1. 基本用法

import logging
logging.basicConfig(level = logging.INFO,format = '%(asctime)s - %(name)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)

logger.info("Start print log")
logger.debug("Do something")
logger.warning("Something maybe fail.")
logger.info("Finish")

运行时，控制台输出：

===
2019-02-16 22:06:37,453 - __main__ - INFO - 5 Start print log
2019-02-16 22:06:37,476 - __main__ - WARNING - 7 Something maybe fail.
2019-02-16 22:06:37,503 - __main__ - INFO - 8 Finished

logging中可以选择很多消息级别，如debug、info、warning、error以及critical。通过赋予logger或者handler不同的级别，开发者就可以只输出错误信息到特定的记录文件，或者在调试时只记录调试信息。
例如，将logger的级别改为DEBUG，再观察一下输出结果，

===
2019-02-16 22:08:14,875 - __main__ - INFO - 5 Start print log
2019-02-16 22:08:14,923 - __main__ - DEBUG - 6 Do something
2019-02-16 22:08:14,947 - __main__ - WARNING - 7 Something maybe fail.
2019-02-16 22:08:14,968 - __main__ - INFO - 8 Finished

logging.basicConfig函数各参数：

filename：指定日志文件名；

filemode：和file函数意义相同，指定日志文件的打开模式，‘w’或者’a’；

format：指定输出的格式和内容，format可以输出很多有用的信息，

参数：作用

%(levelno)s：打印日志级别的数值
%(levelname)s：打印日志级别的名称
%(pathname)s：打印当前执行程序的路径，其实就是sys.argv[0]
%(filename)s：打印当前执行程序名
%(funcName)s：打印日志的当前函数
%(lineno)d：打印日志的当前行号
%(asctime)s：打印日志的时间
%(thread)d：打印线程ID
%(threadName)s：打印线程名称
%(process)d：打印进程ID
%(message)s：打印日志信息

2. 将日志写入文件
   设置logging，创建一个FileHandler，并对输出消息的格式进行设置，将其添加到logger，然后将日志写入到指定的文件中，

import logging
logger = logging.getLogger(__name__)
logger.setLevel(level = logging.INFO)
handler = logging.FileHandler("log.txt")
handler.setLevel(logging.INFO)
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
handler.setFormatter(formatter)
logger.addHandler(handler)

logger.info("Start print log")
logger.debug("Do something")
logger.warning("Something maybe fail.")
logger.info("Finished")

log.txt中日志数据为：

2019-02-16 22:16:05,131 - __main__ - INFO - Start print log
2019-02-16 22:16:05,132 - __main__ - WARNING - Something maybe fail.
2019-02-16 22:16:05,132 - __main__ - INFO - Finished

3. 将日志同时输出到屏幕和日志文件
   logger中添加StreamHandler，可以将日志输出到屏幕上:

import logging
logger = logging.getLogger(__name__)
logger.setLevel(level = logging.INFO)
handler = logging.FileHandler("log.txt")
handler.setLevel(logging.INFO)
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
handler.setFormatter(formatter)

console = logging.StreamHandler()
console.setLevel(logging.INFO)
console.setFormatter(formatter)

logger.addHandler(handler)
logger.addHandler(console)

logger.info("Start print log")
logger.debug("Do something")
logger.warning("Something maybe fail.")
logger.info("Finished")

可以在log.txt文件和控制台中看到:

2019-02-16 22:16:05,131 - __main__ - INFO - Start print log
2019-02-16 22:16:05,132 - __main__ - WARNING - Something maybe fail.
2019-02-16 22:16:05,132 - __main__ - INFO - Finished

可以发现，logging有一个日志处理的主对象，其他处理方式都是通过addHandler添加进去，logging中包含的handler主要有如下几种，
handler名称：位置；作用

StreamHandler：logging.StreamHandler；日志输出到流，可以是sys.stderr，sys.stdout或者文件
FileHandler：logging.FileHandler；日志输出到文件
BaseRotatingHandler：logging.handlers.BaseRotatingHandler；基本的日志回滚方式
RotatingHandler：logging.handlers.RotatingHandler；日志回滚方式，支持日志文件最大数量和日志文件回滚
TimeRotatingHandler：logging.handlers.TimeRotatingHandler；日志回滚方式，在一定时间区域内回滚日志文件
SocketHandler：logging.handlers.SocketHandler；远程输出日志到TCP/IP sockets
DatagramHandler：logging.handlers.DatagramHandler；远程输出日志到UDP sockets
SMTPHandler：logging.handlers.SMTPHandler；远程输出日志到邮件地址
SysLogHandler：logging.handlers.SysLogHandler；日志输出到syslog
NTEventLogHandler：logging.handlers.NTEventLogHandler；远程输出日志到Windows NT/2000/XP的事件日志
MemoryHandler：logging.handlers.MemoryHandler；日志输出到内存中的指定buffer
HTTPHandler：logging.handlers.HTTPHandler；通过"GET"或者"POST"远程输出到HTTP服务器

4. 日志回滚
   使用RotatingFileHandler，可以实现日志回滚：

import logging
from logging.handlers import RotatingFileHandler
logger = logging.getLogger(__name__)
logger.setLevel(level = logging.INFO)
#定义一个RotatingFileHandler,最多备份3个日志文件，每个日志文件最大1k
rHandler = RotatingFileHandler("log.txt", maxBytes = 1*1024, backupCount = 3)
rHandler.setLevel(logging.INFO)
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
rHandler.setFormatter(formatter)

console = logging.StreamHandler()
console.setLevel(logging.INFO)
console.setFormatter(formatter)

logger.addHandler(rHandler)
logger.addHandler(console)

logger.info("Start print log")
logger.debug("Do something")
logger.warning("Something maybe fail.")
logger.info("Finished")

输出目录中：

log.txt
log.txt.1
log.txt.2
log.txt.3

5. 设置消息的等级
   日志等级：使用范围

FATAL：致命错误
CRITICAL：特别糟糕的事情，如内存耗尽、磁盘空间为空，一般很少使用
ERROR：发生错误时，如IO操作失败或者连接问题
WARNING：发生很重要的事件，但是并不是错误时，如用户登录密码错误
INFO：处理请求或者状态变化等日常事务
DEBUG：调试过程中使用DEBUG等级，如算法中每个循环的中间状态

6. 捕获traceback
   Python中的traceback模块被用于跟踪异常返回信息，可以在logging中记录下traceback

import logging
logger = logging.getLogger(__name__)
logger.setLevel(level = logging.INFO)
handler = logging.FileHandler("log.txt")
handler.setLevel(logging.INFO)
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
handler.setFormatter(formatter)

console = logging.StreamHandler()
console.setLevel(logging.INFO)

logger.addHandler(handler)
logger.addHandler(console)

logger.info("Start print log")
logger.debug("Do something")
logger.warning("Something maybe fail.")
try:
    open("sklearn.txt","rb")
except (SystemExit,KeyboardInterrupt):
    raise
except Exception:
    logger.error("Faild to open sklearn.txt from logger.error",exc_info = True)
    
logger.info("Finish")

也可以使用logger.exception(msg,_args)，它等价于logger.error(msg,exc_info = True,_args)
控制台和日志文件log.txt中输出:

Start print log
Something maybe fail.
Faild to open sklearn.txt from logger.error
Traceback (most recent call last):
  File "D:\Users\xk\Desktop\新建文件夹 (2)\test.py", line 19, in <module>
    open("sklearn.txt","rb")
FileNotFoundError: [Errno 2] No such file or directory: 'sklearn.txt'
Finish

7. 多模块使用logging
   主模块mainModule.py:

import logging
import subModule
logger = logging.getLogger("mainModule")
logger.setLevel(level = logging.INFO)
handler = logging.FileHandler("log.txt")
handler.setLevel(logging.INFO)
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
handler.setFormatter(formatter)

console = logging.StreamHandler()
console.setLevel(logging.INFO)
console.setFormatter(formatter)

logger.addHandler(handler)
logger.addHandler(console)


logger.info("creating an instance of subModule.subModuleClass")
a = subModule.SubModuleClass()
logger.info("calling subModule.subModuleClass.doSomething")
a.doSomething()
logger.info("done with  subModule.subModuleClass.doSomething")
logger.info("calling subModule.some_function")
subModule.som_function()
logger.info("done with subModule.some_function")

子模块subModule.py:

import logging

module_logger = logging.getLogger("mainModule.sub")
class SubModuleClass(object):
    def __init__(self):
        self.logger = logging.getLogger("mainModule.sub.module")
        self.logger.info("creating an instance in SubModuleClass")
    def doSomething(self):
        self.logger.info("do something in SubModule")
        a = []
        a.append(1)
        self.logger.debug("list a = " + str(a))
        self.logger.info("finish something in SubModuleClass")

def som_function():
    module_logger.info("call function some_function")

执行之后，在控制和日志文件log.txt输出：

===
2019-02-17 09:04:49,599 - mainModule - INFO - creating an instance of subModule.subModuleClass
2019-02-17 09:04:49,630 - mainModule.sub.module - INFO - creating an instance in SubModuleClass
2019-02-17 09:04:49,646 - mainModule - INFO - calling subModule.subModuleClass.doSomething
2019-02-17 09:04:49,674 - mainModule.sub.module - INFO - do something in SubModule
2019-02-17 09:04:49,685 - mainModule.sub.module - INFO - finish something in SubModuleClass
2019-02-17 09:04:49,699 - mainModule - INFO - done with  subModule.subModuleClass.doSomething
2019-02-17 09:04:49,706 - mainModule - INFO - calling subModule.some_function
2019-02-17 09:04:49,706 - mainModule.sub - INFO - call function some_function
2019-02-17 09:04:49,706 - mainModule - INFO - done with subModule.some_function

首先在主模块定义了logger’mainModule’，并对它进行了配置，就可以在解释器进程里面的其他地方通过getLogger(‘mainModule’)得到的对象都是一样的，不需要重新配置，可以直接使用。定义的该logger的子logger，都可以共享父logger的定义和配置，所谓的父子logger是通过命名来识别，任意以’mainModule’开头的logger都是它的子logger，例如’mainModule.sub’。

8. 通过JSON或者YAML文件配置logging模块
   logging.yaml配置文件：

version: 1
disable_existing_loggers: False
formatters:
        simple:
            format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
handlers:
    console:
            class: logging.StreamHandler
            level: DEBUG
            formatter: simple
            stream: ext://sys.stdout
    info_file_handler:
            class: logging.handlers.RotatingFileHandler
            level: INFO
            formatter: simple
            filename: info.log
            maxBytes: 10485760
            backupCount: 20
            encoding: utf8
    error_file_handler:
            class: logging.handlers.RotatingFileHandler
            level: ERROR
            formatter: simple
            filename: errors.log
            maxBytes: 10485760
            backupCount: 20
            encoding: utf8
loggers:
    my_module:
            level: ERROR
            handlers: [info_file_handler]
            propagate: no
root:
    level: INFO
    handlers: [console,info_file_handler,error_file_handler]

通过YAML加载配置文件，然后通过logging.dictConfig配置logging:

import yaml
import logging.config
import os

def setup_logging(default_path = "logging.yaml",default_level = logging.INFO,env_key = "LOG_CFG"):
    path = default_path
    value = os.getenv(env_key,None)
    if value:
        path = value
    if os.path.exists(path):
        with open(path,"r") as f:
            config = yaml.load(f)
            logging.config.dictConfig(config)
    else:
        logging.basicConfig(level = default_level)

def func():
    logging.info("start func")

    logging.info("exec func")

    logging.info("end func")

if __name__ == "__main__":
    setup_logging(default_path = "logging.yaml")
    func()

info.log输出为：

2019-02-17 10:21:52,319 - root - INFO - start func
2019-02-17 10:21:52,424 - root - INFO - exec func
2019-02-17 10:21:52,444 - root - INFO - end func

转：https://www.cnblogs.com/liujiacai/p/7804848.html