mirror of
https://github.com/Nezumi-2711/astrbot_plugin_qq_group_daily_analysis.git
synced 2026-09-22 20:01:04 +00:00
* fix(max_query_rounds): 弃用的 max_query_rounds
* feat(provider): 根据配置键获取 Provider，支持多级回退
回退顺序：
1. 尝试从配置获取指定的 provider_id（如 topic_provider_id）
2. 回退到主 LLM provider_id（llm_provider_id）
3. 回退到当前会话的 Provider（通过 umo）
4. 回退到第一个可用的 Provider
* feat: 添加 _special: select_provider 支持并删除 custom_api_key 相关逻辑
* style: 使用 ruff 格式化代码
* refactor: 重构 provider 选择逻辑并修复兼容性问题
* improve: 改进 Provider 选择和 LLM 调用的日志输出
* [v3.7.0] 根据配置键获取 Provider，支持多级回退
This commit is contained in:
SXP-Simon
@@ -32,6 +32,16 @@ class BaseAnalyzer(ABC):
|
||||
self.context = context
|
||||
self.config_manager = config_manager
|
||||
|
||||
def get_provider_id_key(self) -> str:
|
||||
"""
|
||||
获取 Provider ID 配置键名
|
||||
子类可重写以指定特定的 provider，默认返回 None（使用主 LLM Provider）
|
||||
|
||||
Returns:
|
||||
Provider ID 配置键名，如 'topic_provider_id'
|
||||
"""
|
||||
return None
|
||||
|
||||
@abstractmethod
|
||||
def get_data_type(self) -> str:
|
||||
"""
|
||||
@@ -128,12 +138,19 @@ class BaseAnalyzer(ABC):
|
||||
)
|
||||
return [], TokenUsage()
|
||||
|
||||
# 2. 调用LLM
|
||||
# 2. 调用LLM（使用配置的 provider）
|
||||
max_tokens = self.get_max_tokens()
|
||||
temperature = self.get_temperature()
|
||||
provider_id_key = self.get_provider_id_key()
|
||||
|
||||
response = await call_provider_with_retry(
|
||||
self.context, self.config_manager, prompt, max_tokens, temperature, umo
|
||||
self.context,
|
||||
self.config_manager,
|
||||
prompt,
|
||||
max_tokens,
|
||||
temperature,
|
||||
umo,
|
||||
provider_id_key,
|
||||
)
|
||||
|
||||
if response is None:
|
||||
|
||||
@@ -18,6 +18,10 @@ class GoldenQuoteAnalyzer(BaseAnalyzer):
|
||||
专门处理群聊金句的提取和分析
|
||||
"""
|
||||
|
||||
def get_provider_id_key(self) -> str:
|
||||
"""获取 Provider ID 配置键名"""
|
||||
return "golden_quote_provider_id"
|
||||
|
||||
def get_data_type(self) -> str:
|
||||
"""获取数据类型标识"""
|
||||
return "金句"
|
||||
|
||||
@@ -19,6 +19,10 @@ class TopicAnalyzer(BaseAnalyzer):
|
||||
专门处理群聊话题的提取和分析
|
||||
"""
|
||||
|
||||
def get_provider_id_key(self) -> str:
|
||||
"""获取 Provider ID 配置键名"""
|
||||
return "topic_provider_id"
|
||||
|
||||
def get_data_type(self) -> str:
|
||||
"""获取数据类型标识"""
|
||||
return "话题"
|
||||
|
||||
@@ -16,6 +16,10 @@ class UserTitleAnalyzer(BaseAnalyzer):
|
||||
专门处理用户称号分配和MBTI类型分析
|
||||
"""
|
||||
|
||||
def get_provider_id_key(self) -> str:
|
||||
"""获取 Provider ID 配置键名"""
|
||||
return "user_title_provider_id"
|
||||
|
||||
def get_data_type(self) -> str:
|
||||
"""获取数据类型标识"""
|
||||
return "用户称号"
|
||||
|
||||
@@ -193,23 +193,31 @@ class LLMAnalyzer:
|
||||
max_tokens: int,
|
||||
temperature: float,
|
||||
umo: str = None,
|
||||
provider_id_key: str = None,
|
||||
):
|
||||
"""
|
||||
向后兼容的LLM调用方法
|
||||
现在委托给llm_utils模块处理
|
||||
|
||||
Args:
|
||||
provider: LLM服务商实例或None
|
||||
provider: LLM服务商实例或None（已弃用，现在使用 provider_id_key）
|
||||
prompt: 输入的提示语
|
||||
max_tokens: 最大生成token数
|
||||
temperature: 采样温度
|
||||
umo: 指定使用的模型唯一标识符
|
||||
provider_id_key: 配置中的 provider_id 键名（可选）
|
||||
|
||||
Returns:
|
||||
LLM生成的结果
|
||||
"""
|
||||
return await call_provider_with_retry(
|
||||
self.context, self.config_manager, prompt, max_tokens, temperature, umo
|
||||
self.context,
|
||||
self.config_manager,
|
||||
prompt,
|
||||
max_tokens,
|
||||
temperature,
|
||||
umo,
|
||||
provider_id_key,
|
||||
)
|
||||
|
||||
def _fix_json(self, text: str) -> str:
|
||||
|
||||
+204
-119
@@ -9,6 +9,171 @@ from astrbot.api import logger
|
||||
import aiohttp
|
||||
|
||||
|
||||
def _try_get_provider_by_id(
|
||||
context, provider_id: str, description: str
|
||||
) -> Optional[Any]:
|
||||
"""
|
||||
尝试通过 ID 获取 Provider 的辅助函数
|
||||
|
||||
Args:
|
||||
context: AstrBot上下文对象
|
||||
provider_id: Provider ID
|
||||
description: 描述信息，用于日志
|
||||
|
||||
Returns:
|
||||
Provider 实例或 None
|
||||
"""
|
||||
if not provider_id or not isinstance(provider_id, str) or not provider_id.strip():
|
||||
return None
|
||||
|
||||
provider_id = provider_id.strip()
|
||||
logger.info(f"尝试使用{description}: {provider_id}")
|
||||
try:
|
||||
provider = context.get_provider_by_id(provider_id=provider_id)
|
||||
if provider:
|
||||
logger.info(f"✓ 使用{description}: {provider_id}")
|
||||
return provider
|
||||
except Exception as e:
|
||||
logger.warning(f"无法找到{description} '{provider_id}': {e}")
|
||||
return None
|
||||
|
||||
|
||||
def _try_get_session_provider(context, umo: str) -> Optional[Any]:
|
||||
"""
|
||||
尝试获取会话 Provider 的辅助函数
|
||||
|
||||
Args:
|
||||
context: AstrBot上下文对象
|
||||
umo: unified_msg_origin
|
||||
|
||||
Returns:
|
||||
Provider 实例或 None
|
||||
"""
|
||||
try:
|
||||
provider = context.get_using_provider(umo=umo)
|
||||
if provider:
|
||||
try:
|
||||
meta = provider.meta()
|
||||
provider_id = meta.id
|
||||
logger.info(f"✓ 使用当前会话的 Provider: {provider_id}")
|
||||
except Exception:
|
||||
logger.info("✓ 使用当前会话的默认 Provider")
|
||||
return provider
|
||||
except Exception as e:
|
||||
logger.warning(f"无法获取会话 Provider: {e}")
|
||||
return None
|
||||
|
||||
|
||||
def _try_get_first_available_provider(context) -> Optional[Any]:
|
||||
"""
|
||||
尝试获取第一个可用 Provider 的辅助函数
|
||||
|
||||
Args:
|
||||
context: AstrBot上下文对象
|
||||
|
||||
Returns:
|
||||
Provider 实例或 None
|
||||
"""
|
||||
try:
|
||||
all_providers = context.get_all_providers()
|
||||
if all_providers and len(all_providers) > 0:
|
||||
provider = all_providers[0]
|
||||
logger.info(f"✓ 使用第一个可用 Provider: {type(provider).__name__}")
|
||||
return provider
|
||||
except Exception as e:
|
||||
logger.warning(f"无法获取任何 Provider: {e}")
|
||||
return None
|
||||
|
||||
|
||||
def get_provider_with_fallback(
|
||||
context, config_manager, provider_id_key: str, umo: str = None
|
||||
) -> Optional[Any]:
|
||||
"""
|
||||
根据配置键获取 Provider，支持多级回退
|
||||
|
||||
回退顺序：
|
||||
1. 尝试从配置获取指定的 provider_id（如 topic_provider_id）
|
||||
2. 回退到主 LLM provider_id（llm_provider_id）
|
||||
3. 回退到当前会话的 Provider（通过 umo）
|
||||
4. 回退到第一个可用的 Provider
|
||||
|
||||
Args:
|
||||
context: AstrBot上下文对象
|
||||
config_manager: 配置管理器
|
||||
provider_id_key: 配置中的 provider_id 键名（如 'topic_provider_id'）
|
||||
umo: unified_msg_origin，用于获取会话默认 Provider
|
||||
|
||||
Returns:
|
||||
Provider 实例或 None
|
||||
"""
|
||||
try:
|
||||
# 输出Provider选择开始日志
|
||||
task_desc = provider_id_key if provider_id_key else "默认任务"
|
||||
logger.info(f"[Provider 选择] 开始为 {task_desc} 选择 Provider...")
|
||||
|
||||
# 定义回退策略列表
|
||||
strategies = []
|
||||
strategy_names = []
|
||||
|
||||
# 1. 特定任务的 provider_id
|
||||
if provider_id_key:
|
||||
getter_method = f"get_{provider_id_key}"
|
||||
if hasattr(config_manager, getter_method):
|
||||
specific_provider_id = getattr(config_manager, getter_method)()
|
||||
if specific_provider_id:
|
||||
strategies.append(
|
||||
lambda pid=specific_provider_id: _try_get_provider_by_id(
|
||||
context, pid, f"配置的 {provider_id_key}"
|
||||
)
|
||||
)
|
||||
strategy_names.append(f"1. 配置的 {provider_id_key}")
|
||||
|
||||
# 2. 主 LLM provider_id
|
||||
main_provider_id = config_manager.get_llm_provider_id()
|
||||
if main_provider_id:
|
||||
strategies.append(
|
||||
lambda pid=main_provider_id: _try_get_provider_by_id(
|
||||
context, pid, "主 LLM Provider"
|
||||
)
|
||||
)
|
||||
strategy_names.append("2. 主 LLM Provider")
|
||||
|
||||
# 3. 当前会话的 Provider
|
||||
strategies.append(lambda: _try_get_session_provider(context, umo))
|
||||
strategy_names.append("3. 当前会话 Provider")
|
||||
|
||||
# 4. 第一个可用的 Provider
|
||||
strategies.append(lambda: _try_get_first_available_provider(context))
|
||||
strategy_names.append("4. 第一个可用 Provider")
|
||||
|
||||
# 输出回退策略列表
|
||||
logger.info(f"[Provider 选择] 回退策略顺序：{' -> '.join(strategy_names)}")
|
||||
|
||||
# 依次尝试每个策略
|
||||
for idx, strategy in enumerate(strategies):
|
||||
provider = strategy()
|
||||
if provider:
|
||||
# 获取最终的 provider ID 用于日志
|
||||
final_provider_id = "unknown"
|
||||
try:
|
||||
meta = provider.meta()
|
||||
final_provider_id = meta.id
|
||||
except Exception:
|
||||
final_provider_id = type(provider).__name__
|
||||
|
||||
logger.info(
|
||||
f"[Provider 选择] ✓ 成功！使用策略 #{idx + 1}，Provider ID: {final_provider_id}"
|
||||
)
|
||||
return provider
|
||||
|
||||
logger.error("[Provider 选择] ✗ 失败：所有回退策略均无法获取可用 Provider")
|
||||
return None
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"[Provider 选择] ✗ 异常：Provider 选择过程出错: {e}")
|
||||
return None
|
||||
|
||||
|
||||
async def call_provider_with_retry(
|
||||
context,
|
||||
config_manager,
|
||||
@@ -16,9 +181,10 @@ async def call_provider_with_retry(
|
||||
max_tokens: int,
|
||||
temperature: float,
|
||||
umo: str = None,
|
||||
provider_id_key: str = None,
|
||||
) -> Optional[Any]:
|
||||
"""
|
||||
调用LLM提供者，带超时、重试与退避。支持自定义服务商。
|
||||
调用LLM提供者，带超时、重试与退避。支持自定义服务商和配置化 Provider 选择。
|
||||
|
||||
Args:
|
||||
context: AstrBot上下文对象
|
||||
@@ -27,6 +193,7 @@ async def call_provider_with_retry(
|
||||
max_tokens: 最大生成token数
|
||||
temperature: 采样温度
|
||||
umo: 指定使用的模型唯一标识符
|
||||
provider_id_key: 配置中的 provider_id 键名（如 'topic_provider_id'），用于选择特定的 Provider
|
||||
|
||||
Returns:
|
||||
LLM生成的结果，失败时返回None
|
||||
@@ -35,129 +202,47 @@ async def call_provider_with_retry(
|
||||
retries = config_manager.get_llm_retries()
|
||||
backoff = config_manager.get_llm_backoff()
|
||||
|
||||
# 获取自定义服务商参数
|
||||
custom_api_key = config_manager.get_custom_api_key()
|
||||
custom_api_base = config_manager.get_custom_api_base_url()
|
||||
custom_model = config_manager.get_custom_model_name()
|
||||
|
||||
last_exc = None
|
||||
for attempt in range(1, retries + 1):
|
||||
try:
|
||||
if custom_api_key and custom_api_base and custom_model:
|
||||
logger.info(
|
||||
f"使用自定义LLM提供商: {custom_api_base} model={custom_model}, max_tokens={max_tokens}, temperature={temperature}"
|
||||
)
|
||||
logger.debug(
|
||||
f"自定义LLM提供商 prompt 长度: {len(prompt) if prompt else 0}"
|
||||
)
|
||||
logger.debug(
|
||||
f"自定义LLM提供商 prompt 前100字符: {prompt[:100] if prompt else 'None'}..."
|
||||
# 使用新的 provider 选择逻辑，支持配置化选择和多级回退
|
||||
provider = get_provider_with_fallback(
|
||||
context, config_manager, provider_id_key, umo
|
||||
)
|
||||
|
||||
provider_id = "unknown"
|
||||
if provider:
|
||||
try:
|
||||
meta = provider.meta()
|
||||
provider_id = meta.id
|
||||
except Exception as e:
|
||||
logger.debug(f"获取提供商ID失败: {e}")
|
||||
|
||||
if not provider:
|
||||
logger.error("provider 为空，无法调用 text_chat，直接返回 None")
|
||||
return None
|
||||
|
||||
logger.info(
|
||||
f"[LLM 调用] 使用 Provider: {provider_id} | "
|
||||
f"max_tokens={max_tokens} | temperature={temperature} | "
|
||||
f"prompt长度={len(prompt) if prompt else 0}字符"
|
||||
)
|
||||
|
||||
logger.debug(
|
||||
f"[LLM 调用] Prompt 前100字符: {prompt[:100] if prompt else 'None'}..."
|
||||
)
|
||||
|
||||
# 检查 prompt 是否为空
|
||||
if not prompt or not prompt.strip():
|
||||
logger.error(
|
||||
"LLM provider: prompt 为空或只包含空白字符，无法调用 text_chat"
|
||||
)
|
||||
return None
|
||||
|
||||
# 检查 prompt 是否为空
|
||||
if not prompt or not prompt.strip():
|
||||
logger.error(
|
||||
"自定义LLM提供商: prompt 为空或只包含空白字符，无法发送请求"
|
||||
)
|
||||
return None
|
||||
|
||||
async with aiohttp.ClientSession() as session:
|
||||
headers = {
|
||||
"Authorization": f"Bearer {custom_api_key}",
|
||||
"Content-Type": "application/json",
|
||||
}
|
||||
payload = {
|
||||
"model": custom_model,
|
||||
"messages": [{"role": "user", "content": prompt}],
|
||||
"max_tokens": max_tokens,
|
||||
"temperature": temperature,
|
||||
}
|
||||
aio_timeout = aiohttp.ClientTimeout(total=timeout)
|
||||
async with session.post(
|
||||
custom_api_base,
|
||||
json=payload,
|
||||
headers=headers,
|
||||
timeout=aio_timeout,
|
||||
) as resp:
|
||||
if resp.status != 200:
|
||||
error_text = await resp.text()
|
||||
logger.error(
|
||||
f"自定义LLM服务商请求失败: HTTP {resp.status}, 内容: {error_text}"
|
||||
)
|
||||
try:
|
||||
response_json = await resp.json()
|
||||
except Exception as json_err:
|
||||
error_text = await resp.text()
|
||||
logger.error(
|
||||
f"自定义LLM服务商响应JSON解析失败: {json_err}, 内容: {error_text}"
|
||||
)
|
||||
return None
|
||||
# 兼容 OpenAI 格式，安全访问嵌套字段
|
||||
content = None
|
||||
try:
|
||||
choices = response_json.get("choices")
|
||||
if (
|
||||
choices
|
||||
and isinstance(choices, list)
|
||||
and len(choices) > 0
|
||||
):
|
||||
message = choices[0].get("message")
|
||||
if message and isinstance(message, dict):
|
||||
content = message.get("content")
|
||||
if content is None:
|
||||
logger.error(f"自定义LLM响应格式异常: {response_json}")
|
||||
return None
|
||||
except Exception as key_err:
|
||||
logger.error(
|
||||
f"自定义LLM响应结构解析失败: {key_err}, 响应内容: {response_json}"
|
||||
)
|
||||
return None
|
||||
|
||||
# 构造一个兼容原有逻辑的对象
|
||||
class CustomResponse:
|
||||
completion_text = content
|
||||
raw_completion = response_json
|
||||
|
||||
return CustomResponse()
|
||||
else:
|
||||
# 确保使用当前指定的模型
|
||||
provider = context.get_using_provider(umo=umo)
|
||||
provider_id = "unknown"
|
||||
if provider:
|
||||
try:
|
||||
meta = provider.meta()
|
||||
provider_id = meta.id
|
||||
except Exception as e:
|
||||
logger.debug(f"获取提供商ID失败: {e}")
|
||||
logger.info(f"获取到的 provider ID: {provider_id}")
|
||||
if not provider or provider_id == "unknown":
|
||||
logger.warning(f"获取的提供商不正确 (Provider ID: {provider_id})")
|
||||
|
||||
logger.info(
|
||||
f"使用LLM provider: {provider}, max_tokens={max_tokens}, temperature={temperature}"
|
||||
)
|
||||
if not provider:
|
||||
logger.error("provider 为空，无法调用 text_chat，直接返回 None")
|
||||
return None
|
||||
|
||||
logger.debug(
|
||||
f"LLM provider prompt 长度: {len(prompt) if prompt else 0}"
|
||||
)
|
||||
logger.debug(
|
||||
f"LLM provider prompt 前100字符: {prompt[:100] if prompt else 'None'}..."
|
||||
)
|
||||
|
||||
# 检查 prompt 是否为空
|
||||
if not prompt or not prompt.strip():
|
||||
logger.error(
|
||||
"LLM provider: prompt 为空或只包含空白字符，无法调用 text_chat"
|
||||
)
|
||||
return None
|
||||
|
||||
coro = provider.text_chat(
|
||||
prompt=prompt, max_tokens=max_tokens, temperature=temperature
|
||||
)
|
||||
return await asyncio.wait_for(coro, timeout=timeout)
|
||||
coro = provider.text_chat(
|
||||
prompt=prompt, max_tokens=max_tokens, temperature=temperature
|
||||
)
|
||||
return await asyncio.wait_for(coro, timeout=timeout)
|
||||
except asyncio.TimeoutError as e:
|
||||
last_exc = e
|
||||
logger.warning(f"LLM请求超时: 第{attempt}次, timeout={timeout}s")
|
||||
|
||||
+13
-18
@@ -69,10 +69,6 @@ class ConfigManager:
|
||||
"""获取最大金句数量"""
|
||||
return self.config.get("max_golden_quotes", 5)
|
||||
|
||||
def get_max_query_rounds(self) -> int:
|
||||
"""获取最大查询轮数"""
|
||||
return self.config.get("max_query_rounds", 35)
|
||||
|
||||
def get_llm_timeout(self) -> int:
|
||||
"""获取LLM请求超时时间（秒）"""
|
||||
return self.config.get("llm_timeout", 30)
|
||||
@@ -97,17 +93,21 @@ class ConfigManager:
|
||||
"""获取用户称号分析最大token数"""
|
||||
return self.config.get("user_title_max_tokens", 4096)
|
||||
|
||||
def get_custom_api_key(self) -> str:
|
||||
"""获取自定义 LLM 服务的 API Key"""
|
||||
return self.config.get("custom_api_key", "")
|
||||
def get_llm_provider_id(self) -> str:
|
||||
"""获取主 LLM Provider ID"""
|
||||
return self.config.get("llm_provider_id", "")
|
||||
|
||||
def get_custom_api_base_url(self) -> str:
|
||||
"""获取自定义 LLM 服务的 Base URL"""
|
||||
return self.config.get("custom_api_base_url", "")
|
||||
def get_topic_provider_id(self) -> str:
|
||||
"""获取话题分析专用 Provider ID"""
|
||||
return self.config.get("topic_provider_id", "")
|
||||
|
||||
def get_custom_model_name(self) -> str:
|
||||
"""获取自定义 LLM 服务的模型名称"""
|
||||
return self.config.get("custom_model_name", "")
|
||||
def get_user_title_provider_id(self) -> str:
|
||||
"""获取用户称号分析专用 Provider ID"""
|
||||
return self.config.get("user_title_provider_id", "")
|
||||
|
||||
def get_golden_quote_provider_id(self) -> str:
|
||||
"""获取金句分析专用 Provider ID"""
|
||||
return self.config.get("golden_quote_provider_id", "")
|
||||
|
||||
def get_pdf_output_dir(self) -> str:
|
||||
"""获取PDF输出目录"""
|
||||
@@ -264,11 +264,6 @@ class ConfigManager:
|
||||
self.config["max_golden_quotes"] = count
|
||||
self.config.save_config()
|
||||
|
||||
def set_max_query_rounds(self, rounds: int):
|
||||
"""设置最大查询轮数"""
|
||||
self.config["max_query_rounds"] = rounds
|
||||
self.config.save_config()
|
||||
|
||||
def set_pdf_output_dir(self, directory: str):
|
||||
"""设置PDF输出目录"""
|
||||
self.config["pdf_output_dir"] = directory
|
||||
|
||||
Reference in New Issue
Block a user