Shell Programming and Scripting

Hi,

I am trying to paste together two files which are like this-

Code:

1 
2
4

and

Code:

2
3
4

The paste command would work by default for this case, but when there are cases where the number of entries are different in each file. for eg:

Code:

1 
3

and

Code:

1
3
4

I want to make it such that the odd entry is removed and put these two together.

Is there a way to do it using sed or awk

Thanks

Code:

awk '{ A[$1]++ } END { for(X in A) if(A[X]==2) print X }' file1 file2

Thanks a lot
If you have the time could you give hints on how this works, I find it a bit confusing, I am still getting in grips with awk n sed.

Thanks again.

---------- Post updated at 12:34 AM ---------- Previous update was at 12:29 AM ----------

Hi,
for the case of

Code:

1
2
4
5

and

Code:

2
4
4

I get just 2 as the output.

Just tried this at random.

THanks a lot!

That's because 4 occurs three times, not two. If that's okay, try A[X]>=2 instead of A[X]==2.

It works by counting how many times each letter does, and storing it in an associative array. Each time a number is found, it's incremented. Then it goes through the whole array and prints ones that have been found the appropriate number of times.

Hi,
Thanks again..
I think I should've posted the intended output to make it clear. For the previous input, I expect something like

Code:

1 2
2 4
4 4

In that case, just

Code:

paste file1 file2 | awk 'NF>2'

Here you go: (use nawk with Solaris)

Code:

awk '
NR == FNR { first[FNR]=$0 ; c1++}
NR != FNR { second[FNR]=$0 ; c2++}
END {
  count=c1>c2?c2:c1;
  for(i=1;i<=count;i++)
  {
    printf("%s\t%s\n",first[i],second[i]);
  }
}
' file1 file2