Question

我是Awk的新手，我有以下查询。

input.csv

"11111","TRUE","aa"
"456789","TRUE","aa;bb;cc"
"2345","TRUE","bb"
"434566","","cc"

我正在尝试创建一个awk命令，它应该给我以下输出：

output.csv

"11111","TRUE","aa,,"
"456789","TRUE","aa,bb,cc"
"2345","TRUE",",bb,"
"434566","",",,cc"

我需要每次打印前两个参数，但第三个参数是我需要检查的。所以第三个属性可以包含任何aa，bb，cc或它们的组合，如aa; bb或aa; cc或aa; bb; cc或这些都不是，所以我需要确保其中哪些存在并打印这些分离用逗号如果aa; bb在输入文件中我需要aa，bb，如果没有那么我需要，（两个逗号）。

我在每个中使用正则表达式如果检查aa，bb，cc是这些中的任何一个，如果它不存在那么它将附加在varible值中，（逗号）将附加在值中。

我创建了以下命令。

awk 'BEGIN{FS=",";OFS=","}  { if( $3 ~ /aa/ ) { value="aa" }  else { value="," }; if( $3 ~ /bb/ ) { value="$value,bb" } else { value="$value," }; 
if( $3 ~ /cc/ ) { value="$value,cc" } else { value="$value,"};  print $1 , $2 , $value}' input.csv > output.csv

但是它给了我以下输出。

"11111","TRUE","11111","TRUE","aa"
"456789","TRUE","456789","TRUE","aa;bb;cc"
"2345","TRUE","2345","TRUE","bb"
"434566","","434566","","cc"

我不确定为什么打印前两个属性两次然后打印第三个值。我能够在shell脚本中完成它，但我需要使用awk。

Answer 1

$ cat tst.awk
BEGIN {
    FS=OFS="\""
    split("aa,bb,cc",dflts,/,/)
}
{
    delete vals
    for (i in dflts) {
        vals[i] = ($(NF-1) ~ dflts[i] ? dflts[i] : "")
    }
    $(NF-1) = vals[1] "," vals[2] "," vals[3]
    print
}

$ awk -f tst.awk file
"11111","TRUE","aa,,"
"456789","TRUE","aa,bb,cc"
"2345","TRUE",",bb,"
"434566","",",,cc"

阅读Arnold Robbins撰写的有效Awk编程，第4版。

如何在Awk中使用多个if条件？

1 个答案: