Files
CodexandClaude Sonnet 4.5 d19d5bfc56 refactor(cleanup): 移除 logging 模块死代码
- unified_logger.py: 删除 5 个从未使用的装饰器/函数
- internal_loggers.py: 删除 3 个未使用方法和工厂函数
- action_logger.py: 删除 3 个未使用方法
- log_formatter.py: 删除 6 个未使用格式化函数(保留 GUI 依赖的 4 个)
- 同步更新 __init__.py 和测试文件

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-05-23 21:04:35 +08:00

367 lines
13 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""
统一日志系统模块
================
功能特性:
1. 统一的日志格式规范(时间戳、级别、线程、类名、行号、traceId)
2. MDC(Mapped Diagnostic Context)支持,用于追踪请求链路
3. 敏感信息自动脱敏(密码、身份证、手机号等)
4. 性能优化(DEBUG级别判断、避免不必要的字符串拼接)
5. 异常堆栈完整记录
6. 方法入口/出口自动记录
"""
from __future__ import annotations
import os
import sys
import re
import time
import uuid
import logging
import threading
import inspect
import traceback
from typing import Optional, Dict, Any
from functools import wraps
from contextlib import contextmanager
from douyin.logging import douyin_logging
# 线程本地存储,用于MDC(Mapped Diagnostic Context)
_thread_local = threading.local()
class MDC:
"""Mapped Diagnostic Context - 用于在日志中传递上下文信息(如traceId)"""
@staticmethod # type: ignore[misc]
def put(key: str, value: Any) -> None:
"""设置MDC值"""
if not hasattr(_thread_local, "mdc"):
_thread_local.mdc = {}
_thread_local.mdc[key] = value
@staticmethod # type: ignore[misc]
def get(key: str, default: Any = None) -> Any:
"""获取MDC值"""
if hasattr(_thread_local, "mdc"):
return _thread_local.mdc.get(key, default)
return default
@staticmethod # type: ignore[misc]
def remove(key: str) -> None:
"""移除MDC值"""
if hasattr(_thread_local, "mdc"):
_thread_local.mdc.pop(key, None)
@staticmethod
def clear() -> None:
"""清空所有MDC值"""
if hasattr(_thread_local, "mdc"):
_thread_local.mdc.clear()
@staticmethod
def get_all() -> Dict[str, Any]:
"""获取所有MDC值"""
if hasattr(_thread_local, "mdc"):
return dict(_thread_local.mdc)
return {}
@staticmethod
def get_trace_id() -> Optional[str]:
"""获取当前线程的traceId"""
result = MDC.get("traceId")
return result if result else None
@staticmethod
def set_trace_id(trace_id: Optional[str] = None) -> str:
"""设置当前线程的traceId,如果未提供则自动生成"""
if trace_id is None:
trace_id = str(uuid.uuid4())[:16] # 使用UUID前16位作为traceId
MDC.put("traceId", trace_id)
return str(trace_id) # 确保返回 str 类型
class SensitiveDataMasker:
"""敏感信息脱敏工具"""
# 敏感信息模式
PATTERNS = {
"password": [
r"(?i)(password|pwd|passwd)\s*[:=]\s*([^\s,;}\]]+)",
r'(?i)(password|pwd|passwd)["\']?\s*[:=]\s*["\']?([^"\'\s,;}\]]+)',
],
"phone": [
r"(\d{3})\d{4}(\d{4})", # 手机号:保留前3位和后4位
r"(\+86\s*)?(\d{3})\d{4}(\d{4})", # 带区号的手机号
],
"id_card": [
r"(\d{6})\d{8}(\d{4})", # 身份证号:保留前6位和后4位
r"(\d{3})\d{11}(\d{3})", # 18位身份证号
],
"email": [
r"([a-zA-Z0-9._%+-]+)@([a-zA-Z0-9.-]+\.[a-zA-Z]{2,})", # 邮箱:保留@前1位和域名
],
"bank_card": [
r"(\d{4})\d{8,12}(\d{4})", # 银行卡号:保留前4位和后4位
],
}
@staticmethod # type: ignore[misc]
def mask_password(text: str) -> str:
"""脱敏密码"""
