🚀 go-pugleaf

GNU cp has a lot of nice features (particular the -a and -update options).

I'm somewhat comfused about copying trees of directories to back up.

It seems when I execute:
cp -a srcdir targetdir

if targetdir doesn't exist, it makes it (i.e.
cp -a test1 test2)
After test2 exists, and I execute the command again, it puts
test1 under test2 (i.e. test2/test1).

In addition, if I do:
cp -a /homes/leisner/tmp/test1 test2
it works as
cp -a test1 test2 (when I'm in /home/leisner/tmp).

This is confusing and needs to be documented.

Also, it looks like it makes sense to have a translation table of uid and gid
for cp.  This is so when I execute cp as root, and do an update, I can copy
distributions and map UID/GID from one domain to another domain.  Sound useful?
Looks like I'm going to do it.

I did a study of copying between machines, and using NFS with cp -a -update
wins...seems to be the simplest I've seen.


Marty    leisner.henr801c@xerox.com 
          Member of the League for Programming Freedom 
 The Feynman Problem Solving Algorithm:
 1) Write down the problem
 2) Think very hard
 3) Write down the answer     -- Murray Gell-mann