使用 awk 基于行值动态将行转置为列

使用 awk 基于行值动态将行转置为列

我的文件包含重复行中的 data_0 到 data_4 的数据。我需要将其转换为受尊重的数据集下的列值。如果早期类别的数据丢失,是否有任何方法可以放置空白/空值。例如

TimeStamp,Block,No_of_requests
04:19:12,data_0,4
04:19:12,data_1,6
04:19:12,date_2,8
04:19:12,date_3,10
04:19:12,data_4,12
04:19:14,data_0,5
04:19:14,data_1,6
04:19:14,date_3,7
04:19:14,data_4,8

预期输出是

TimeStamp,data_0,data_1,data_2,data_3,data_4
04:19:12,4,6,8,10,12
04:19:14,5,6,,7,8

等等。它应该放置空数据,以防相应 data_x 的值不可用。

答案1

GNUawk解决方案:

awk 'BEGIN{ 
         FS = OFS = ",";
         PROCINFO["sorted_in"] = "@ind_num_asc";
         print "TimeStamp,data_0,data_1,data_2,data_3,data_4" 
     }
     NR > 1{ a[$1][substr($2, 6) + 1] = $3 }
     END{ 
         for (i in a) { 
             printf "%s,", i;
             for (j=0; j<=4; j++) printf "%s%s", a[i][j+1], (j == 4? ORS:OFS) 
         }
     }' file

输出:

TimeStamp,data_0,data_1,data_2,data_3,data_4
04:19:12,4,6,8,10,12
04:19:14,5,6,,7,8

答案2

与 Roman 的答案类似,但对文件内容的硬编码较少

awk -F, -v OFS=, '
    NR > 1 {data[$1][$2] = $3; blocks[$2]}
    END {
        PROCINFO["sorted_in"] = "@ind_str_asc"

        # header
        printf "TimeStamp"
        for (block in blocks) {
            printf "%s%s", OFS, block
        }
        print ""

        # data
        for (ts in data) {
            printf "%s", ts
            for (block in blocks) {
                printf "%s%s", OFS, data[ts][block]
            }
            print ""
        }
    }
' file
TimeStamp,data_0,data_1,data_4,date_2,date_3
04:19:12,4,6,12,8,10
04:19:14,5,6,8,,7

请注意,您的示例数据使用“数据“ 和 ”日期“ 两个都。

相关内容