Is it possible to create a script that will create the following structure after running it:

/test/1/1 -> A
/test/1/2 -> B
/test/1/3 -> C
/test/1/4 -> D
/test/1/5 -> E
/test/1/6 -> F
/test/1/7 -> G
/test/1/8 -> H
/test/1/9 -> I
/test/1/10 -> J 
/test/1/11 -> K
/test/1/12 -> L
/test/1/13 -> M
/test/1/14 -> N
/test/1/15 -> O
/test/2/1 -> P
/test/2/2 -> Q
/test/2/3 -> R
/test/2/4 -> S
/test/2/5 -> T
/test/2/6 -> U
/test/2/7 -> V
/test/2/8 -> X
/test/2/9 -> Y
/test/2/10 -> Z
/test/2/11 -> A
/test/2/12 -> B
/test/2/13 -> C
/test/2/14 -> D
/test/2/15 -> E
/test/3/1 -> F
/test/3/2 -> G
/test/3/3 -> H
/test/10/15 -> ???

As you can see, in this example, the directory "test" is the parent directory and the numbered ones are subdirectories. The letters are files, not folders. Each time the alphabet ends, it will start from the beginning in the next directory.

How can I create a script that will make this easier rather than just creating them manually?

  • How many alphabets? Feb 1 '16 at 12:36
  • Well, I need it until it reaches /10/15. It doesn't matter.. Feb 1 '16 at 12:37
  • What is the point of this? The files are supposed to be symlinks to files named for upoercase letters in the same directory?
    – mikeserv
    Feb 1 '16 at 19:09

With zsh:

dirs=(test/{1..10}/{1..15}) n=0
mkdir -p $dirs &&
  for d ($dirs) {:>$d/$(([##36]n%26+10)); ((n++))}

The trick above being to use base36 numbers where digits 10 to 35 are expressed with A.. Z. Would be a bit more legible and generalisable with:

dirs=(test/{1..10}/{1..15}) n=0 l=({A..Z})
mkdir -p $dirs && for d ($dirs) : > $d/$l[1 + n++ % $#l]
echo /test/{1..10}/{1..15} " " |
  ( chars=({A..Z}); i=0;
   while read -d " " dir; do
       test -z "$dir" && break;
       echo mkdir -p "$dir";
       echo touch "${dir}/${chars[i]}";
   done )

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.