awk 根据第二个文件的标题从文件中提取列 3

Question

我正在尝试从有关此主题的先前线程中运行一些修改过的代码。 我有一个文件 data.txt，其中第一行是标题。我想创建一个新文件，其中仅包含与第二个文件（list.txt）中的条目匹配的那些列。

数据.txt

1,2,3,4,5,6,7,8,9,10
1.000000,0,0,0,0,0,0,0,0,0
0,1.000000,0.031250,0,0,0,0.031250,0,0,0
0,0.031250,1.000000,0,0,0,0.062500,0,0,0
0,0,0,1.000000,0,0,0,0,0,0
0,0,0,0,1.000000,0,0,0,0,0
0,0,0,0,0,1.000000,0,0.062500,0,0
0,0.031250,0.062500,0,0,0,1.000000,0,0,0
0,0,0,0,0,0.062500,0,1.000000,0,0
0,0,0,0,0,0,0,0,1.000000,0

list.txt

3
5
7
9

想要的输出是

3,5,7,9
0,0,0,0
0.031250,0,0.031250,0
1.000000,0,0.062500,0
0,0,0,0
0,1.000000,0,0
0,0,0,0
0,1.000000,0
0,0,0,0
0,0,0,1.000000

我已经使用了下面的代码

echo "${DATAFILE:-data.txt}"
echo "${COLUMNFILE:-list.txt}"

awk {
     j=1
     while ((getline < COLUMNFILE) > 0) {
        col[j++] = $1
     }
     n=j-1;
     close(COLUMNFILE)
     for (i=1; i<=n; i++) s[col[i]]=i
   }
   NR==1 {
     for (f=1; f<=NF; f++)
       if ($f in s) c[s[$f]]=f
     next
   }
   { sep=","
     for (f=1; f<=n; f++) {
       printf("%c%s",sep,$c[f])
       sep=FS
     }
     print ""
 } 
 DATAFILE

我得到下面的结果，它复制了 data.txt 中的行而不做任何选择。 list.txt 中的条目打印在文件末尾

1,2,3,4,5,6,7,8,9,10
1,2,3,4,5,6,7,8,9,10
1.000000,0,0,0,0,0,0,0,0,0
1.000000,0,0,0,0,0,0,0,0,0

0,1.000000,0.031250,0,0,0,0.031250,0,0,0
0,1.000000,0.031250,0,0,0,0.031250,0,0,0

0,0.031250,1.000000,0,0,0,0.062500,0,0,0
0,0.031250,1.000000,0,0,0,0.062500,0,0,0

0,0,0,1.000000,0,0,0,0,0,0
0,0,0,1.000000,0,0,0,0,0,0

0,0,0,0,1.000000,0,0,0,0,0
0,0,0,0,1.000000,0,0,0,0,0

0,0,0,0,0,1.000000,0,0.062500,0,0
0,0,0,0,0,1.000000,0,0.062500,0,0

0,0.031250,0.062500,0,0,0,1.000000,0,0,0
0,0.031250,0.062500,0,0,0,1.000000,0,0,0

0,0,0,0,0,0.062500,0,1.000000,0,0
0,0,0,0,0,0.062500,0,1.000000,0,0

0,0,0,0,0,0,0,0,1.000000,0
0,0,0,0,0,0,0,0,1.000000,0

3
3

5
5

7
7

9
9

非常感谢任何帮助。

Answer 1

$ awk '
    BEGIN { FS=OFS="," }
    NR==FNR { f[++nf]=$0; next }
    { for (i=1; i<=nf; i++) printf "%s%s", $(f[i]), (i<nf?OFS:ORS) }
' list.txt data.txt
3,5,7,9
0,0,0,0
0.031250,0,0.031250,0
1.000000,0,0.062500,0
0,0,0,0
0,1.000000,0,0
0,0,0,0
0.062500,0,1.000000,0
0,0,0,0
0,0,0,1.000000

Answer 2

您可以将您的位置文件和数据文件都传递到 awk 并在内部执行您的逻辑：

 awk -F"," 'FILENAME=="list.txt"{a[NR]=$1}FILENAME=="data.txt"{for(i=1; i<=length(a); i++){printf (i==length(a)?"%s\n":"%s,"),$a[i]}}' list.txt data.txt

我们在这里：

1. 使用逗号分隔符分割传入文件 (-F",")
2. 如果 FILENAME awk 变量是“list.txt” (FILENAME=="list.txt")
3. ++然后将该行中的值添加到以行号为索引的数组中（a[NR]=$1）
4. 如果 FILENAME awk 变量是“data.txt”(FILENAME=="data.txt")
5. ++然后循环遍历数组for(i=1; i<=length(a); i++)中的每个元素
6. ++++并打印出该位置的项目值（$a[i]）。如果该位置是找到的最后一个位置 (i==length(a))，则使用换行符 ("%s\n") 将其打印出来，否则使用逗号 ($a[i]) 将其打印出来。

另一种选择是通过 -v（变量）标志传递您的职位，但这并不能很好地考虑可变数量的职位：

awk -F"," -v f1=$(awk 'NR==1' list.txt) -v f2=$(awk 'NR==2' list.txt) -v f3=$(awk 'NR==3' list.txt) -v f4=$(awk 'NR==4' list.txt) '{print $f1, $f2, $f3, $f4}' data.txt

Answer 3

awk解决方案：

awk -F, 'function pr(a){ r=""; for(i=1;i<=NF;i++) if(i in a) r=(r!="")? r","$i:$i; print r }
         NR==FNR{ a[$0]; next }{ pr(a) }' list.txt data.txt

输出：

3,5,7,9
0,0,0,0
0.031250,0,0.031250,0
1.000000,0,0.062500,0
0,0,0,0
0,1.000000,0,0
0,0,0,0
0.062500,0,1.000000,0
0,0,0,0
0,0,0,1.000000

Answer 4

比较和对比的非awk 解决方案...

$ join -t, <(sort list) <(<file tr ',' '\n' | pr -10ts, | sort) | 
  sort -n | 
  tr ',' '\n' | 
  pr -4ts,


3,5,7,9
0,0,0,0
0.031250,0,0.031250,0
1.000000,0,0.062500,0
0,0,0,0
0,1.000000,0,0
0,0,0,0
0.062500,0,1.000000,0
0,0,0,0
0,0,0,1.000000

您需要魔术数字10 和4，它们是原始文件和提取文件的列号（这些也可以自动化）。将数字排序转换为字典顺序并返回所需的多种排序（join 需要）。

算法本质是transpose-join-transpose

Answer 5

这是一个将list.txt 处理为字段列表并使用该列表调用另一个awk 来处理data.txt 的awk：

$ awk '
BEGIN { FS=OFS="," }          # set the delimiters for the list file
NR==FNR {                     # process the list file
    p=p (p==""?"":OFS) "$" $1 # make a field list ($3,$5,$7,$9)
    next
}
{                             # process the data or call the processor
    RS=""                     # for getline to return multilined output
    cmd="awk \047BEGIN{FS=OFS=\",\"}{print "p"}\047 " FILENAME   # build awk call
    cmd | getline res         # actual awk call and output to res
    print res                 # output res
    exit                      # exit after first record
}
' list data
3,5,7,9
0,0,0,0
0.031250,0,0.031250,0
1.000000,0,0.062500,0
0,0,0,0
0,1.000000,0,0
0,0,0,0
0.062500,0,1.000000,0
0,0,0,0
0,0,0,1.000000

Answer 6

awk 解决方案，考虑到列名可以是任何东西，而不仅仅是列的索引。

BEGIN { FS=OFS="," }
NR==FNR { l[$0]++; next }                      # save headers from list
FNR==1{ for (i=1; i<=NF; i++) 
            if ($i in l){ max=i; c[i]++ }}         # save column index in c;
                                                   # max index in max 
{ for(j=1; j<=NF; j++)                             # loop over column indices
      if(j in c)                                   # if index in c
          printf "%s%s", $j, (j==max ? ORS : OFS)  # print column
}

有输入：

$ cat list.txt
C
E
G
I

和

$ cat data.txt
A,B,C,D,E,F,G,H,I,J
1.000000,0,0,0,0,0,0,0,0,0
0,1.000000,0.031250,0,0,0,0.031250,0,0,0
0,0.031250,1.000000,0,0,0,0.062500,0,0,0
0,0,0,1.000000,0,0,0,0,0,0
0,0,0,0,1.000000,0,0,0,0,0
0,0,0,0,0,1.000000,0,0.062500,0,0
0,0.031250,0.062500,0,0,0,1.000000,0,0,0
0,0,0,0,0,0.062500,0,1.000000,0,0
0,0,0,0,0,0,0,0,1.000000,0

给出结果：

$ awk 'BEGIN {FS=OFS=","} NR==FNR{l[$0]++;next} FNR==1{ for (i=1; i<=NF; i++) if ($i in l){max=i; c[i]++}}{for (j=1;j<=NF;j++) if(j in c) printf "%s%s",$j,(j==max ?ORS:OFS) }' list.txt data.txt
C,E,G,I
0,0,0,0
0.031250,0,0.031250,0
1.000000,0,0.062500,0
0,0,0,0
0,1.000000,0,0
0,0,0,0
0.062500,0,1.000000,0
0,0,0,0
0,0,0,1.000000