How can I easily bulk rename files with Perl?

Question

I have a lot of files I'm trying to rename, I tried to make a regular expression to match them, but even that I got stuck on the files are named like:

File Name 01

File Name 100

File Name 02

File Name 03

etc, I would like to add a "0" (zero), behind any of file that are less than 100, like this:

File Name 001

File Name 100

File Name 002

File Name 003

The closest I got to so much as matching them was using this find -type d | sort -r | grep ' [1-9][0-9]$' however I could not figure out how to replace them. Thanks in advance for any help you can offer me. Im on CentOS if that is of any help, all this is being done via SSH.

Can you elaborate on the "tools" available on the machine? Do you have perl, python, ruby or similar installed? — R. Martinho Fernandes, Dec 02 '09 at 23:19
could you clarify which are the names of the files you're trying to match are, and which are the names you're trying to avoid matching? I'm a little confused here. — mpobrien, Dec 02 '09 at 23:20
Perl (5.8.8) is on the machine, however python and ruby are not, I am familiar enough to go through yum and install them if necessary. — linux_rookie, Dec 02 '09 at 23:21
mpobrien; im trying to match files named like: "Folder Name 12" "Folder Name 120" There is 2 words in the names the names are all the same, its just the numbers are different, in short I want to add a zero before any number under than 100, so Im trying to turn "File Name 10" into "File name 010", let me know if thats a bit more clear. Sorry for any confusion — linux_rookie, Dec 02 '09 at 23:25
Are you running a program locally that must rename files on a remote system via SSH? Or can you install and run a program on the target system? — daotoad, Dec 02 '09 at 23:26
daotoad, its being done remotely via SSH. I can install stuff on the target system. — linux_rookie, Dec 02 '09 at 23:31

score 13 · Accepted Answer · answered Dec 02 '09 at 23:33

13

perl -e 'foreach $f (glob("File\\ Name*")) { $nf = $f; $nf =~ s/(\d+)$/sprintf("%03d",$1)/e; print `mv \"$f\" \"$nf\"`;}'

A bit overkill maybe, but it does what is asked.

answered Dec 02 '09 at 23:33

Jeff B

29,943
7
61
90

score 10 · Answer 2 · answered Dec 02 '09 at 23:24

10

find . -type d -print0 | xargs -0 rename 's/(\d+)/sprintf "%03d", $1/e'

or something like that, provided

You have GNU find and GNU xargs (for -print0 and -0)
You have the 'rename' utility that comes with perl
There's only one group of digits in the filename. If there's more than one, then you need to do something with the regex to make it only match the number you want to reformat.

answered Dec 02 '09 at 23:24

hobbs

223,387
19
210
288

That's close to something I was going to suggest, using sed instead of xargs, so I'll +1 it. – shoover Dec 02 '09 at 23:27
I knew xargs was the way to go, I just couldn't write the exact command. – R. Martinho Fernandes Dec 02 '09 at 23:29
+1 For listing dependencies (it's possibly the rename utility you are referring to is one yum or apt-get away). – ChristopheD Dec 02 '09 at 23:31
1

I've always been pissed that the cygwin perl distro doesn't include rename. – rmeador Dec 02 '09 at 23:33

score 4 · Answer 3 · answered Dec 02 '09 at 23:25

4

Is this a one-time thing? If so, I'm going to suggest something that might seem to be a cop out by many programmers here:

Pipe the output of your command (find -type d | sort -r | grep ' [1-9][0-9]$') to a file and use an editor along with some global search/replace magic to create a script that does the renames.

Then throw away the script.

There's little fuss and little chance that you'll end up shooting yourself in the foot by having some attempt at a clever (but inadequately debugged) one-liner go off into the weeds on your files.

answered Dec 02 '09 at 23:25

Michael Burr

333,147
50
533
760

++! Sometimes it is better to use a kludge you understand than to learn "deep magic". – daotoad Dec 02 '09 at 23:30
+1 I think this sort of thing is often the most pragmatic solution for one off requirements – Brian Agnew Dec 02 '09 at 23:33
It may be a cop-out, but it's a safe cop-out...kudos. – Mark Krenitsky Dec 02 '09 at 23:34
A possible global search and replace magic in vim for this is `:s/^$.*$$\d\d$$/mv \1 \10\2/` – R. Martinho Fernandes Dec 02 '09 at 23:37
1

In Emacs you can open a directory listing, press C-x C-q, edit the filenames in the buffer (e.g. with search and replace, or any other command), and upon saving Emacs will rename the files. – Phil Dec 03 '09 at 00:16

score 2 · Answer 4 · answered Dec 02 '09 at 23:25

2

Run two commands, in this order:

$ rename 's/File Name (\d)$/File Name 0$1/' *
$ rename 's/File Name (\d\d)$/File Name 0$1/' *

First one renames everything less than 10 and prepends a zero. The second one renames everything less than 100 and prepends a zero. The result should be three digits for all filenames.

answered Dec 02 '09 at 23:25

ire_and_curses

68,372
23
116
141

`rename` is a utility which comes with the standard perl distribution. I ran these under Linux Bash. – ire_and_curses Dec 02 '09 at 23:31

score 1 · Answer 5 · answered Dec 02 '09 at 23:35

1

In my debian it works well with rename, tested with 300 files.

 perl -e 'map `touch door$_.txt`, 1..300;'
 rename 's/(\d+)\.txt/sprintf("%03d.txt", $1)/e' *.txt

answered Dec 02 '09 at 23:35

pavel

377
2
11

score 0 · Answer 6 · answered Dec 02 '09 at 23:21

0

I think mmv is your friend here.

answered Dec 02 '09 at 23:21

Ken

1,066
1
10
17

score 0 · Answer 7 · answered Dec 02 '09 at 23:25

0

you could do something using perl or ruby.

put all this files in the same directory

dirlisting = DIR.entries('.')

dirListing.each do |file| 
 num = file.match(/\d+$/).to_i
 if num < 100
   find the position where start the number, with index and inject the 0 there.
 end
end

answered Dec 02 '09 at 23:25

VP.

5,122
6
46
71

You should probably mention that this code is ruby and not perl – R. Martinho Fernandes Dec 02 '09 at 23:27
And that's why I said you should disambiguate the fact that the code is ruby. – R. Martinho Fernandes Dec 02 '09 at 23:39

chris d · Answer 8 · 2009-12-02T23:52:32.507

0

use strict;
use File::Copy;

my @files = glob 'File*Name*';

foreach my $filename (@files) {
    if ($filename =~ m`^.*File.*Name.*?(\d+)`) {
        my $number = $1;
        next if ($number > 99);
        rename $filename, sprintf("FileName%03d",$number);
    }
}

edited Dec 02 '09 at 23:52

answered Dec 02 '09 at 23:34

chris d

19
3

That renames `File Name 100` to `File Name 0100` which is not intended. – R. Martinho Fernandes Dec 02 '09 at 23:41
You were most certainly right. Added some code to make it so that won't happen. – chris d Dec 02 '09 at 23:52

score 0 · Answer 9 · answered Dec 03 '09 at 00:09

0

if your remote has bash shell

for i in File*; 
do 
    case "${i##* }" in  [0-9][0-9] ) 
      echo  mv "$i" "${i% *} $(printf "%03d" ${i##* })" ;; 
    esac; 
done

remove "echo" to do actual renaming

answered Dec 03 '09 at 00:09

ghostdog74

327,991
56
259
343

How can I easily bulk rename files with Perl?

9 Answers9

Linked