
|
||
presence_penalty: 存在惩罚参数
|
||
|
||
Returns:
|
||
str: 生成的完整文本响应
|
||
|
||
Raises:
|
||
Exception: 如果API调用在所有重试后仍然失败
|
||
"""
|
||
messages = [
|
||
{"role": "system", "content": system_prompt},
|
||
{"role": "user", "content": user_prompt},
|
||
]
|
||
logging.info(f"Generating text stream using model: {self.model_name}")
|
||
|
||
retries = 0
|
||
backoff_time = INITIAL_BACKOFF
|
||
last_exception = None
|
||
global_timeout = self.timeout * 2 # 设置全局超时，是API超时的两倍
|
||
partial_response = "" # 用于存储部分响应
|
||
|
||
while retries < self.max_retries:
|
||
try:
|
||
logging.debug(f"Attempt {retries + 1}/{self.max_retries} to generate text stream.")
|
||
client = OpenAI(
|
||
api_key=self.api,
|
||
base_url=self.base_url,
|
||
timeout=self.timeout,
|
||
max_retries=0 # 禁用内置重试，使用我们自己的重试逻辑
|
||
)
|
||
|
||
stream = client.chat.completions.create(
|
||
model=self.model_name,
|
||
messages=messages,
|
||
temperature=temperature,
|
||
top_p=top_p,
|
||
presence_penalty=presence_penalty,
|
||
stream=True,
|
||
)
|
||
|
||
response_text = ""
|
||
stream_start_time = time.time()
|
||
last_chunk_time = time.time()
|
||
received_any_chunk = False # 标记是否收到过任何块
|
||
|
||
try:
|
||
for chunk in stream:
|
||
current_time = time.time()
|
||
|
||
# 检查流块超时：只有在已经开始接收数据后才检查
|
||
if received_any_chunk and current_time - last_chunk_time > self.stream_chunk_timeout:
|
||
logging.warning(f"Stream chunk timeout: No new chunk received for {self.stream_chunk_timeout} seconds after previous chunk.")
|
||
if response_text:
|
||
response_text += "\n\n[注意: 流式传输中断，内容可能不完整]"
|
||
partial_response = response_text
|
||
return partial_response
|
||
raise Timeout(f"Stream stalled: No new chunk received for {self.stream_chunk_timeout} seconds after previous chunk.")
|
||
|
||
# 检查首次响应超时：如果很长时间没收到第一个块
|
||
if not received_any_chunk and current_time - stream_start_time > self.timeout:
|
||
logging.warning(f"Initial response timeout: No chunk received in {self.timeout} seconds since stream started.")
|
||
raise Timeout(f"No initial response received for {self.timeout} seconds.")
|
||
|
||
# 检查全局超时(保留作为安全措施)
|
||
if current_time - stream_start_time > global_timeout:
|
||
logging.warning(f"Global timeout reached after {global_timeout} seconds.")
|
||
if response_text:
|
||
response_text += "\n\n[注意: 由于全局超时，内容可能不完整]"
|
||
partial_response = response_text
|
||
return partial_response
|
||
else:
|
||
logging.error("Global timeout with no content received.")
|
||
raise Timeout(f"Global timeout after {global_timeout} seconds with no content.")
|
||
|
||
# 处理接收到的数据块
|
||
if chunk.choices and chunk.choices[0].delta and chunk.choices[0].delta.content:
|
||
content = chunk.choices[0].delta.content
|
||
response_text += content
|
||
received_any_chunk = True # 标记已收到块
|
||
last_chunk_time = current_time # 更新最后块时间戳
|
||
|
||
logging.info("Stream completed successfully.")
|
||
return response_text # 成功完成
|
||
|
||
except Timeout as e:
|
||
logging.warning(f"Stream timeout: {e}. Retrying if possible ({retries + 1}/{self.max_retries}).")
|
||
if response_text:
|
||
partial_response = response_text
|
||
last_exception = e
|
||
except (APITimeoutError, APIConnectionError, RateLimitError) as e:
|
||
logging.warning(f"API error during streaming: {type(e).__name__} - {e}. Retrying if possible.")
|
||
if response_text:
|
||
partial_response = response_text
|
||
last_exception = e
|
||
except Exception as e:
|
||
logging.error(f"Unexpected error during streaming: {traceback.format_exc()}")
|
||
if response_text:
|
||
partial_response = response_text
|
||
response_text += f"\n\n[注意: 由于错误中断，内容可能不完整: {str(e)}]"
|
||
return response_text
|
||
last_exception = e
|
||
|
||
# 执行重试逻辑
|
||
retries += 1
|
||
if retries < self.max_retries:
|
||
logging.info(f"Retrying stream in {backoff_time} seconds...")
|
||
time.sleep(backoff_time + random.uniform(0, 1)) # 添加随机抖动
|
||
backoff_time = min(backoff_time * 2, MAX_BACKOFF)
|
||
else:
|
||
logging.error(f"Stream generation failed after {self.max_retries} retries.")
|
||
# 如果已经获取了部分响应，返回并标记不完整
|
||
if partial_response:
|
||
logging.info(f"Returning partial response of length {len(partial_response)}")
|
||
return partial_response + "\n\n[注意: 达到最大重试次数，内容可能不完整]"
|
||
# 如果没有获取到任何内容，抛出异常
|
||
raise last_exception or Exception("Stream generation failed after max retries with no content.")
|
||
|
||
except (Timeout, APITimeoutError, APIConnectionError, RateLimitError) as e:
|
||
retries += 1
|
||
last_exception = e
|
||
logging.warning(f"Attempt {retries}/{self.max_retries} failed: {type(e).__name__} - {e}")
|
||
if retries < self.max_retries:
|
||
logging.info(f"Retrying in {backoff_time} seconds...")
|
||
time.sleep(backoff_time + random.uniform(0, 1))
|
||
backoff_time = min(backoff_time * 2, MAX_BACKOFF)
|
||
else:
|
||
logging.error(f"API call failed after {self.max_retries} retries.")
|
||
# 如果已经获取了部分响应，返回并标记不完整
|
||
if partial_response:
|
||
return partial_response + f"\n\n[注意: API调用失败，内容可能不完整: {str(e)}]"
|
||
return f"[生成内容失败: {str(last_exception)}]"
|
||
except Exception as e:
|
||
logging.error(f"Unexpected error setting up stream: {traceback.format_exc()}")
|
||
# 如果已经获取了部分响应，返回并标记不完整
|
||
if partial_response:
|
||
return partial_response + f"\n\n[注意: 意外错误，内容可能不完整: {str(e)}]"
|
||
return f"[生成内容失败: {str(e)}]"
|
||
|
||
# 作为安全措施，虽然通常不应该到达此处
|
||
logging.error("Exited stream generation loop unexpectedly.")
|
||
# 如果已经获取了部分响应，返回并标记不完整
|
||
if partial_response:
|
||
return partial_response + "\n\n[注意: 意外退出流处理，内容可能不完整]"
|
||
return "[生成内容失败: 超过最大重试次数]"
|
||
|
||
def generate_text_stream_with_callback(self, system_prompt, user_prompt, callback,
|
||
temperature=0.7, top_p=0.9, presence_penalty=0.0, accumulate=False):
|
||
"""
|
||
Generates text based on prompts using a streaming connection with a callback.
|
||
|
||
Args:
|
||
system_prompt: The system prompt for the AI.
|
||
user_prompt: The user prompt for the AI.
|
||
callback: Function to call with each chunk of text (callback(chunk, accumulated_response))
|
||
temperature: Sampling temperature, defaults to 0.7.
|
||
top_p: Nucleus sampling parameter, defaults to 0.9.
|
||
presence_penalty: Presence penalty parameter, defaults to 0.0.
|
||
accumulate: If True, pass the accumulated response to the callback function, defaults to False.
|
||
|
||
Returns:
|
||
str: The complete generated text.
|
||
|
||
Raises:
|
||
Exception: If the API call fails after all retries.
|
||
"""
|
||
messages = [
|
||
{"role": "system", "content": system_prompt},
|
||
{"role": "user", "content": user_prompt},
|
||
]
|
||
logging.info(f"Generating text stream with callback, model: {self.model_name}")
|
||
|
||
retries = 0
|
||
backoff_time = INITIAL_BACKOFF
|
||
last_exception = None
|
||
full_response = ""
|
||
global_timeout = self.timeout * 2 # 设置全局超时，是API超时的两倍
|
||
|
||
while retries < self.max_retries:
|
||
try:
|
||
logging.debug(f"Attempt {retries + 1}/{self.max_retries} to generate text stream with callback.")
|
||
stream = self.client.chat.completions.create(
|
||
model=self.model_name,
|
||
messages=messages,
|
||
temperature=temperature,
|
||
top_p=top_p,
|
||
presence_penalty=presence_penalty,
|
||
stream=True,
|
||
timeout=self.timeout
|
||
)
|
||
|
||
chunk_iterator = iter(stream)
|
||
last_chunk_time = time.time()
|
||
start_time = time.time() # 记录开始时间用于全局超时检查
|
||
received_any_chunk = False # 标记是否收到过任何块
|
||
|
||
while True:
|
||
try:
|
||
current_time = time.time()
|
||
|
||
# 检查流块超时：只有在已经开始接收数据后才检查
|
||
if received_any_chunk and current_time - last_chunk_time > self.stream_chunk_timeout:
|
||
logging.warning(f"Stream chunk timeout: No new chunk received for {self.stream_chunk_timeout} seconds after previous chunk.")
|
||
if full_response:
|
||
logging.info(f"Returning partial response of length {len(full_response)} due to chunk timeout.")
|
||
# 通知回调函数超时情况
|
||
timeout_msg = "\n\n[注意: 流式传输中断，内容可能不完整]"
|
||
callback(timeout_msg, full_response + timeout_msg if accumulate else None)
|
||
return full_response + timeout_msg
|
||
raise Timeout(f"Stream stalled: No new chunk received for {self.stream_chunk_timeout} seconds after previous chunk.")
|
||
|
||
# 检查首次响应超时：如果很长时间没收到第一个块
|
||
if not received_any_chunk and current_time - start_time > self.timeout:
|
||
logging.warning(f"Initial response timeout: No chunk received in {self.timeout} seconds since stream started.")
|
||
raise Timeout(f"No initial response received for {self.timeout} seconds.")
|
||
|
||
# 检查全局超时(保留作为安全措施)
|
||
if current_time - start_time > global_timeout:
|
||
logging.warning(f"Global timeout reached after {global_timeout} seconds.")
|
||
if full_response:
|
||
logging.info(f"Returning partial response of length {len(full_response)} due to global timeout.")
|
||
# 通知回调函数超时情况
|
||
timeout_msg = "\n\n[注意: 由于全局超时，内容可能不完整]"
|
||
callback(timeout_msg, full_response + timeout_msg if accumulate else None)
|
||
return full_response + timeout_msg
|
||
else:
|
||
logging.error("Global timeout with no content received.")
|
||
raise Timeout(f"Global timeout after {global_timeout} seconds with no content.")
|
||
|
||
# 获取下一个数据块
|
||
chunk = next(chunk_iterator)
|
||
|
||
# 处理接收到的数据块
|
||
if chunk.choices and chunk.choices[0].delta and chunk.choices[0].delta.content:
|
||
content = chunk.choices[0].delta.content
|
||
full_response += content
|
||
callback(content, full_response if accumulate else None)
|
||
received_any_chunk = True # 标记已收到块
|
||
last_chunk_time = current_time # 更新最后块时间戳
|
||
elif chunk.choices and chunk.choices[0].finish_reason == 'stop':
|
||
logging.info("Stream with callback finished normally.")
|
||
return full_response
|
||
|
||
except StopIteration:
|
||
logging.info("Stream iterator with callback exhausted normally.")
|
||
return full_response
|
||
except Timeout as e:
|
||
logging.warning(f"Stream timeout: {e}. Retrying if possible ({retries + 1}/{self.max_retries}).")
|
||
last_exception = e
|
||
break
|
||
except (APITimeoutError, APIConnectionError, RateLimitError) as e:
|
||
logging.warning(f"API error during streaming with callback: {type(e).__name__} - {e}. Retrying if possible.")
|
||
last_exception = e
|
||
break
|
||
except Exception as e:
|
||
logging.error(f"Unexpected error during streaming with callback: {traceback.format_exc()}")
|
||
# 如果已经获取了部分响应，可以返回，否则重试
|
||
if full_response:
|
||
error_msg = f"\n\n[注意: 由于错误中断，内容可能不完整: {str(e)}]"
|
||
logging.info(f"Returning partial response of length {len(full_response)} due to error: {type(e).__name__}")
|
||
callback(error_msg, full_response + error_msg if accumulate else None)
|
||
return full_response + error_msg
|
||
last_exception = e
|
||
break
|
||
|
||
# If we broke from the inner loop due to an error that needs retry
|
||
retries += 1
|
||
if retries < self.max_retries:
|
||
logging.info(f"Retrying stream with callback in {backoff_time} seconds...")
|
||
time.sleep(backoff_time + random.uniform(0, 1))
|
||
backoff_time = min(backoff_time * 2, MAX_BACKOFF)
|
||
else:
|
||
logging.error(f"Stream generation with callback failed after {self.max_retries} retries.")
|
||
# 如果已经获取了部分响应，可以返回，否则引发异常
|
||
if full_response:
|
||
error_msg = "\n\n[注意: 达到最大重试次数，内容可能不完整]"
|
||
logging.info(f"Returning partial response of length {len(full_response)} after max retries.")
|
||
callback(error_msg, full_response + error_msg if accumulate else None)
|
||
return full_response + error_msg
|
||
raise last_exception or Exception("Stream generation with callback failed after max retries.")
|
||
|
||
except (Timeout, APITimeoutError, APIConnectionError, RateLimitError) as e:
|
||
retries += 1
|
||
last_exception = e
|
||
logging.warning(f"Attempt {retries}/{self.max_retries} failed: {type(e).__name__} - {e}")
|
||
if retries < self.max_retries:
|
||
logging.info(f"Retrying in {backoff_time} seconds...")
|
||
time.sleep(backoff_time + random.uniform(0, 1))
|
||
backoff_time = min(backoff_time * 2, MAX_BACKOFF)
|
||
else:
|
||
logging.error(f"API call with callback failed after {self.max_retries} retries.")
|
||
# 如果已经获取了部分响应，可以返回，否则返回错误消息
|
||
if full_response:
|
||
error_msg = f"\n\n[注意: API调用失败，内容可能不完整: {str(e)}]"
|
||
callback(error_msg, full_response + error_msg if accumulate else None)
|
||
return full_response + error_msg
|
||
error_text = f"[生成内容失败: {str(last_exception)}]"
|
||
callback(error_text, error_text if accumulate else None)
|
||
return error_text
|
||
except Exception as e:
|
||
logging.error(f"Unexpected error setting up stream with callback: {traceback.format_exc()}")
|
||
# 如果已经获取了部分响应，可以返回，否则返回错误消息
|
||
if full_response:
|
||
error_msg = f"\n\n[注意: 意外错误，内容可能不完整: {str(e)}]"
|
||
callback(error_msg, full_response + error_msg if accumulate else None)
|
||
return full_response + error_msg
|
||
error_text = f"[生成内容失败: {str(e)}]"
|
||
callback(error_text, error_text if accumulate else None)
|
||
return error_text
|
||
|
||
# Should not be reached if logic is correct, but as a safeguard:
|
||
logging.error("Exited stream with callback generation loop unexpectedly.")
|
||
# 如果已经获取了部分响应，可以返回，否则返回错误消息
|
||
if full_response:
|
||
error_msg = "\n\n[注意: 意外退出流处理，内容可能不完整]"
|
||
callback(error_msg, full_response + error_msg if accumulate else None)
|
||
return full_response + error_msg
|
||
error_text = "[生成内容失败: 超过最大重试次数]"
|
||
callback(error_text, error_text if accumulate else None)
|
||
return error_text
|
||
|
||
async def async_generate_text_stream(self, system_prompt, user_prompt, temperature=0.7, top_p=0.9, presence_penalty=0.0):
|
||
"""
|
||
异步生成文本流，返回一个异步迭代器
|
||
|
||
Args:
|
||
system_prompt: 系统提示词
|
||
user_prompt: 用户提示词
|
||
temperature: 温度参数
|
||
top_p: 核采样参数
|
||
presence_penalty: 存在惩罚参数
|
||
|
||
Returns:
|
||
AsyncGenerator: 异步生成器，生成文本块
|
||
|
||
Raises:
|
||
Exception: 如果API调用在所有重试后仍然失败
|
||
"""
|
||
messages = [
|
||
{"role": "system", "content": system_prompt},
|
||
{"role": "user", "content": user_prompt},
|
||
]
|
||
logging.info(f"Async generating text stream using model: {self.model_name}")
|
||
|
||
retries = 0
|
||
backoff_time = INITIAL_BACKOFF
|
||
last_exception = None
|
||
full_response = ""
|
||
global_timeout = self.timeout * 2 # 设置全局超时，是API超时的两倍
|
||
|
||
while retries < self.max_retries:
|
||
try:
|
||
logging.debug(f"Attempt {retries + 1}/{self.max_retries} to async generate text stream.")
|
||
aclient = OpenAI(
|
||
api_key=self.api,
|
||
base_url=self.base_url,
|
||
timeout=self.timeout,
|
||
max_retries=0 # 禁用内置重试，使用我们自己的重试逻辑
|
||
)
|
||
|
||
stream = aclient.chat.completions.create(
|
||
model=self.model_name,
|
||
messages=messages,
|
||
temperature=temperature,
|
||
top_p=top_p,
|
||
presence_penalty=presence_penalty,
|
||
stream=True,
|
||
)
|
||
|
||
stream_start_time = time.time()
|
||
last_chunk_time = time.time()
|
||
received_any_chunk = False # 标记是否收到过任何块
|
||
|
||
try:
|
||
async for chunk in stream:
|
||
current_time = time.time()
|
||
|
||
# 检查流块超时：只有在已经开始接收数据后才检查
|
||
if received_any_chunk and current_time - last_chunk_time > self.stream_chunk_timeout:
|
||
logging.warning(f"Async stream chunk timeout: No new chunk received for {self.stream_chunk_timeout} seconds after previous chunk.")
|
||
if full_response:
|
||
timeout_msg = "\n\n[注意: 流式传输中断，内容可能不完整]"
|
||
yield timeout_msg
|
||
return # 结束生成器
|
||
raise Timeout(f"Async stream stalled: No new chunk received for {self.stream_chunk_timeout} seconds after previous chunk.")
|
||
|
||
# 检查首次响应超时：如果很长时间没收到第一个块
|
||
if not received_any_chunk and current_time - stream_start_time > self.timeout:
|
||
logging.warning(f"Async initial response timeout: No chunk received in {self.timeout} seconds since stream started.")
|
||
raise Timeout(f"No initial response received for {self.timeout} seconds.")
|
||
|
||
# 检查全局超时(保留作为安全措施)
|
||
if current_time - stream_start_time > global_timeout:
|
||
logging.warning(f"Async global timeout reached after {global_timeout} seconds.")
|
||
if full_response:
|
||
timeout_msg = "\n\n[注意: 由于全局超时，内容可能不完整]"
|
||
yield timeout_msg
|
||
return # 结束生成器
|
||
else:
|
||
logging.error("Async global timeout with no content received.")
|
||
raise Timeout(f"Async global timeout after {global_timeout} seconds with no content.")
|
||
|
||
# 处理接收到的数据块
|
||
if chunk.choices and chunk.choices[0].delta and chunk.choices[0].delta.content:
|
||
content = chunk.choices[0].delta.content
|
||
full_response += content
|
||
yield content
|
||
received_any_chunk = True # 标记已收到块
|
||
last_chunk_time = current_time # 更新最后块时间戳
|
||
|
||
logging.info("Async stream completed normally.")
|
||
return
|
||
|
||
except Timeout as e:
|
||
logging.warning(f"Async stream timeout: {e}. Retrying if possible ({retries + 1}/{self.max_retries}).")
|
||
last_exception = e
|
||
except (APITimeoutError, APIConnectionError, RateLimitError) as e:
|
||
logging.warning(f"Async API error during streaming: {type(e).__name__} - {e}. Retrying if possible.")
|
||
last_exception = e
|
||
except Exception as e:
|
||
logging.error(f"Async unexpected error during streaming: {traceback.format_exc()}")
|
||
# 如果已经获取了部分响应，可以返回错误消息后结束
|
||
if full_response:
|
||
error_msg = f"\n\n[注意: 由于错误中断，内容可能不完整: {str(e)}]"
|
||
logging.info(f"Yielding error message after partial response of length {len(full_response)}")
|
||
yield error_msg
|
||
return # 结束生成器
|
||
last_exception = e
|
||
|
||
# 执行重试逻辑
|
||
retries += 1
|
||
if retries < self.max_retries:
|
||
logging.info(f"Async retrying stream in {backoff_time} seconds...")
|
||
await asyncio.sleep(backoff_time + random.uniform(0, 1)) # 添加随机抖动
|
||
backoff_time = min(backoff_time * 2, MAX_BACKOFF)
|
||
else:
|
||
logging.error(f"Async stream generation failed after {self.max_retries} retries.")
|
||
# 如果已经获取了部分响应，返回错误消息后结束
|
||
if full_response:
|
||
error_msg = "\n\n[注意: 达到最大重试次数，内容可能不完整]"
|
||
logging.info(f"Yielding error message after partial response of length {len(full_response)}")
|
||
yield error_msg
|
||
return # 结束生成器
|
||
# 如果没有获取到任何内容，抛出异常
|
||
raise last_exception or Exception("Async stream generation failed after max retries with no content.")
|
||
|
||
except (Timeout, APITimeoutError, APIConnectionError, RateLimitError) as e:
|
||
retries += 1
|
||
last_exception = e
|
||
logging.warning(f"Async attempt {retries}/{self.max_retries} failed: {type(e).__name__} - {e}")
|
||
if retries < self.max_retries:
|
||
logging.info(f"Async retrying in {backoff_time} seconds...")
|
||
await asyncio.sleep(backoff_time + random.uniform(0, 1))
|
||
backoff_time = min(backoff_time * 2, MAX_BACKOFF)
|
||
else:
|
||
logging.error(f"Async API call failed after {self.max_retries} retries.")
|
||
# 如果已经获取了部分响应，返回错误消息后结束
|
||
if full_response:
|
||
error_msg = f"\n\n[注意: API调用失败，内容可能不完整: {str(e)}]"
|
||
yield error_msg
|
||
return # 结束生成器
|
||
error_text = f"[生成内容失败: {str(last_exception)}]"
|
||
yield error_text
|
||
return # 结束生成器
|
||
except Exception as e:
|
||
logging.error(f"Async unexpected error setting up stream: {traceback.format_exc()}")
|
||
# 如果已经获取了部分响应，返回错误消息后结束
|
||
if full_response:
|
||
error_msg = f"\n\n[注意: 意外错误，内容可能不完整: {str(e)}]"
|
||
yield error_msg
|
||
return # 结束生成器
|
||
error_text = f"[生成内容失败: {str(e)}]"
|
||
yield error_text
|
||
return # 结束生成器
|
||
|
||
# 作为安全措施，虽然通常不应该到达此处
|
||
logging.error("Async exited stream generation loop unexpectedly.")
|
||
# 如果已经获取了部分响应，返回错误消息后结束
|
||
if full_response:
|
||
error_msg = "\n\n[注意: 意外退出流处理，内容可能不完整]"
|
||
yield error_msg
|
||
return # 结束生成器
|
||
error_text = "[生成内容失败: 超过最大重试次数]"
|
||
yield error_text
|
||
|
||
async def async_work_stream(self, system_prompt, user_prompt, file_folder, temperature, top_p, presence_penalty):
|
||
"""异步完整工作流程：读取文件夹（如果提供），然后生成文本流"""
|
||
logging.info(f"Starting async 'work_stream' process. File folder: {file_folder}")
|
||
if file_folder:
|
||
logging.info(f"Reading context from folder: {file_folder}")
|
||
context = self.read_folder(file_folder)
|
||
if context:
|
||
user_prompt = f"{user_prompt.strip()}\n\n--- 参考资料 ---\n{context.strip()}"
|
||
else:
|
||
logging.warning(f"Folder {file_folder} provided but no content read.")
|
||
|
||
logging.info("Calling async_generate_text_stream...")
|
||
return self.async_generate_text_stream(system_prompt, user_prompt, temperature, top_p, presence_penalty)
|
||
|
||
def work_stream(self, system_prompt, user_prompt, file_folder, temperature, top_p, presence_penalty):
|
||
"""完整的工作流程（流式）：读取文件夹（如果提供），然后生成文本流"""
|
||
logging.info(f"Starting 'work_stream' process. File folder: {file_folder}")
|
||
if file_folder:
|
||
logging.info(f"Reading context from folder: {file_folder}")
|
||
context = self.read_folder(file_folder)
|
||
if context:
|
||
user_prompt = f"{user_prompt.strip()}\n\n--- 参考资料 ---\n{context.strip()}"
|
||
else:
|
||
logging.warning(f"Folder {file_folder} provided but no content read.")
|
||
|
||
full_response = self.generate_text_stream(system_prompt, user_prompt, temperature, top_p, presence_penalty)
|
||
return full_response
|
||
# --- End Streaming Methods --- |