awk 分隔方括号，空值当列问题

skypeleo

2014-11-11 09:39:48: ip[192.168.10.15] uid[585721] msg[get cache userinfo]
2014-11-05 09:58:33: ip[192.168.10.14] uid[585721] msg[get cache userinfo] uid[] word[555]
2014-11-06 09:58:33: ip[192.168.10.15] uid[585721] msg[get cache userinfo] key[we]
2014-11-12 09:58:34: ip[192.168.10.16] uid[585721] msg[get cache userinfo] uid[]
2014-11-05 09:58:34: ip[192.168.10.17] uid[] msg[get cache userinfo] search[you]

1.去除第2列后面的冒号：
2.如果这个列有方括号，取方括号内的内容
3.用“|”分隔
4.前5列一样且一定有，后面可追加自定义的列
5.遇到某些列方括号里面为空 也要当成列
6.文件大小2-4G


结果为
2014-11-11|09:39:48|192.168.10.15|585721|get cache userinfo
2014-11-05|09:58:33|192.168.10.14|585721|get cache userinfo||555
2014-11-06|09:58:33|192.168.10.15|585721|get cache userinfo|we
2014-11-12|09:58:34|192.168.10.16|585721|get cache userinfo||
2014-11-05|09:58:34|192.168.10.17||get cache userinfo|you

ly5066113

回复 1# skypeleo


try:

1. awk -F '[][]' '{printf gensub(/(.*) (.*):[^:]*$/,"\\1|\\2",1,$1);for(i=2;i<NF;i+=2)printf "|"$i;print ""}' file

复制代码

skypeleo

回复 2# ly5066113

秒杀  谢谢哒哒


   

reyleon

1. #!/usr/bin/python
   
2. # -*- coding: utf-8 -*-
   
3. 
   
4. import fileinput, sys, re
   
5. for line in fileinput.input(sys.argv[1]):
   
6.     l = []
   
7.     record = re.findall('(\d+-\d+-\d+ \d+:\d+:\d+)|\[(.*?)\]',line)
   
8.     try:
   
9.         l.append(record[0][0])
   
10.         for t in record[1:]:
    
11.             l.append(t[1])
    
12.     except Exception, e:
    
13.         pass
    
14.     print '|'.join(l)

复制代码

也可用于你之前的那个贴: http://bbs.chinaunix.net/thread-4160126-1-1.html

chengchow

1. sed -r 's/(^\S+)\s+(\S+):/\1|\2|/g;s/(\s+)\S+\[([^]]*)\]\S*/\1\2|/g' file
   
2. 2014-11-11|09:39:48| 192.168.10.15| 585721| get cache userinfo|
   
3. 2014-11-05|09:58:33| 192.168.10.14| 585721| get cache userinfo| | 555|
   
4. 2014-11-06|09:58:33| 192.168.10.15| 585721| get cache userinfo| we|
   
5. 2014-11-12|09:58:34| 192.168.10.16| 585721| get cache userinfo| |
   
6. 2014-11-05|09:58:34| 192.168.10.17| | get cache userinfo| you|
   

复制代码

ntwarren

gawk 'BEGIN{FS="[]:] +[a-z]+\\[|] *";OFS="|"}{sub(" ","|",$1);NF=NF-1}1'