首页 > 代码库 > python logging
python logging
参考：
https://docs.python.org/2/howto/logging.html#logging-basic-tutorial
https://docs.python.org/2/library/logging.config.html#logging-config-api
https://docs.python.org/2/howto/logging-cookbook.html#logging-cookbook
http://victorlin.me/posts/2012/08/26/good-logging-practice-in-python
背景
Logging 用来追踪程序运行过程中发生的事件, 事件按其重要性可划分为不同的级别: DEBUG, INFO, WARNING, ERROR, CRITICAL.
- DEBUG 运行中的详细信息, 特别用于代码调试
- INFO 确认事情按照期待的方式发生了
- WARNING 一些超出预期的事情发生或将要发生, 但是程序仍可以按照期望的方式继续运行
- ERROR 发生了更严重的事情, 程序的一些功能会受到影响, 无法进行
- CRITICAL 程序炸了
默认的级别是 WARNING, DEBUG 和 INFO 会被忽略.
输出到终端
import logginglogging.info("info information") # will print nothinglogging.warn("warn infotmation") # will print a message to the consoleOutput:WARNING:root:warn infotmation
观察输出, WARNING 是级别, warn infomation 是记录的信息, root是啥? root 是 logger 的名字.
虽然我们没有显式创建 root logger, root logger 是默认创建的.
输出到文件
import logginglogging.basicConfig(filename="example.log", level=DEBUG)logging.debug("debug information")logging.info("info information")logging.error("error information")example.logDEBUG:root:debug informationINFO:root:info informationERROR:root:error information
注意到 basicConfig 方法, 该方法创建一个使用默认 Formatter 的 StreamHandler 并添加到 root logger. 如果 root logger 已经配置了 handler, 调用该方法 will do nothing.
参数 filename 用来指定日志输出的文件名, 一旦指定了 filename, StreamHandler 不会被创建, FileHandler 会代替 SteamHandler 添加到 root logger.
名词解释
Logger 提供应用直接调用的接口
Handler 将 logger 创建的日志发送到设定的目的地
Filter 过滤日志, 决定哪些可以输出(暂时用不到, 略)
Formatter 确定日志最终的输出格式
Logger
- Logger.setLevel() 设定级别
- Logger.addHandler(), Logger.removeHandler() 添加或删除 Handler
- Logger.addFilter(), Logger.removeFilter() 添加或删除 Filter
- Logger.debug(), Logger.info(), Logger.warn(), Logger.error(), Logger.critical()
- Logger.exception() 输出与 Logger.error() 相似, 附加异常详情, 用于异常处理
Handler
- Handler.setLevel() 设定级别, 决定是否继续传递
- Handler.setFormatter() 设定输出格式
- Handler.addFilter(), Handler.removeFilter() 添加或删除Filter
Formatter
- Formatter.__init__(fmt=None, datefmt=None) datefmt 默认为 "%Y-%m-%d %H:%M:%S" with the milliseconds tacked on at the end, fmt 默认为 "%(levelname)s:%(name)s:%(message)s"
同时输出到终端和文件, 终端级别为DEBUG, 文件级别为ERROR
import logginglogger = logging.getLogger(__name__)logger.setLevel(level=logging.DEBUG)formatter = logging.Formatter("%(asctime)s-%(name)s-%(levelname)s-%(message)s")console = logging.StreamHandler()console.setLevel(level=logging.DEBUG)console.setFormatter(formatter)logfile = logging.FileHandler("example.log")logfile.setLevel(level=logging.ERROR)logfile.setFormatter(formatter)logger.addHandler(console)logger.addHandler(logfile)logger.debug("debug information")logger.info("info information")logger.warn("warn information")logger.error("error information")logger.critical("critical information")Output:2016-09-03 23:49:17,207-__main__-DEBUG-debug information2016-09-03 23:49:17,207-__main__-INFO-info information2016-09-03 23:49:17,207-__main__-WARNING-warn information2016-09-03 23:49:17,207-__main__-ERROR-error information2016-09-03 23:49:17,207-__main__-CRITICAL-critical informationexample.log2016-09-03 23:49:17,207-__main__-ERROR-error information2016-09-03 23:49:17,207-__main__-CRITICAL-critical information
logging.FileHandler() 支持相对路径和绝对路径, 习惯上使用绝对路径会好一些
logging.config.dictConfig
从字典中读取配置信息, 配置信息可以存储为 json 或 yaml 格式, yaml 更易读, pip install pyyaml
version: 1disable_existing_logger: Falseformatters: brief: format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s"handlers: console: class: logging.StreamHandler level: DEBUG formatter: brief file: class: logging.handlers.RotatingFileHandler level: INFO formatter: brief filename: info.log maxBytes: 10500000 backupCount: 5 encoding: utf8 mail: class: logging.handlers.SMTPHandler level: ERROR formatter: brief mailhost: localhost fromaddr: username@test.com toaddrs: [receiver0@test.com, receiver1@test.com] subject: Program crashed!!!loggers: warthog: level: ERROR handlers: [file, mail] propagate: noroot: level: DEBUG handlers: [console]
varsion: 1 必须存在且必须为1
disable_existing_logger 默认为 True, 会禁用除当前模块的所有 logger
handlers console, file, mail 标识不同的 handler, class, level, formatter 指定特定的 Handler, 级别, 输出格式, 其他的参数由不同的 Handler 决定.
loggers warthog 标识不同的 logger, 使用方式为 logger.getLogger("warthog"), propagate 默认为 yes, warthog logger 的日志会向上传递到 root logger.
如果一个模块叫做 warthog.py, logger = logging.getLogger(__name__), 该 logger 就是 logger warthog
进阶
use __name__ as the logger name
__name__ 是当前模块的名字, 假设当前模块是 foo.bar.my_module, 只要对 logger foo 进行设置, 所有 foo. 中模块的 logger 会自动继承.
logger.exception
try: open(‘/path/to/does/not/exist‘, ‘rb‘)except (SystemExit, KeyboardInterrupt): raiseexcept Exception as e: logger.exception("Failed to open file") #logger.error("Failed to open file", exc_info=True)Output:ERROR:__main__:Failed to open fileTraceback (most recent call last): File "example.py", line 6, in <module> open(‘/path/to/does/not/exist‘, ‘rb‘)IOError: [Errno 2] No such file or directory: ‘/path/to/does/not/exist‘
Do not get logger at the module level unless disable_existing_loggers is False
# not goodimport logginglogger = logging.getLogger(__name__)def foo(): logger.info("info message")# betterimport loggingdef foo(): logger = logging.getLogger(__name__) logger.info("info message")
python logging