logstash grok插件中用正则表达式匹配java log

用logstash的grok插件，匹配下面格式的log

2018-04-17 23:42:10.335  INFO [main] s.b.c.e.t.TomcatEmbeddedServletContainer : Tomcat initialized with port(s): 8080 (http)

用表达式

%{DATA:day} %{DATA:time} %{DATA:level} \[%{DATA:thread}\] %{DATA:className} \: %{GREEDYDATA:msg}

匹配出来的结果是

{
  "day": [
    [
      "2018-04-17"
    ]
  ],
  "time": [
    [
      "23:42:10.335"
    ]
  ],
  "level": [
    [
      " INFO 7304 ---"
    ]
  ],
  "thread": [
    [
      "main"
    ]
  ],
  "className": [
    [
      "s.b.c.e.t.TomcatEmbeddedServletContainer"
    ]
  ],
  "msg": [
    [
      "Tomcat initialized with port(s): 8080"
    ]
  ]
}

可以看到时间被分成了两个字段，官方的表达式没有匹配中国时间的，于是我想到能不能自定义正则表达式，后来终于找到了。下面是我改进的表达式：

(?\S{10} \S{12}) %{DATA:level} \[%{DATA:thread}\] %{DATA:className} \: %{GREEDYDATA:msg}

结果时间字段完美解析了：

{
  "fullTime": [
    [
      "2018-04-17 23:42:10.335"
    ]
  ],
  "level": [
    [
      " INFO"
    ]
  ],
  "thread": [
    [
      "main"
    ]
  ],
  "className": [
    [
      "s.b.c.e.t.TomcatEmbeddedServletContainer"
    ]
  ],
  "msg": [
    [
      "Tomcat initialized with port(s): 8080"
    ]
  ]
}