Shell Programming and Scripting

I have a text file of this format...

55 55-45345.xml 20070615
55 55-87655.xml 20070613
34 34-56753.xml 20070614

The text file has values like a number,xml file name, and a date.The first column can have n number of duplicates.And no two dates are equal.Now I sorted out the file.So, it is sorted by first column wise.But itis not sorted by date-wise.

My actual problem is,

I need to write a perl script to find the duplicates in first column.If n number of duplicates are found then n number of XML files are also duplicates.So,on the basis of Date i need to sort it down and take down the least date with it's corresponding XML file and put in a text file.(so that the it is sorted on basis of date wise !) Then,i have to shift the least dated XML file(top most) to a particular directory if the wc(word count) of the n file is equal...if the wc is not equal, then i have to take the bottom most(highest dated XML file) to a folder.

Plz help me out in solving the problem. !

With this code, you can get the latest xml file list with unique ID by first column

Code:

sort -k1n -k3n urfile |awk '{a[$1]=$3; b[$1]=$0}END{for (i in a) print b[i]|"sort -n"}'

For your rest requests, such as shift, wc, you need explain more.