shell - Transpose values from multiple files with awk

Question

I have 10 files. For example:

$ cat ifile01.txt 
1 0.22
2 0.01
4 0.32
5 0.10
. .

$ cat ifile02.txt 
1 0.23
2 0.11
3 0.12
4 0.20
. .

$ cat ifile03.txt 
1 0.32
2 0.64
3 0.12
5 0.90
. .

and so on for 10 files

Here 1st column is the serial number and 2nd column is their index

I would to rearrange them in the following way in one file:

outfile.txt
0.22  0.01     ?  0.32  0.10     (Transpose of the index from ifile01.txt with "?" for serial number 3, as the index for 3 is missing)
0.23  0.11  0.12  0.20     ?     (Transpose of the index from ifile02.txt with "?" for serial number 5, as the index for 5 is missing)
0.32  0.64  0.12     ?  0.90     (Transpose of the index from ifile03.txt with "?" for serial number 4, as the index for 4 is missing)

I was trying with the following way in fortran, but I am looking for a awk script.

   for each ifile.txt, I check the follwoing
      for i in {1..50};do
       if [ $i != $1 ]; then i="?"
    Then append print transpose of $2 for each ifile.txt in outfile.txt 

标签： shellawk

Answer 1

Could you please try following.

awk '
FNR==1{
  count++
}
{
  a[count,$1]=$2
  if(!c[$1]++){
    d[++occ]=$1
  }
}
END{
  asorti(d,e)
  for(i=1;i<=count;i++){
    for(k=1;k<=occ;k++){
       printf("%s ",a[i,e[k]]?a[i,e[k]]:"?")
    }
    print ""
  }
}
'  Input_file1   Input_file2  Input_file3 | column -t

Output will be as follows.

0.22  0.01  ?     0.32  0.10
0.23  0.11  0.12  0.20  ?
0.32  0.64  0.12  ?     0.90

Explanation: Adding explanation for above code.

awk '                                            ##Starting awk program from here.
FNR==1{                                          ##Checking condition if this is first line of Input_file.
  count++                                        ##increment variable count value with 1 here.
}
{
  a[count,$1]=$2                                 ##Creating an array named a with index count and $1 whose value is $2 of current line.
  if(!c[$1]++){                                  ##Checking condition if $1 is NOT present in array c then do following.
    d[++occ]=$1                                  ##Creating an array named d whose index is occ variable and value is $1 of current line.
  }                                              ##Closing BLOCK for if condition.
}                                                ##Closing main BLOCK.
END{                                             ##starting END block for this awk program here.
  asorti(d,e)                                    ##Using asorti to sort array d and creating array e with it(which has sorted values in it).
  for(i=1;i<=count;i++){                         ##Starting a for loop from i=1 to till value of count(number of files actually).
    for(k=1;k<=occ;k++){                         ##Starting a for loop from k=1 to till value of occ.
       printf("%s ",a[i,e[k]]?a[i,e[k]]:"?")     ##Printing value of array a whose index is variable i AND array e with index of k if its NOT NULL else print ? as per OP
    }                                            ##Closing BLOCK for, for Loop here.
    print ""                                     ##Printing NULL value to get a new line here.
  }                                              ##Closing BLOCK for outer for loop here.
}                                                ##Closing BLOCK for END block of this awk program here.
'  file1  file2 file3 | column -t                ##Mentioning Input_file names here and using colunm -t to put equal spacing in their output.

---

---

As per @jhnc's nice comments, adding his suggested tweaked solution here too.

awk '
FNR==1{
  f++
}
{
  a[f,$1]=$2
}
m<$1{
  m=$1
}
END{
  for(i=0;i++<f;){
    for(j=1;j<=m;j++){
      printf"%s%s",a[i,j]?a[i,j]:"?",j<m?OFS:ORS
    }
  }
}'  file1 file2 file3