Question

将行转换为多列的Shell脚本

输入CSV文件：

Driver Id,Driver Name,Measure Names,Measure Values
123,XYZ,Total Offers,10
123,XYZ,Driver Reject,0
123,XYZ,Driver Accept ,4
123,XYZ,Expired Offers,3
123,XYZ,Total Bookings,6
123,XYZ,Rider Cancels,2
123,XYZ,Driver Cancels,0
123,XYZ,Rider No-Show,0
123,XYZ,Completed Rides,4
124,PQR,Total Offers,2
124,PQR,Driver Reject,0
124,PQR,Driver Accept ,1
124,PQR,Expired Offers,1
124,PQR,Total Bookings,1
124,PQR,Rider Cancels,0
124,PQR,Driver Cancels,0
124,PQR,Rider No-Show,0
124,PQR,Completed Rides,1

需要输出：

驱动程序ID，驱动程序名称，总优惠，驱动程序拒绝，驱动程序接受，过期优惠，总预订，车手取消，驱动程序取消，车手未显示，已完成游乐设施

123,XYZ,10,0,4,3,6,2,0,0,4
124,PQR,2,0,1,1,1,0,0,0,1

我尝试使用awk，但结果不正确。

awk -F\, '
    BEGIN{
        P["Total Offers"]="%s;%s;%s;;;;;;;;;\n"
       P["Driver Reject"]="%s;%s;;%s;;;;;;;;\n"
       P["Driver Accept"]="%s;%s;;;%s;;;;;;;\n"
      P["Expired Offers"]="%s;%s;;;;%s;;;;;;\n"
      P["Total Bookings"]="%s;%s;;;;;%s;;;;;\n"     
       P["Rider Cancels"]="%s;%s;;;;;;%s;;;;\n"
      P["Driver Cancels"]="%s;%s;;;;;;;%s;;;\n"     
       P["Rider No-Show"]="%s;%s;;;;;;;;%s;;\n"     
     P["Completed Rides"]="%s;%s;;;;;;;;;%s;\n" 
        }                         
    FNR==1{
        print "Driver Id,Driver Name,Total Offers,Driver Reject,Driver Accept,Expired Offers,Total Bookings,Rider Cancels,Driver Cancels,Rider No-Show,Completed Rides"
        next
        }
    {
        printf(P[$3],$1,$2,$4)
        }
    ' sample1.csv

有人可以帮我或者告诉我任何其他方法来实现这个。在此先感谢

Answer 1

考虑到您的Input_file与显示的示例相同，如果您不关心输出序列应该作为输入，那么以下内容可以帮助您。

awk -F, 'FNR>1{a[$1,$2]=a[$1,$2]?a[$1,$2] FS $NF:$NF} END{for(i in a){print i FS a[i]}}' SUBSEP=","   Input_file

Answer 2

如果有任何

，下面会考虑输出顺序和缺失值

awk '
     BEGIN{
       FS=OFS=SUBSEP=","; 
     }
     FNR==1{
        printf("%s%s%s",$1,OFS,$2);
        next
     }
     {
      if(!(($1,$2) in tmp)){
        usr[++u] = $1 OFS $2
        tmp[$1,$2]
      }
      if(!($3 in tmp)){
        names[++n] = $3;
        tmp[$3]
        printf("%s%s",OFS,$3)
      }
       arr[$1,$2,$3] = $4
     }
     END{
       print ""
       for(u=1; u in usr; u++){
           printf("%s", usr[u]);
           for(n=1; n in names; n++){
               indexkey = usr[u] SUBSEP names[n]
               printf("%s%s",OFS, (indexkey in arr) ? arr[indexkey]:"")
           }
           print ""
       }
     }
    ' infile

<强>解释

FS=OFS=SUBSEP=","; - 设置字段分隔符，输出字段分隔符和内置变量subsep为逗号，在当前程序中至少至少OFS和SUBSEP应该相同，因为我用它来访问数组{{1}如果您有任何其他输入字段分隔符（比如管道），那么请indexkey = usr[u] SUBSEP names[n]
FS="|"; OFS=SUBSEP="," 如果是第一行，则打印前2个字段并转到下一行
FNR==1{ printf("%s%s%s",$1,OFS,$2); next }由于您需要有序输出，因此在此程序中使用连续（按顺序）数组（if(!(($1,$2) in tmp)){ usr[++u] = $1 OFS $2 tmp[$1,$2] }）。 usr是数组，其中索引为tmp和$1，$2为数组，其中index为变量usr，值为u且$1，$2如果之前不存在则负责。
if(!(($1,$2) in tmp))与上述类似，if(!($3 in tmp)){ names[++n] = $3; tmp[$3] printf("%s%s",OFS,$3) }数组是连续的，值为names
$3数组arr[$1,$2,$3] = $4键为3个字段arr，值为$1,$2,$3
最后在$4块循环中通过END和usr数组，构建indexkey和print数组值，如果数组names中存在indexkey

输入：

arr

输出：

$ cat infile
Driver Id,Driver Name,Measure Names,Measure Values
123,XYZ,Total Offers,10
123,XYZ,Driver Reject,0
123,XYZ,Driver Accept ,4
123,XYZ,Expired Offers,3
123,XYZ,Total Bookings,6
123,XYZ,Rider Cancels,2
123,XYZ,Driver Cancels,0
123,XYZ,Rider No-Show,0
123,XYZ,Completed Rides,4
124,PQR,Total Offers,2
124,PQR,Driver Reject,0
124,PQR,Driver Accept ,1
124,PQR,Expired Offers,1
124,PQR,Total Bookings,1
124,PQR,Rider Cancels,0
124,PQR,Driver Cancels,0
124,PQR,Rider No-Show,0
124,PQR,Completed Rides,1

Answer 3

如果行未在必填字段中排序，则必须使用关联数组。

$ awk -F, -v cols='Total Offers,Driver Reject,Driver Accept ,Expired Offers,Total Bookings,Rider Cancels,Driver Cancels,Rider No-Show,Completed Rides' '
   BEGIN {n=split(cols,f)} 
   NR>1  {k=$1 FS $2; keys[k]; a[k,$3]=$4} 
   END   {for(k in keys) 
             {printf "%s", k; 
              for(i=1;i<=n;i++) printf "%s%d", FS,+a[k,f[i]]; 
              print ""}}' file

124,PQR,2,0,1,1,1,0,0,0,1
123,XYZ,10,0,4,3,6,2,0,0,4

如果缺少任何度量行，则需要注意

PS。请注意＆＃34;驱动程序接受＆＃34;有一个尾随空间，我保留。

LINUX Shell脚本将行转换为多列

3 个答案: