Python 日志记录 - 不同的日志但相同的输出
Posted
技术标签:
【中文标题】Python 日志记录 - 不同的日志但相同的输出【英文标题】：Python logging - Different logs but the same output 【发布时间】：2012-10-26 15:31:57 【问题描述】：我有几个不同的解析器用于不同的网站，我还有一个名为 shared.py 的文件，它具有用于特殊解析的 lxml 函数和一个 base .py 文件，负责处理数据库（保存等）。 如果某事失败（未找到图像）或通过（我们找到图像），我需要将其记录到日志文件中，为此我使用标准日志记录模块。
我编写了一个 logger.py 文件，在其中创建了一个 Log 类，以便我可以在我的解析器或 base.py/shared.py 文件中调用它看起来像这样：
import logging
import os.path
__metaclass__ = type
class Log:
def __init__(self, filename, loggername="fetchers", path="logs/", formater="%(asctime)s %(levelname)s %(message)s"):
self.logger = logging.getLogger(loggername)
self.hdlr = logging.FileHandler(path + os.path.splitext(filename)[0] + ".log")
self.formater = logging.Formatter(formater)
self.hdlr.setFormatter(self.formater)
self.logger.addHandler(self.hdlr)
self.logger.setLevel(logging.INFO)
def info(self, msg):
self.logger.info(msg)
def warning(self, msg):
self.logger.warning(msg)
def error(self, msg):
self.logger.error(msg)
def critical(self, msg):
self.logger.critical(msg)
if __name__ == "__main__":
pass
文件夹层次结构如下所示；
/解析器 /Parsers/base.py /Parsers/shared.py /Parsers/parserone.py /Parsers/parsertwo.py /Parsers/parsersthree.py /解析器/... /Parsers/logs/base.log /Parsers/logs/shared.log /Parsers.logs/parserOne.log每个解析器都会导入记录器文件（base.py 和 shared.py 也一样），并像这样初始化记录器：
logger = Log(os.path.basename(__file__))
logger.warning("Something has happened..")
这工作正常，它确实写入日志，但问题是，base.py 写入有关查询等的日志，以及有关失败等的解析器（与 shared.py 相同） 问题是它正在写日志，但它们看起来都一样..
➜ logs tail parserOne.log
2012-10-26 16:35:21,250 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/scherpe-omzetdaling-televisiereclame/
2012-10-26 16:35:21,286 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/nominaties-mercurs-bekend2/
2012-10-26 16:35:21,322 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/traditionele-media-nog-steeds-populair-bij-jongeren/
2012-10-26 16:35:21,371 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/persgroep-blijft-sponsor-san/
2012-10-26 16:35:21,407 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/pg-overtreft-verwachtingen/
2012-10-26 16:35:21,443 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/discovery-networks-introduceert-discovery-client-productions/
2012-10-26 16:35:21,479 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/stoelendans-bij-wehkamp.nl/
2012-10-26 16:35:21,563 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/amazon-duikt-in-rode-cijfers/
2012-10-26 16:35:21,599 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/linkedin-rabobank-meest-populaire-werkgever/
2012-10-26 16:35:21,683 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/veronica-uitgeverij-wil-naar-amsterdam/
➜ logs tail base.log
2012-10-26 16:35:21,250 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/scherpe-omzetdaling-televisiereclame/
2012-10-26 16:35:21,286 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/nominaties-mercurs-bekend2/
2012-10-26 16:35:21,322 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/traditionele-media-nog-steeds-populair-bij-jongeren/
2012-10-26 16:35:21,371 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/persgroep-blijft-sponsor-san/
2012-10-26 16:35:21,407 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/pg-overtreft-verwachtingen/
2012-10-26 16:35:21,443 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/discovery-networks-introduceert-discovery-client-productions/
2012-10-26 16:35:21,479 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/stoelendans-bij-wehkamp.nl/
2012-10-26 16:35:21,563 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/amazon-duikt-in-rode-cijfers/
2012-10-26 16:35:21,599 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/linkedin-rabobank-meest-populaire-werkgever/
2012-10-26 16:35:21,683 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/veronica-uitgeverij-wil-naar-amsterdam/
➜ logs tail shared.log
2012-10-26 16:35:21,250 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/scherpe-omzetdaling-televisiereclame/
2012-10-26 16:35:21,286 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/nominaties-mercurs-bekend2/
2012-10-26 16:35:21,322 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/traditionele-media-nog-steeds-populair-bij-jongeren/
2012-10-26 16:35:21,371 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/persgroep-blijft-sponsor-san/
2012-10-26 16:35:21,407 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/pg-overtreft-verwachtingen/
2012-10-26 16:35:21,443 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/discovery-networks-introduceert-discovery-client-productions/
2012-10-26 16:35:21,479 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/stoelendans-bij-wehkamp.nl/
2012-10-26 16:35:21,563 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/amazon-duikt-in-rode-cijfers/
2012-10-26 16:35:21,599 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/linkedin-rabobank-meest-populaire-werkgever/
2012-10-26 16:35:21,683 INFO Data was sucessfuly saved: http://www.adformatie.nl/nieuws/bericht/veronica-uitgeverij-wil-naar-amsterdam/
为什么所有的日志文件都一样？如果它们是不同的文件？！
干杯。
【问题讨论】：
@Lanaru 你是对的，按照你说的做为我解决了这个问题。 太棒了，我发表了我的评论作为答案。请接受它以解决此问题。谢谢！ 【参考方案1】：看起来它使用loggerName 获取记录器，并且始终设置为"fetchers"。因此，您在任何地方都使用相同的记录器，这解释了为什么输出相同。
【讨论】：
【参考方案2】：您的记录器都称为“提取器”（因为您没有提供loggername 参数），因此您会在所有日志中看到相同的消息。
我建议为您的记录器添加一个名称，也许使用filters。
您可以将其应用于您的代码，如下所示：
logger = Log(os.path.basename(__file__), loggername='test')
def __init__(self, filename, loggername="fetchers", path="logs/", formater="%(asctime)s %(levelname)s %(message)s"):
self.logger = logging.getLogger(loggername)
self.hdlr = logging.FileHandler(path + os.path.splitext(filename)[0] + ".log")
self.hdlr.addFilter(logging.Filter(name=loggername))
...
【讨论】：
这个问题也解决了，你和 Lanaru 发布了答案，因为我不能同时接受，所以我接受了第一个，尽管我个人都接受了 ;-) 谢谢！ @BenMezger 不用担心，很高兴为您提供帮助！ :)以上是关于Python 日志记录 - 不同的日志但相同的输出的主要内容，如果未能解决你的问题，请参考以下文章