Manipulating field differences using ksh

Shell Programming and Scripting
1

I have a list of files that should contain the following

Im trying to find the items of interest that are missing from each file and create a csv.

cat *.txt | while read file
do
grep 3500[1-6] file | tr '\012' ','
done

My problem is this possible output

one.txt 35001,35002,35005,35006
two.txt 35002,35003,35004,35005,35006
three.txt 35001,35002, 35003,35004,35005,35006
four.txt 35001,35006

What Im after is
one.txt 35001,35002,,,35005,35006
two.txt ,35002,35003,35004,35005,35006
three.txt 35001,35002, 35003,35004,35005,35006
four.txt 35001,,,,,35006

This way, when I move it into the Microsoft environment the colums will be aligned correctly.

2

try awk solution

$ cat test_file
one.txt 35001,35002,35005,35006
two.txt 35002,35003,35004,35005,35006
three.txt 35001,35002, 35003,35004,35005,35006
four.txt 35001,35006

$  awk -F'[ ,]' '{printf $1 " ";for(i=$2;i<=$NF;i++)printf ($0~i)?(i<$NF)? i OFS:i:OFS;printf RS}' OFS="," test_file

Resulting

one.txt 35001,35002,,,35005,35006
two.txt 35002,35003,35004,35005,35006
three.txt 35001,35002,35003,35004,35005,35006
four.txt 35001,,,,,35006
3

Compare with field.

awk -F"," '{for(i=$1;i<=$NF;i++) {++j;if ( $j != i) { printf FS ;--j } else { printf (j==NF)?i: i FS}} printf "\n";j=0}'  filename