Shell Programming and Scripting

Hey ppl
I have two columns with random values. i need to insert the 1st row of the first column with the highest number of the two rows in the first column and vice versa. some thing like this. I'm sorry If my question is unclear...

input
col1 col2
12...11
11...14
34...45
14...67
31...21

output
col1 col2
12...11
14...11
45...34
67...14
31...21

Not able to understand..make it clear plz....?

Thanks
Sha

its about the two values of two columns
I need to make the first column values will be higher compared to the subsequent column 2 values

11 ..12 to 12.. 11

use below one ...hope this should work..

sed 's/\.\.\./,/g' inputfile |awk -F"," '{if($2 > $1) {print $2,$1} else {print $1,$2}}'

output:
12 11
14 11
45 34
67 14
31 21

if you want the same format output then use below one:

sed 's/\.\.\./,/g' out.lst|awk -F"," 'OFS="..." {if($2 > $1) {print $2,$1} else {print $1,$2}}'

output:
12...11
14...11
45...34
67...14
31...21


Thanks
Sha

sed 's/\.\.\./,/g'

wht this mean

In your input file ..you have used delimeter..something like this "..."
so before i proceed with comparing column to make it easy myself...i changed delimeter from "..." to ","

Thanks
Sha

actually I have tab in between two file and I need result with tab

thanx