时间戳

首页 > 解决方案 > awk 根据单列的唯一值组合其他列的唯一值

csv - awk 根据单列的唯一值组合其他列的唯一值

问题描述

我的输入文件看起来像

Item1,200,a,four,five,six,seven,eight1,nine1
Item2,500,b,four,five,six,seven,eight2,nine2
Item3,900,c,four,five,six,seven,eight3,nine3
Item2,800,d,four,five,six,seven,eight4,nine4
Item1,,e,four,five,six,seven,eight5,nine5

基于第一列的唯一值，我想组合所有其他列的唯一值。到目前为止我尝试的是：

awk -F, '{
a[$1]=a[$1]?a[$1]"_"$2:$2;
b[$1]=b[$1]?b[$1]"_"$3:$3;
c[$1]=c[$1]?c[$1]"_"$4:$4;
d[$1]=d[$1]?d[$1]"_"$5:$5;
e[$1]=e[$1]?e[$1]"_"$6:$6;
f[$1]=f[$1]?f[$1]"_"$7:$7;
g[$1]=g[$1]?g[$1]"_"$8:$8;
h[$1]=h[$1]?h[$1]"_"$9:$9;
}END{for (i in a)print i, a[i], b[i], c[i], d[i], e[i], f[i], g[i], h[i];}' OFS=, input.txt

上面的输出是：

Item3,900,c,four,five,six,seven,eight3,nine3
Item1,200_,a_e,four_four,five_five,six_six,seven_seven,eight1_eight5,nine1_nine5
Item2,500_800,b_d,four_four,five_five,six_six,seven_seven,eight2_eight4,nine2_nine4

但我期待的是：

Item3,900,c,four,five,six,seven,eight3,nine3
Item1,200,a_e,four,five,six,seven,eight1_eight5,nine1_nine5
Item2,500_800,b_d,four,five,six,seven,eight2_eight4,nine2_nine4

我正在寻求一些帮助：

  1. 如何在组合值时只取唯一值？
  2. 每当存在空白值时，在组合时不应在末尾附加分隔符（在我上面的例子中为下划线）？
  3. 如何根据第 1 列的值对输出进行排序？

非常感谢你的帮助。

标签： csvawk

解决方案

任何awk加号sort：

$ cat tst.awk
BEGIN { FS=OFS="," }
{
    key = $1
    keys[key]
    for (i=2; i<=NF; i++) {
        if ( ($i ~ /[^[:space:]]/) && (!seen[key,i,$i]++) ) {
            idx = key FS i
            vals[idx] = (idx in vals ? vals[idx] "_" : "") $i
        }
    }
}
END {
    for (key in keys) {
        printf "%s%s", key, OFS
        for (i=2; i<=NF; i++) {
            idx = key FS i
            printf "%s%s", vals[idx], (i<NF ? OFS : ORS)
        }
    }
}

.

$ awk -f tst.awk file | sort -t, -k1,1
Item1,200,a_e,four,five,six,seven,eight1_eight5,nine1_nine5
Item2,500_800,b_d,four,five,six,seven,eight2_eight4,nine2_nine4
Item3,900,c,four,five,six,seven,eight3,nine3

或者使用 GNUawk来处理数组（参见https://www.gnu.org/software/gawk/manual/gawk.html#Multidimensional和https://www.gnu.org/software/gawk/manual/gawk.html #Arrays-of-Arrays了解两者之间的区别）和sorted_in（参见https://www.gnu.org/software/gawk/manual/gawk.html#Controlling-Array-Traversal和https://www.gnu。 org/software/gawk/manual/gawk.html#Controlling-Scanning）：

$ cat tst.awk
BEGIN { FS=OFS="," }
{
    for ( i=2; i<=NF; i++ ) {
        vals[$1][i][$i]
    }
}
END {
    PROCINFO["sorted_in"] = "@ind_str_asc"
    for ( key in vals ) {
        printf "%s%s", key, OFS
        for ( i=2; i<=NF; i++ ) {
            sep = ""
            for ( val in vals[key][i] ) {
                if ( val ~ /[^[:space:]]/ ) {
                    printf "%s%s", sep, val
                    sep = "_"
                }
            }
            printf "%s", (i<NF ? OFS : ORS)
        }
    }
}

.

$ awk -f tst.awk file
Item1,200,a_e,four,five,six,seven,eight1_eight5,nine1_nine5
Item2,500_800,b_d,four,five,six,seven,eight2_eight4,nine2_nine4
Item3,900,c,four,five,six,seven,eight3,nine3

推荐阅读