awk 分组问题，请指教

whatislinux

有这样一个日志文件，文件中 有errmsg，希望实现errmsg分类，并计算出现的次数，通过awk如何得到下面的输入，请高手指教！
                  类别                                                             次数
"errmsg":"请先开通个人账户"                                               5次
"errmsg":"当前>订单状态是已离店, 无法进行申请返现操作"         8次
"errmsg":"请先>开通个人账户"                                            4次


{"ver":"1.0","ret":false,"errmsg":"请先开通个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:02:43"}
###200####{"ver":"1.0","ret":false,"errmsg":"当前>订单状态是已离店, 无法进行申请返现操作","errcode":404,"time":"2014-06-22 00:03:00"}####7EBA2C6B-BAD7-4714-B
00####{"ver":"1.0","ret":false,"errmsg":"请先开通>个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:18:54"}
{"ver":"1.0","ret":false,"errmsg":"请先开通个人账>户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:20:29"}
####200####{"ver":"1.0","ret":false,"errmsg":"请先
开通个人账户","errcode":202,"data":
###200####{"ver":"1.0","ret":false,"errmsg":"请先>开通个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:24:18"}
#{"ver":"1.0","ret":false,"errmsg":"请先开通个人账
户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:29:36"}
{"ver":"1.0","ret":false,"errmsg":"请先
开通个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:29:41"}
##{"ver":"1.0","ret":false,"errmsg":"请先
开通个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:30:12"}
#{"ver":"1.0","ret":false,"errmsg":"当前订单状态是已
离店, 无法进行申请返现操作","errcode":404,"time":"2014-06-22 00:31:52"}####D8764
#{"ver":"1.0","ret":false,"errmsg":"请先
开通个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:38:22"}
#{"ver":"1.0","ret":false,"errmsg":"请先
开通个人账户","errcode":202,"data":"https://pay.ip-localhost/userAccount/account/my.htm","time":"2014-06-22 00:38:39"}

关阴月飞

假设errmsg中的内容只存在一个,号分隔：

1. awk -F, '/errmsg/{a[$3~/"$/?$3:$3FS$4]++}END{for(i in a)print i,a[i]}'  urfile

复制代码

reyleon

珍爱积分，远离世界杯......

whatislinux

回复 2# 关阴月飞


    感谢，我试试。

MeRcy_PM

1. sed ':a;N;s/\n//g;ta' urfile | grep -Po "\"errmsg\":\".*?\"" | awk  '{a[$0]++}END{for(i in a)print i,a[i]}'

复制代码

先去掉换行再匹配，最后awk。

whatislinux

回复 5# MeRcy_PM


    感谢