OK. You lucked out... The BRE sampletest*\\_[0-9] tells grep to match and print lines that contain the string sampletes followed by zero or more occurrences of t followed by whatever unspecified characters are matched by the character sequence \_ on the regular expression matching engine used on your operating system followed by a decimal digit. It looks like your operating system's RE matching engine chooses to use that sequence to match an underscore character,

To match the filenames you want to process, the following BRE would work more reliably:

Code:

grep 'sampletest_[0-9][0-9]*.xml'

If you want to exclude matching filenames like sampletest_112.xml.0001, you could force the xml to only be matched at the end of a filename with:

Code:

grep 'sampletest_[0-9][0-9]*.xml$'

Now that we have gotten past that... What statement in your script is failing to do what you want it to do? What are the arguments being passed to that command according to the trace output you're seeing? What arguments did you hope would be passed to that command instead of the arguments that are actually being passed to that command?