UNIX for Dummies Questions & Answers

I have a directory which contains files that are the some name with only one exception -- one of the files is a duplicate of its original and has "copy" at the end of the name, and they all end in .doc

What I want to do is delete only the originals and keep the copy, but I can't figure out how to do it. I've tried listing them, but I don't know how to restrict what I list/delete only the desired files using just the standard wildcard characters. Can anyone help me?

Code:

for file in $(ls *.doc |grep -v copy); do
echo rm $file
done

Verify your list is correct, then remove the echo...

If I understand correctly, you want to delete the original files if copies exist but leave them intact if no copy is there. Try

Code:

for FN in *copy.doc; do echo rm "${FN/copy}"; done
rm f1.doc
rm f2.doc

Remove echo if happy with the result. Should the be spaces etc. in the filenames, try

Code:

ls *copy.doc | { while read FN; do echo rm ${FN/copy}; done; }
rm f1.doc
rm f2.doc

I think previous solutions are not exact about where the copy is.
E.g.

Code:

FN=copy123_copy.doc
echo ${FN/copy}
123_copy.doc

The following is more exact and also works in all Posix shells:

Code:

for file in *copy.doc
do
   ofile=${file%copy.doc}.doc # strip copy.doc from the end and add .doc again
   echo rm "$ofile"
done

I'd guess the pattern *copy.doc will allow for only the desired files.

BTW, what is the star in your %* needed for?

You are right, the * shouldn't be there because % always strips from the end.
Removed in my previous post.