for pattern in SensitiveDataMasker.PATTERNS["password"]:
text = re.sub(pattern, r"\1=***", text)
return text
@staticmethod # type: ignore[misc]
def mask_phone(text: str) -> str:
"""脱敏手机号"""
# 手机号脱敏:保留前3位和后4位
text = re.sub(r"(\d{3})\d{4}(\d{4})", r"\1****\2", text)
return text
@staticmethod # type: ignore[misc]
def mask_id_card(text: str) -> str:
"""脱敏身份证号"""
# 18位身份证:保留前6位和后4位
text = re.sub(r"(\d{6})\d{8}(\d{4})", r"\1********\2", text)
# 15位身份证:保留前6位和后3位
text = re.sub(r"(\d{6})\d{6}(\d{3})", r"\1******\2", text)
return text
@staticmethod # type: ignore[misc]
def mask_email(text: str) -> str:
"""脱敏邮箱"""
def replace_email(match):
username = match.group(1)
domain = match.group(2)
if len(username) > 1:
masked_username = username[0] + "*" * (len(username) - 1)
else:
masked_username = "*"
return f"{masked_username}@{domain}"
text = re.sub(SensitiveDataMasker.PATTERNS["email"][0], replace_email, text)
return text
@staticmethod # type: ignore[misc]
def mask_bank_card(text: str) -> str:
"""脱敏银行卡号"""
text = re.sub(r"(\d{4})\d{8,12}(\d{4})", r"\1********\2", text)
return text
@staticmethod # type: ignore[misc]
def mask_all(text: str) -> str:
"""对所有敏感信息进行脱敏(性能优化版本)"""
if not isinstance(text, str):
text = str(text)
# 优化:快速检查,如果文本很短或明显不包含敏感信息,跳过部分脱敏
text_lower = text.lower()
has_sensitive_keywords = any(
keyword in text_lower
for keyword in ["password", "pwd", "phone", "mobile", "id", "card", "@"]
)
# 只在可能包含敏感信息时才执行脱敏
if has_sensitive_keywords:
text = SensitiveDataMasker.mask_password(text)
text = SensitiveDataMasker.mask_phone(text)
text = SensitiveDataMasker.mask_id_card(text)
text = SensitiveDataMasker.mask_email(text)
text = SensitiveDataMasker.mask_bank_card(text)
return text
@staticmethod # type: ignore[misc]
def mask_dict(data: Dict[str, Any]) -> Dict[str, Any]:
"""对字典中的敏感字段进行脱敏"""
sensitive_keys = [
"password",
"pwd",
"passwd",
"phone",
"mobile",
"tel",
"id_card",
"idcard",
"email",
"bank_card",
"bankcard",
"card_no",
"card_number",
"secret",
"token",
"api_key",
]
masked_data = {}
for key, value in data.items():
if any(sk in key.lower() for sk in sensitive_keys):
if isinstance(value, str):
masked_data[key] = SensitiveDataMasker.mask_all(value)
else:
masked_data[key] = "***"
elif isinstance(value, dict):
masked_data[key] = SensitiveDataMasker.mask_dict(value)
elif isinstance(value, str):
masked_data[key] = SensitiveDataMasker.mask_all(value)
else:
masked_data[key] = value
return masked_data
class UnifiedFormatter(logging.Formatter):
"""统一的日志格式化器,包含traceId和MDC信息"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
# 简化的格式:时间戳 - [级别] - [模块] - 消息
# 去掉了冗余的线程名、函数名、行号、traceId等信息,使日志更简洁
self._detailed_format = (
"%(asctime)s - [%(levelname)s] - [%(name)s] - %(message)s"
)
# 缓存:避免重复检查MDC
self._last_mdc_check_time = 0.0
self._last_mdc_data = {}
self._mdc_cache_ttl = (
0.1 # 100ms缓存,减少MDC查询频率(在高频日志场景下提升性能)
)
self._mdc_check_counter = 0 # 计数器,用于进一步优化
def format(self, record: logging.LogRecord) -> str:
"""格式化日志记录,简化显示以提高可读性"""
# 只在DEBUG级别或错误时显示traceId
if record.levelno <= logging.DEBUG or record.levelno >= logging.ERROR:
trace_id = MDC.get_trace_id()
if trace_id:
record.traceId = trace_id # type: ignore[attr-defined]
else:
record.traceId = "N/A" # type: ignore[attr-defined]
else:
record.traceId = "" # type: ignore[attr-defined]
# 只在DEBUG级别显示完整的MDC信息
if record.levelno <= logging.DEBUG:
# 优化:使用计数器和时间戳双重检查,减少MDC查询频率
self._mdc_check_counter += 1
current_time = time.time()
should_check_mdc = (
current_time - self._last_mdc_check_time
) > self._mdc_cache_ttl or self._mdc_check_counter >= 50 # 每50条记录强制检查一次
if should_check_mdc:
self._last_mdc_data = MDC.get_all()
self._last_mdc_check_time = current_time
self._mdc_check_counter = 0
mdc_data = self._last_mdc_data
if mdc_data:
# 将MDC信息添加到消息中(排除traceId,避免重复)
mdc_items = [(k, v) for k, v in mdc_data.items() if k != "traceId"]
if mdc_items:
# 优化:使用列表推导式,避免多次字符串拼接
mdc_str = ", ".join(f"{k}={v}" for k, v in mdc_items)
if mdc_str:
record.msg = f"{record.msg} | MDC: {mdc_str}"
# 敏感信息脱敏(只对可能包含敏感信息的消息进行)
if isinstance(record.msg, str):
msg = record.msg
# 快速检查:如果消息很短(<50字符),直接跳过脱敏
if len(msg) > 50:
# 使用更高效的检查:先检查长度,再检查关键词
msg_lower = msg.lower()
if any(
keyword in msg_lower
for keyword in [
"password",
"pwd",
"phone",
"mobile",
"id",
"card",
"email",
"@",
]
):
record.msg = SensitiveDataMasker.mask_all(msg)
# 格式化
return super().format(record)
# 缓存logger实例,避免重复创建和检查
_logger_cache: Dict[str, logging.Logger] = {}
_logger_cache_lock = threading.Lock()
def get_unified_logger(name: Optional[str] = None) -> logging.Logger:
"""获取统一配置的logger实例(性能优化版本)"""
# 确保name不为None
if name is None:
# 优化:避免使用inspect.currentframe(),直接使用调用栈
# 如果调用者没有提供name,使用默认值
name = "root"
try:
# 尝试从调用栈获取模块名(最多向上查找3层)
frame = inspect.currentframe()
if frame:
for _ in range(3): # 最多向上查找3层
frame = frame.f_back
if frame is None:
break
module_name = frame.f_globals.get("__name__")
if module_name and module_name != __name__:
name = module_name
break
except Exception:
# 如果获取失败,使用默认值
pass
# 确保name是str类型(类型检查)
logger_name: str = name if isinstance(name, str) else "root"
# 使用缓存避免重复创建和检查
with _logger_cache_lock:
if logger_name in _logger_cache:
return _logger_cache[logger_name]
logger = douyin_logging.get_logger(logger_name)
# 如果logger还没有自定义formatter,添加统一formatter
# 优化:只检查一次,避免每次调用都检查
if logger.handlers:
for handler in logger.handlers:
if isinstance(handler.formatter, logging.Formatter) and not isinstance(
handler.formatter, UnifiedFormatter
):
handler.setFormatter(
UnifiedFormatter(
fmt=(
handler.formatter._fmt
if hasattr(handler.formatter, "_fmt")
else None
),
datefmt="%Y-%m-%d %H:%M:%S",
)
)
_logger_cache[logger_name] = logger
return logger
# 导出主要接口
__all__ = [
"MDC",
"SensitiveDataMasker",
"UnifiedFormatter",
"get_unified_logger",
]