Question

我有以下文件，我需要创建一个awk命令来计算字段$ 15（错误代码）在每个字段$ 2（eventkey）中的次数。例如：

Video_load_time，错误代码：111，发生次数：2

Video_load_time，错误代码：3805，出现次数：3

app_launch_time，错误代码：111，出现次数：1

time|eventKey|Version|Model|SVersion|signal|net|State|atitude|long|subd|bupUsername|tvAccount|assetId|errorCode|errorDescription|duration

14201|video_load_time|5.7|i3|8.3|0.0|FI|GT|52.1147619|9829672936714|777|ghouso|444|6789|111|4464|7149       
4399784|playback_error|8.0|W8|33.2|0.0|FI|TED|468|071|078410X_.ca||2314831||3805|152rorDescript|0      
762|playback_error|70|ALFiee|4.2.2|0.0|IFI|AUED|4325|795|||81321761|3805|05|1529|
634|app_launch_time|5.0.0|SGHI317M|4.1.2|0.0|CELAR|AUTO_ATED|4588|64|180||||3805|yes|0     
1418|video_load_time|5.0.0.37|iP1|7.0.6|0.0|IN_HE_WIFI|AUTHEATED|45.47099941453838|477109099|||8455700500884828|111|N/A|77|9398     
1420|playback_error|5.0.0.37|iPa1|8.1.1|0.0|WIFI|BUP|9863786|6799798072||ta99|841759656|Be000|1601|Video|22
594|app_launch_time|5.0.0|S7M|4.3|0.0|CLAR|AUTO_ATED|5010226|-110.673567|6167612959947-023Xca|||111|N/A||11
421|video_load_time|5.0.0.37|iP5|8.1.2|0.0|WIFI|BUP|528950658168|06.613394189||cpcpb15|84551050401|1601|N/A||182

所以我创建了以下代码：

awk -F \| '{ duration[$15] = $NF; ++counter[$15]; duration2[$2] = $NF; } END {for(d in duration); for(e in duration2) {print e,d, counter[d]} }'errors.out

但它似乎没有返回好的数字:( 任何人都知道如何处理这个问题？谢谢大家！

Answer 1

如果你想计算不同的（$ 2，$ 15）对，这就是我理解你对问题的描述，你可以这样做，（虽然你的示例输出似乎与数据不匹配））。

$ cat vid.txt
14201|video_load_time|5.7|i3|8.3|0.0|FI|GT|52.1147619|9829672936714|777|ghouso|444|6789|111|4464|7149       
4399784|playback_error|8.0|W8|33.2|0.0|FI|TED|468|071|078410X_.ca||2314831||3805|152rorDescript|0      
762|playback_error|70|ALFiee|4.2.2|0.0|IFI|AUED|4325|795|||81321761|3805|05|1529|
634|app_launch_time|5.0.0|SGHI317M|4.1.2|0.0|CELAR|AUTO_ATED|4588|64|180||||3805|yes|0     
1418|video_load_time|5.0.0.37|iP1|7.0.6|0.0|IN_HE_WIFI|AUTHEATED|45.47099941453838|477109099|||8455700500884828|111|N/A|77|9398     
1420|playback_error|5.0.0.37|iPa1|8.1.1|0.0|WIFI|BUP|9863786|6799798072||ta99|841759656|Be000|1601|Video|22
594|app_launch_time|5.0.0|S7M|4.3|0.0|CLAR|AUTO_ATED|5010226|-110.673567|6167612959947-023Xca|||111|N/A||11
421|video_load_time|5.0.0.37|iP5|8.1.2|0.0|WIFI|BUP|528950658168|06.613394189||cpcpb15|84551050401|1601|N/A||182

$ cat vid.awk
{ ++count[$2,$15] }

END { for (i in count) {
        split(i, parts, SUBSEP);
        print parts[1] ", errcode: " parts[2] ", number of occurrences: " count[i] 
      }
}

$ awk -F\| -f vid.awk vid.txt
app_launch_time, errcode: N/A, number of occurrences: 1
playback_error, errcode: 1601, number of occurrences: 1
playback_error, errcode: 3805, number of occurrences: 1
video_load_time, errcode: 111, number of occurrences: 1
video_load_time, errcode: N/A, number of occurrences: 2
app_launch_time, errcode: 3805, number of occurrences: 1
playback_error, errcode: 05, number of occurrences: 1

注意虽然awk只支持一维数组，但是当你使用多个逗号分隔值作为数组索引时 awk通过连接将它们转换为单个索引值与内置变量SUBSEP一起作为分隔符。这个让你在某种程度上模仿多维数组。迭代时但是，通过for循环中的索引，由你来决定使用split将值分开。

awk命令使用for循环计算两个字段

1 个答案: