Merging files in a single one

Serg · #1 (**permalink**) 05-24-04, 13:18

Hi all.

Hope someone can help on this one.

I have several files, each of them containing a single colum
of numbers. How can I create a single file containing all these
different columns ?

file1 :
1
3
4
5

file2 :
-3
0
10
4

file3 :
3
6
3
7

after merging files :
1 -3 3
3 0 6
4 10 3
5 4 7

I appreciate any help ! (the real problem contains 400 files
with 200 rows each...)

Thanks,

Serg

vgersh99 · #2 (**permalink**) 05-24-04, 14:17

man paste
man pr

Serg · #3 (**permalink**) 05-24-04, 15:01

thanks for the help !

supinformix · #4 (**permalink**) 05-25-04, 03:48

open different file descriptors :
exec 3 <f1
exec 4 <f2
exec 5 <f3
while read F1
do
read F2 < &4
read F3 < &5
echo "$F1 \n $F2 \n $F3" >>output
done < &3

problem is what if file f2 has less rows than f1.

Damian Ibbotson · #5 (**permalink**) 05-25-04, 07:13

Quote:

Originally Posted by supinformix

open different file descriptors :
exec 3 <f1
exec 4 <f2
exec 5 <f3
while read F1
do
read F2 < &4
read F3 < &5
echo "$F1 \n $F2 \n $F3" >>output
done < &3

problem is what if file f2 has less rows than f1.

You can get round that something like this...

Code:

exec 3<file1 4<file2 5<file3
while read file1Var <&3 && isVarRead=true;
      read file2Var <&4 && isVarRead=true;
      read file3Var <&5 && isVarRead=true;
      ${isVarRead-false}
do
  print ${file1Var} ${file2Var} ${file3Var}
  unset isVarRead
done

Admittedly, this does the same as 'paste' but it allows you to tinker with the 'print' line to output exactly what you would like.

You could also use awk, e.g. while (getline < "file1" > 0) { ...

Damian

Serg · #6 (**permalink**) 05-25-04, 08:43

Thanks guys.

I was looking for some awk solution but your
suggestions were also great.

Serg

skd · #7 (**permalink**) 05-28-04, 15:40

Hi Damian,
Can you walk through me the above code?
alsosintereseted to see awk solution for my post 'to find a match between two files'
dated May 28.

Thanks,

vgersh99 · #8 (**permalink**) 05-28-04, 16:51

Assuming all your input files have the same numbe rof records/lines:

nawk -f sdk.awk file*

Code:

{
  arr[FNR] = (NR == FNR) ? $0 : arr[FNR] OFS $0
}

END {
  for(i=1; i <= FNR; i++)
    print arr[i];
}

Internet Services

Web Development

Digital Marketing

Consumer Tech