在准备环境前提交次全部更改。

This commit is contained in:
Neo
2026-02-19 08:35:13 +08:00
parent ded6dfb9d8
commit 4eac07da47
1387 changed files with 6107191 additions and 33002 deletions

View File

@@ -0,0 +1,292 @@
# -*- coding: utf-8 -*-
"""统一任务日志器
提供统一的日志输出格式,支持:
- 任务开始/结束记录
- 进度追踪
- 统计计数
- 格式化的任务总结
"""
import logging
import time
from datetime import datetime
from typing import Any, Dict, Optional
# 统一日志格式
UNIFIED_LOG_FORMAT = "[%(asctime)s] %(levelname)-5s | %(name)s | %(message)s"
UNIFIED_DATE_FORMAT = "%Y-%m-%d %H:%M:%S"
class TaskLogger:
"""任务日志器,统一 print 和 logging 输出"""
def __init__(
self,
task_code: str,
logger: Optional[logging.Logger] = None,
):
"""
初始化任务日志器
Args:
task_code: 任务代码
logger: 底层日志器,如果不提供则创建新的
"""
self.task_code = task_code
self.logger = logger or logging.getLogger(f"task.{task_code}")
# 任务状态
self.start_time: Optional[datetime] = None
self.end_time: Optional[datetime] = None
self.status: str = "pending"
# 统计计数
self.counts: Dict[str, int] = {
"fetched": 0,
"inserted": 0,
"updated": 0,
"skipped": 0,
"errors": 0,
}
# 额外信息
self.extra_info: Dict[str, Any] = {}
# 校验结果(如果有)
self.verification_result: Optional[dict] = None
def start(self, message: str = "任务开始"):
"""
记录任务开始
Args:
message: 开始消息
"""
self.start_time = datetime.now()
self.status = "running"
self.logger.info(
"%s | %s | 开始时间: %s",
self.task_code, message,
self.start_time.strftime(UNIFIED_DATE_FORMAT)
)
def progress(self, message: str, **kwargs):
"""
记录进度
Args:
message: 进度消息
**kwargs: 额外的统计信息
"""
# 更新计数
for key, value in kwargs.items():
if key in self.counts:
if isinstance(value, int):
self.counts[key] += value
else:
self.counts[key] = value
else:
self.extra_info[key] = value
# 构建进度字符串
counts_str = ", ".join(f"{k}={v}" for k, v in self.counts.items() if v > 0)
if counts_str:
self.logger.info("%s | %s | %s", self.task_code, message, counts_str)
else:
self.logger.info("%s | %s", self.task_code, message)
def info(self, message: str, *args):
"""记录信息级别日志"""
if args:
self.logger.info(f"{self.task_code} | {message}", *args)
else:
self.logger.info(f"{self.task_code} | {message}")
def warning(self, message: str, *args):
"""记录警告级别日志"""
if args:
self.logger.warning(f"{self.task_code} | {message}", *args)
else:
self.logger.warning(f"{self.task_code} | {message}")
def error(self, message: str, *args, exc_info: bool = False):
"""记录错误级别日志"""
self.counts["errors"] += 1
if args:
self.logger.error(f"{self.task_code} | {message}", *args, exc_info=exc_info)
else:
self.logger.error(f"{self.task_code} | {message}", exc_info=exc_info)
def set_counts(self, **counts):
"""直接设置计数"""
for key, value in counts.items():
if key in self.counts:
self.counts[key] = value
def add_counts(self, **counts):
"""累加计数"""
for key, value in counts.items():
if key in self.counts:
self.counts[key] += value
def set_verification_result(self, result: dict):
"""设置校验结果"""
self.verification_result = result
def end(self, status: str = "成功", error_message: Optional[str] = None) -> str:
"""
记录任务结束,返回格式化的总结
Args:
status: 状态 ("成功" / "失败" / "取消")
error_message: 错误信息(如果失败)
Returns:
格式化的任务总结字符串
"""
self.end_time = datetime.now()
self.status = status
# 计算耗时
if self.start_time:
elapsed = (self.end_time - self.start_time).total_seconds()
elapsed_str = self._format_duration(elapsed)
else:
elapsed = 0
elapsed_str = "-"
# 生成总结
summary = self._format_summary(status, elapsed_str, error_message)
# 记录日志
if status == "成功":
self.logger.info("\n%s", summary)
else:
self.logger.error("\n%s", summary)
return summary
def _format_duration(self, seconds: float) -> str:
"""格式化时长"""
if seconds < 60:
return f"{seconds:.1f}"
elif seconds < 3600:
mins = int(seconds // 60)
secs = seconds % 60
return f"{mins}{secs:.0f}"
else:
hours = int(seconds // 3600)
mins = int((seconds % 3600) // 60)
return f"{hours}{mins}"
def _format_summary(
self,
status: str,
elapsed_str: str,
error_message: Optional[str] = None,
) -> str:
"""格式化任务总结"""
lines = [
"╔══════════════════════════════════════════════════════════════╗",
"║ 任务执行总结 ║",
"╠══════════════════════════════════════════════════════════════╣",
f"║ 任务代码: {self.task_code:<50}",
f"║ 执行状态: {status:<50}",
]
if self.start_time and self.end_time:
time_range = f"{self.start_time.strftime('%Y-%m-%d %H:%M:%S')} ~ {self.end_time.strftime('%H:%M:%S')} ({elapsed_str})"
lines.append(f"║ 执行时间: {time_range:<50}")
lines.extend([
"╠══════════════════════════════════════════════════════════════╣",
"║ 数据统计 ║",
f"║ - 获取记录: {self.counts['fetched']:>10,}",
f"║ - 新增记录: {self.counts['inserted']:>10,}",
f"║ - 更新记录: {self.counts['updated']:>10,}",
f"║ - 跳过记录: {self.counts['skipped']:>10,}",
f"║ - 错误记录: {self.counts['errors']:>10,}",
])
# 校验结果
if self.verification_result:
backfilled_missing = self.verification_result.get(
"backfilled_missing_count",
self.verification_result.get("backfilled_count", 0),
)
backfilled_mismatch = self.verification_result.get("backfilled_mismatch_count", 0)
lines.extend([
"╠══════════════════════════════════════════════════════════════╣",
"║ 校验结果 ║",
f"║ - 源数据量: {self.verification_result.get('source_count', 0):>10,}",
f"║ - 目标数据量: {self.verification_result.get('target_count', 0):>10,}",
f"║ - 缺失补齐: {backfilled_missing:>10,}",
f"║ - 不一致补齐: {backfilled_mismatch:>10,}",
])
# 错误信息
if error_message:
lines.extend([
"╠══════════════════════════════════════════════════════════════╣",
f"║ 错误信息: {error_message[:50]:<50}",
])
lines.append("╚══════════════════════════════════════════════════════════════╝")
return "\n".join(lines)
def get_result(self) -> dict:
"""获取任务结果字典"""
elapsed = 0
if self.start_time and self.end_time:
elapsed = (self.end_time - self.start_time).total_seconds()
return {
"task_code": self.task_code,
"status": self.status,
"start_time": self.start_time.isoformat() if self.start_time else None,
"end_time": self.end_time.isoformat() if self.end_time else None,
"elapsed_seconds": elapsed,
"counts": self.counts.copy(),
"extra_info": self.extra_info.copy(),
"verification_result": self.verification_result,
}
def configure_task_logging(
name: str = "fq_etl",
level: str = "INFO",
) -> logging.Logger:
"""
配置任务日志
Args:
name: 日志器名称
level: 日志级别
Returns:
配置好的日志器
"""
logger = logging.getLogger(name)
logger.setLevel(getattr(logging, level.upper(), logging.INFO))
# 清除已有处理器
logger.handlers.clear()
# 添加控制台处理器
handler = logging.StreamHandler()
handler.setLevel(logging.DEBUG)
# 设置格式
formatter = logging.Formatter(
UNIFIED_LOG_FORMAT,
UNIFIED_DATE_FORMAT,
)
handler.setFormatter(formatter)
logger.addHandler(handler)
logger.propagate = False
return logger