If you like to test your script output, which is a noble thing, then I would code the reference XML either by hand or take a more standard tool for it then Excel.

You can compare both XML like you said. I would use XML::Simple to read both XML files into separate hashes which can that be compared. Test::More gives you the function is_deeply to do this, but I'm not sure if you can easily use it in a non-Test Perl script. On the other hand your script is basically one.

Do you mean to convert the xml file which is generated as a output of perl script to excel. Then compare both the spreadsheets i.e. the spreadsheet which I created manually with the one converted from xml. will this be more convinient than comparing 2 xmls?

No, comparing the two XML files would be better. My point was that you take the XML generated by Excel as reference, i.e. you assume that Excel will produce correct, valid XML and differences to the Perl XML would mean that the Perl script made an error. I personally trust the Perl XML generator more than Excel. So I would take the Perl XML output as reference if Excel is generating correct XML.

But don't start a Linux/Windows or Perl/MS Office flame war here ... you can use the modules I mentioned in my last post to compare the two XMLs, you just have to critical interpret the results.

It's a bit of a weird thing you're doing. Why create the control data in excel instead of in a more programmer friendly format?

One way to do what you want would be to use DBD::AnyData to read both the XML and the excel file (exported as CSV, maybe using Spreadsheet::ParseExcel), and compare them.

Compare the data, not the XML! You know what differences are meaningful, not some generic XML comparison tool/function. For example are the following two snippets equivalent?

<foo>
 <bar>1</bar>
 <baz>2</baz>
</foo>

<foo>
 <baz>2</baz>
 <bar>1</bar>
</foo>
[download]

Parse the XMLs you need to verify, parse the Excel file (save is as CSV, that's gonna be easier to work with) and compare the data that should be the same.