堆栈内存溢出
  繁体   English   中英

如何打印与 awk 多次匹配的列的第一次出现

[英]how to print the first occurence of a column matching more than once with awk

 原文  2021-04-22 18:21:54  5   3    bash/ awk/ grep

问题描述

我有一个包含所有备份的 log_file 和一个值为 yes 的列，表示保留策略不会删除它（保留）。 对于特定的 vmname，可能有 1 行或更多行具有该保留列 = yes。

我的输入是：

=    FULL     ==   20210105   ==     2100     == ASR-FULL-20210105-2100 ==  YES
=    FULL     ==   20210202   ==     2100     == ASR-FULL-20210202-2100 ==  YES
=    FULL     ==   20210302   ==     2100     == ASR-FULL-20210302-2100 ==  YES
=    FULL     ==   20210406   ==     2100     == ASR-FULL-20210406-2100 ==  YES
=    FULL     ==   20210105   ==     2146     == DNS10_7-FULL-20210105-2146 ==  YES
=    FULL     ==   20210202   ==     2153     == DNS10_7-FULL-20210202-2153 ==  YES
=    FULL     ==   20210302   ==     2148     == DNS10_7-FULL-20210302-2148 ==  YES
=    FULL     ==   20210406   ==     2122     == DNS10_7-FULL-20210406-2122 ==  YES
=    FULL     ==   20210105   ==     2105     == execnet.0-FULL-20210105-2105 ==  YES
=    FULL     ==   20210202   ==     2106     == execnet.0-FULL-20210202-2106 ==  YES
=    FULL     ==   20210302   ==     2106     == execnet.0-FULL-20210302-2106 ==  YES
=    FULL     ==   20210406   ==     2105     == execnet.0-FULL-20210406-2105 ==  YES
=    FULL     ==   20210106   ==     0200     == Prtgadmin.0-FULL-20210106-0200 ==  YES
=    FULL     ==   20210105   ==     2216     == sandbox.0-FULL-20210105-2216 ==  YES
=    FULL     ==   20210202   ==     2227     == sandbox.0-FULL-20210202-2227 ==  YES
=    FULL     ==   20210406   ==     2152     == sandbox.0-FULL-20210406-2152 ==  YES
=    FULL     ==   20210105   ==     2236     == wwwp.0-FULL-20210105-2236 ==  YES
=    FULL     ==   20210202   ==     2249     == wwwp.0-FULL-20210202-2249 ==  YES
=    FULL     ==   20210105   ==     2259     == wwws.0-FULL-20210105-2259 ==  YES
=    FULL     ==   20210202   ==     2314     == wwws.0-FULL-20210202-2314 ==  YES
=    FULL     ==   20210105   ==     2259     == webhost.0-FULL-20210105-2259 ==  YES

我想要的 output 是打印 n-1 个最旧的匹配项（前 n-1 个）

ASR-FULL-20210105-2100        
ASR-FULL-20210202-2100         
ASR-FULL-20210302-2100         
DNS10_7-FULL-20210105-2146     
DNS10_7-FULL-20210202-2153     
DNS10_7-FULL-20210302-2148     
execnet.0-FULL-20210105-2105  
execnet.0-FULL-20210202-2106   
execnet.0-FULL-20210302-2106   
sandbox.0-FULL-20210105-2216   
sandbox.0-FULL-20210202-2227   
wwwp.0-FULL-20210105-2236     
wwws.0-FULL-20210105-2259

到目前为止，我可以通过运行以下 awk 命令获得以下结果，但它显示的是最近的匹配项。 理想情况下，我还希望有一个 awk 命令。 年份过滤器并不那么重要。

# cat bkp_list.log| grep -E '*2021.*YES'| awk -F[==-] 'cnt[$8]++{if (cnt[$8]>1) print prev=$0;next}' |awk -F[==] '{print $8}' 

ASR-FULL-20210202-2100
ASR-FULL-20210302-2100
ASR-FULL-20210406-2100
DNS10_7-FULL-20210202-2153
DNS10_7-FULL-20210302-2148
DNS10_7-FULL-20210406-2122
execnet.0-FULL-20210202-2106
execnet.0-FULL-20210302-2106
execnet.0-FULL-20210406-2105
sandbox.0-FULL-20210202-2227
sandbox.0-FULL-20210406-2152
wwwp.0-FULL-20210202-2249
wwws.0-FULL-20210202-2314

谢谢

3 个解决方案

解决方案1
 3 已采纳  2021-04-22 19:49:22

要打印$8的 substring 的最后一场比赛，您可以使用此awk ：

awk '
$NF != "YES" {next}
{
   s = $8
   sub(/-FULL-.*/, "", s)
}
s == ps {
   print pval
}
{
   ps = s
   pval = $8
}' file

ASR-FULL-20210105-2100
ASR-FULL-20210202-2100
ASR-FULL-20210302-2100
DNS10_7-FULL-20210105-2146
DNS10_7-FULL-20210202-2153
DNS10_7-FULL-20210302-2148
execnet.0-FULL-20210105-2105
execnet.0-FULL-20210202-2106
execnet.0-FULL-20210302-2106
sandbox.0-FULL-20210105-2216
sandbox.0-FULL-20210202-2227
wwwp.0-FULL-20210105-2236
wwws.0-FULL-20210105-2259

或一个班轮：

awk '$NF != "YES"{next} {s=$8; sub(/-FULL-.*/, "", s)} s == ps {print pval} {ps = s; pval=$8}' file

解决方案2
 1  2021-04-22 18:37:04

如果要过滤带有 YES 的列，可以使用块前的条件表达式进行过滤

$ cat file
=    FULL     ==   20210105   ==     2100     == ASR-FULL-20210105-2100 ==  NO
=    FULL     ==   20210202   ==     2100     == ASR-FULL-20210202-2100 ==  YES
=    FULL     ==   20210302   ==     2100     == ASR-FULL-20210302-2100 ==  YES
=    FULL     ==   20210406   ==     2100     == ASR-FULL-20210406-2100 ==  YES
=    FULL     ==   20210105   ==     2146     == DNS10_7-FULL-20210105-2146 ==  YES
=    FULL     ==   20210202   ==     2153     == DNS10_7-FULL-20210202-2153 ==  YES
=    FULL     ==   20210302   ==     2148     == DNS10_7-FULL-20210302-2148 ==  YES
=    FULL     ==   20210406   ==     2122     == DNS10_7-FULL-20210406-2122 ==  YES
=    FULL     ==   20210105   ==     2105     == execnet.0-FULL-20210105-2105 ==  YES
=    FULL     ==   20210202   ==     2106     == execnet.0-FULL-20210202-2106 ==  YES
=    FULL     ==   20210302   ==     2106     == execnet.0-FULL-20210302-2106 ==  YES
=    FULL     ==   20210406   ==     2105     == execnet.0-FULL-20210406-2105 ==  YES
=    FULL     ==   20210106   ==     0200     == Prtgadmin.0-FULL-20210106-0200 ==  YES
=    FULL     ==   20210105   ==     2216     == sandbox.0-FULL-20210105-2216 ==  YES
=    FULL     ==   20210202   ==     2227     == sandbox.0-FULL-20210202-2227 ==  YES
=    FULL     ==   20210406   ==     2152     == sandbox.0-FULL-20210406-2152 ==  YES
=    FULL     ==   20210105   ==     2236     == wwwp.0-FULL-20210105-2236 ==  YES
=    FULL     ==   20210202   ==     2249     == wwwp.0-FULL-20210202-2249 ==  YES
=    FULL     ==   20210105   ==     2259     == wwws.0-FULL-20210105-2259 ==  YES
=    FULL     ==   20210202   ==     2314     == wwws.0-FULL-20210202-2314 ==  YES
=    FULL     ==   20210105   ==     2259     == webhost.0-FULL-20210105-2259 ==  YES


$ awk ' $NF == "YES" { print $(NF-2) }' file
ASR-FULL-20210202-2100
ASR-FULL-20210302-2100
ASR-FULL-20210406-2100
DNS10_7-FULL-20210105-2146
DNS10_7-FULL-20210202-2153
DNS10_7-FULL-20210302-2148
DNS10_7-FULL-20210406-2122
execnet.0-FULL-20210105-2105
execnet.0-FULL-20210202-2106
execnet.0-FULL-20210302-2106
execnet.0-FULL-20210406-2105
Prtgadmin.0-FULL-20210106-0200
sandbox.0-FULL-20210105-2216
sandbox.0-FULL-20210202-2227
sandbox.0-FULL-20210406-2152
wwwp.0-FULL-20210105-2236
wwwp.0-FULL-20210202-2249
wwws.0-FULL-20210105-2259
wwws.0-FULL-20210202-2314
webhost.0-FULL-20210105-2259


$ awk ' $NF == "NO" { print $(NF-2) }' file
ASR-FULL-20210105-2100
$

** 注意我将第一行 YES 更改为 NO 以检查正确的行为

无论如何，如果您需要进行任何其他特殊过滤，例如检查年份，请指定

解决方案3
 1  2021-04-22 21:21:23

对于 gensub() 使用 GNU awk：

$ tac file | awk '$NF=="YES" && seen[gensub(/-.*/,"",1,$8)]++{print $8}' | tac
ASR-FULL-20210105-2100
ASR-FULL-20210202-2100
ASR-FULL-20210302-2100
DNS10_7-FULL-20210105-2146
DNS10_7-FULL-20210202-2153
DNS10_7-FULL-20210302-2148
execnet.0-FULL-20210105-2105
execnet.0-FULL-20210202-2106
execnet.0-FULL-20210302-2106
sandbox.0-FULL-20210105-2216
sandbox.0-FULL-20210202-2227
wwwp.0-FULL-20210105-2236
wwws.0-FULL-20210105-2259

或使用任何 awk：

$ tac file | awk '$NF!="YES"{next} {k=$8; sub(/-.*/,"",k)} seen[k]++{print $8}' | tac
ASR-FULL-20210105-2100
ASR-FULL-20210202-2100
ASR-FULL-20210302-2100
DNS10_7-FULL-20210105-2146
DNS10_7-FULL-20210202-2153
DNS10_7-FULL-20210302-2148
execnet.0-FULL-20210105-2105
execnet.0-FULL-20210202-2106
execnet.0-FULL-20210302-2106
sandbox.0-FULL-20210105-2216
sandbox.0-FULL-20210202-2227
wwwp.0-FULL-20210105-2236
wwws.0-FULL-20210105-2259

暂无
暂无

声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 在第n次出现匹配模式之间打印awk AWK打印bash变量不止一次 awk 访问变量不止一次 使用带有出现多次的shell重命名文件 匹配文件的第一列与awk，引号有困难 awk在第1列中打印大于零的行 使用awk或sed打印第一列并通过电子邮件发送 如何使用awk计算列中负值和正值的出现？ csv中的awk命令：如何打印多个条目？ awk：select 匹配词后的第一列和列中的值
相关标签
 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM