Need script for changing DIRECTORY names only to upper case

PaxRomana · 05-07-2007, 07:53 AM

EDIT: Good start - I meant to say in the title 'change to LOWER case' not upper duhh.

Hello all

I have just installed Ubuntu and am moving a lot of files from my Windows machine.

Now, as I have many, many directories with upper case characters in them, I'd like to change them so they are only lower case. However, I only want the DIRECTORIES changed this way, NOT the filenames (got my reasons).

I've made a script (or I got one off the net I should say

), here:

Code:

#!/bin/bash
ls | while read file
do
new=`echo $file | tr [:upper:] [:lower:]`
mv -v "$file" "$new"
done

This is fine, but does both filenames and directories. Now, if I do ls -F I get the directory list output with directory names appended with a '/' . So perhaps I could check the name string and only echo those that end with a '/'. Except, of course, I have no clue how to do this, and need showing (I've looked, but no joy) .

Or perhaps my whole approach is wrong and it can be done a different way?

Thx for replies

acid_kewpie · 05-07-2007, 08:21 AM

#!/bin/bash
for i in *
do
[ -f "$i" ] && mv -v "$file" "$(echo $file | tr [:upper:] [:lower:])"
done

should do it.

cfaj · 05-07-2007, 11:32 AM

Quote:

Originally Posted by PaxRomana

EDIT: Good start - I meant to say in the title 'change to LOWER case' not upper duhh.

I have just installed Ubuntu and am moving a lot of files from my Windows machine.

Now, as I have many, many directories with upper case characters in them, I'd like to change them so they are only lower case. However, I only want the DIRECTORIES changed this way, NOT the filenames (got my reasons).

I've made a script (or I got one off the net I should say

), here:

Code:

#!/bin/bash
ls | while read file
do
new=`echo $file | tr [:upper:] [:lower:]`
mv -v "$file" "$new"
done

This is fine, but does both filenames and directories. Now, if I do ls -F I get the directory list output with directory names appended with a '/' . So perhaps I could check the name string and only echo those that end with a '/'. Except, of course, I have no clue how to do this, and need showing (I've looked, but no joy) .

Or perhaps my whole approach is wrong and it can be done a different way?

Code:
for dir in */
do
  new=`printf "%s\n" "${dir%/}" | tr [:upper:] [:lower:]`
  mv -v "$dir" "$new"
done
You can speed it up with a dynamically loadable builtin command (lcase: external commands, e.g., tr, are slow). See http://www.unixreview.com/documents/...6a/ur0606a.htm for details. Then you can use:
Code:
for dir in */
do
  lcase -p new ${file%/}"
  mv -v "$file" "$new"
done

PaxRomana · 05-07-2007, 02:00 PM

acid_kewpie's code didn't work - much appreciated tho m8, as I just spent a couple of hours analysing what you did and finding out more for myself and got it done

.

cfaj - maybe tomorrow if I get time I'll see about working yours in just for the learning experience

. Thx folks.

Here's mine:

Code:

#!/bin/bash

echo

for i in *; do

if [ -d "$i" ]
then
	echo "$i is a directory"
	new=`echo $i | tr [:upper:] [:lower:]`
	if [[ "$i" == "$new" ]]
	then		
		echo "...but is already lower case, so will not be changed"
		echo
	else
		echo "...changing to lower case."
		mv -v "$i" "$new"
		echo
	fi
else
	echo "$i is not a directory"
	echo
fi
done

acid_kewpie · 05-07-2007, 02:16 PM

oh yeah, i didn't replace all the "$file"'s with "$i"'s. my code is *much* simpler though... there's no point using intermediate variables like you are, just a waste of space.

PaxRomana · 05-07-2007, 08:02 PM

Your code still doesn't work dude if I replace the "file"s with "i"'s

. Changes filenames as well as directories

.

PaxRomana · 05-07-2007, 08:14 PM

Oh I see it it's 'cos it has "-f" instead of "-d" . I must be learning something

.

cfaj · 05-07-2007, 08:18 PM

Quote:

Originally Posted by PaxRomana

Your code still doesn't work dude if I replace the "file"s with "i"'s

. Changes filenames as well as directories

.

Code:

for i in */
do
  mv -v "$i" "$(printf "%s\n" "$i" | tr [:upper:] [:lower:])"
done

cfaj · 05-07-2007, 08:21 PM

Quote:

Originally Posted by acid_kewpie

oh yeah, i didn't replace all the "$file"'s with "$i"'s. my code is *much* simpler though... there's no point using intermediate variables like you are, just a waste of space.

Intermediate variables serve various useful purposes. They can:
make code clearer

encapsulate a task

make lines shorter

make modifications easier

ghostdog74 · 05-07-2007, 11:19 PM

Code:

#!/bin/bash
ls -l | awk '/^d/ && $9 ~ /[[:upper:]]/{
		new = tolower($9)
		cmd = "mv " "\047" $9 "\047" " " "\047" new "\047"	
		system(cmd)	
}'

acid_kewpie · 05-08-2007, 01:42 AM

Quote:

Originally Posted by cfaj

Intermediate variables serve various useful purposes. They can:
make code clearer

encapsulate a task

make lines shorter

make modifications easier

yes, but not in such a trviial case as this.

acid_kewpie · 05-08-2007, 01:42 AM

Quote:

Originally Posted by ghostdog74

Code:

#!/bin/bash
ls -l | awk '/^d/ && $9 ~ /[[:upper:]]/{
		new = tolower($9)
		cmd = "mv " "\047" $9 "\047" " " "\047" new "\047"	
		system(cmd)	
}'

mr Sledgehammer, meet mr Nut.

bigearsbilly · 05-08-2007, 04:10 AM

if you only want directories, using
the shell wildcard glob */ is what you want (in bash and ksh anyhow)
eg:

ls -d */

PaxRomana · 05-08-2007, 04:33 AM

Hehe code wars

. Guess it's always like this around here eh?

. All good for educating myself though, even 'Mr Sledgehammer'

. FWIW, I wanted to put the error checking in my version to help me see what was going on and for practice (not written any code in about 5 years).

One final question - can somebody give me an example of a negative IF conditional line? In C you'd just use "!=". Doesn't seem to work here and I can't find a single example of it (every tutorial I've found just does an '=='). Thx again folks

.

bigearsbilly · 05-08-2007, 04:51 AM

in shell scripting
!= == > < are for string comparison

-eq -lt -ne

etc. are for numerical expressions.

man bash or ksh