日志分析想要一个正则

帮你写好了Python的正则的代码了：

# Version: 20190521
# Function: answer 
#           日志分析想要一个正则 - OSCHINA
#           https://www.oschina.net/question/251973_2306245
# Author: Crifan

import re

gInputLogStrList = [
  "[DEBUG][20190513 00:00:00,300][org.apache.ibatis.logging.commons.JakartaCommonsLoggingImpl:49][] 任意文本内容",
  # "[INFO][30-Oct-2017 10:04:44.671][org.apache.catalina.startup.VersionLoggerListener.log Server:51][] Apache Tomcat/8.5.15",
  "[INFO][20190521 16:15:21,528][org.apache.catalina.startup.VersionLoggerListener.log Server:51][] Apache Tomcat/8.5.15",
]

def extractLog():
  for eachLogStr in gInputLogStrList:
    print("eachLogStr=%s" % eachLogStr)

    # foundLog = re.search("\[(?P[A-Z]+)\]", eachLogStr)
    # foundLog = re.search("\[(?P[A-Z]+)\]\s*\[(?P[\w\s\-\.\,\:]+)\]", eachLogStr)
    # foundLog = re.search("\[(?P[A-Z]+)\]\s*\[(?P[^\[\]]+)\]", eachLogStr)
    # foundLog = re.search("\[(?P[A-Z]+)\]\[(?P[^\[\]]+)\]\[(?P[^:]+):(?P\d+)\]", eachLogStr)

    # Named Group = 带 命名组
    # foundLog = re.search("^\[(?P[A-Z]+)\]\[(?P[^\[\]]+)\]\[(?P[^:]+):(?P\d+)\]\[\]\s*(?P.+)$", eachLogStr)
    # 后记：如果确保时间格式是固定的：20190513 00:00:00,300，那么可以改为：
    foundLog = re.search("^\[(?P[A-Z]+)\]\[(?P\d+\s+\d+:\d+:\d+),\d+\]\[(?P[^:]+):(?P\d+)\]\[\]\s*(?P.+)$", eachLogStr)
    if foundLog:
      logLevel = foundLog.group("logLevel")
      logTime = foundLog.group("logTime")
      logModule = foundLog.group("logModule")
      logLineNum = foundLog.group("logLineNum")
      logContent = foundLog.group("logContent")
      print("logLevel=%s, logTime=%s, logModule=%s, logLineNum=%s, logContent=%s" % (logLevel, logTime, logModule, logLineNum, logContent))

    # No Named Group = 不带命名组
    foundLog = re.search("^\[([A-Z]+)\]\[(\d+\s+\d+:\d+:\d+),\d+\]\[([^:]+):(\d+)\]\[\]\s*(.+)$", eachLogStr)
    if foundLog:
      logLevel = foundLog.group(1)
      logTime = foundLog.group(2)
      logModule = foundLog.group(3)
      logLineNum = foundLog.group(4)
      logContent = foundLog.group(5)
      print("logLevel=%s, logTime=%s, logModule=%s, logLineNum=%s, logContent=%s" % (logLevel, logTime, logModule, logLineNum, logContent))

    # Final Result:
    # logLevel=DEBUG, logTime=20190513 00:00:00,300, logModule=org.apache.ibatis.logging.commons.JakartaCommonsLoggingImpl, logLineNum=49, logContent=任意文本内容
    # logLevel=INFO, logTime=30-Oct-2017 10:04:44.671, logModule=org.apache.catalina.startup.VersionLoggerListener.log Server, logLineNum=51, logContent=Apache Tomcat/8.5.15

    # logLevel=DEBUG, logTime=20190513 00:00:00, logModule=org.apache.ibatis.logging.commons.JakartaCommonsLoggingImpl, logLineNum=49, logContent=任意文本内容
    # logLevel=INFO, logTime=20190521 16:15:21, logModule=org.apache.catalina.startup.VersionLoggerListener.log Server, logLineNum=51, logContent=Apache Tomcat/8.5.15

if __name__ == "__main__":
    extractLog()

说明：

• 根据你贴出的“参考示例”，代码中的logLineNum，估计对应的是ITEM_TID，但是你此处明显只是 行号
  • 是你自己搞错了，还是说明有误，需要你自己查证
• 代码中 包含2套规则，你根据自己情况选择使用
  • 一个是命名的组
  • 一个是 不带命名的组
• 如果想要把上述python的正则换成其他语言的正则，比如perl，php，java等，适当修改即可得到，我就不赘述了