Renaming hundreds of files at once for proper sorting

Ubuntu comes with a script called rename. It's just a little Perl script that features a number of powerful bulk-renaming features but the best (in this case) is the ability for it to run Perl during the replacement. The result is a truly compact solution:

rename 's/\d+/sprintf("%05d", $&)/e' *.jpg

This is similar to the other printf-style answers here but it's all handled for us. The code above is for a 5-digit number (including a variable number of leading zeros).

It will search and replace the first number-string it finds with a zero-padded version and leave the rest of the filename alone. This means you don't have to worry too much about carrying any extension or prefix over.

Note: this is not completely portable. Many distributions use rename.ul from the util-linux package as their default rename binary. This is a significantly stunted alternative (see man rename.ul) which won't understand the above. If you'd like this on a platform that isn't using Perl's rename, find out how to install that first.

And here's a test harness:

$ touch {1..19}.jpg

$ ls
10.jpg  12.jpg  14.jpg  16.jpg  18.jpg  1.jpg  3.jpg  5.jpg  7.jpg  9.jpg
11.jpg  13.jpg  15.jpg  17.jpg  19.jpg  2.jpg  4.jpg  6.jpg  8.jpg

$ rename 's/\d+/sprintf("%05d", $&)/e' *.jpg

$ ls
00001.jpg  00005.jpg  00009.jpg  00013.jpg  00017.jpg
00002.jpg  00006.jpg  00010.jpg  00014.jpg  00018.jpg
00003.jpg  00007.jpg  00011.jpg  00015.jpg  00019.jpg
00004.jpg  00008.jpg  00012.jpg  00016.jpg

And an example prefixes (we aren't doing anything different):

$ touch track_{9..11}.mp3 && ls
track_10.mp3  track_11.mp3  track_9.mp3

$ rename 's/\d+/sprintf("%02d", $&)/e' *.mp3 && ls
track_09.mp3  track_10.mp3  track_11.mp3

for f in *.jpg ; do if [[ $f =~ [0-9]+\. ]] ; then  mv $f `printf "%.5d" "${f%.*}"`.jpg  ; fi ; done

Edit

Explanation:

• if [[ $f =~ [0-9]+\. ]] makes sure that only files whose names are numbers (followed by a dot) are being renamed.
• printf "%.5d" NUMBER adds the leading zeroes
• "${f%.*}" cuts the extension (.jpg) and leaves just the number
• .jpgafter the second backtick adds the file extension again.

Note that this will work only on file names that are numbers. Left-padding leading zeroes to non-numbered files would require different format.

If you want to experiment try this command:

for f in *.jpg ; do if [[ $f =~ [0-9]+\. ]] ; then echo mv $f `printf "%.5d" "${f%.*}"`.jpg  ; fi ; done

Edit 2

Made the command safer by making sure that only file names that are numbers are being renamed. Note that any pre-existing files named like 00001.jpg will be overwritten.

Below a python script.

The script adds leading zeros up to the defined number of digits. If the name is larger than that, the file(name) is untouched.

Combining different extensions in one rename action might add some convenience. To add extension(s), simply add them to the tuple, for example extensions = (".jpg", ".jpeg", ".tiff").

Copy the text into an empty file, save it as rename.py, enter the correct path to the files directory (sourcedir), the number of digits you'd like the new names to have (number_ofdigits) and the file extension(s) to rename (extensions)

Run it by the command:

python3 /path/to/script/rename.py

The script:

#!/usr/bin/python3

import shutil
import os

sourcedir = "/path/to/files"; number_ofdigits = 5; extensions = (".jpg", ".jpeg")

files = os.listdir(sourcedir)
for item in files:
    if item.endswith(extensions):
        name = item.split("."); zeros = number_ofdigits-len(name[0])
        newname = str(zeros*"0")+name[0]+"."+name[1]
        shutil.move(sourcedir+"/"+item, sourcedir+"/"+newname)

edit:

Below a slightly improved version. It automatically determines the longest name in the directory, and adds leading zeros up to the length of the longest name.

example:

1.jpg
12.jpg
123.jpg

becomes:

001.jpg
012.jpg
123.jpg

No need to set the number of digits.

#!/usr/bin/python3

import shutil
import os

sourcedir = "/path/to/files"; extensions = (".jpg", ".jpeg")
files = [(f, f[f.rfind("."):], f[:f.rfind(".")]) for f in os.listdir(sourcedir)if f.endswith(extensions)]
maxlen = len(max([f[2] for f in files], key = len))

for item in files:
    zeros = maxlen-len(item[2])
    shutil.move(sourcedir+"/"+item[0], sourcedir+"/"+str(zeros*"0")+item[0])

There is a rename utility implemented in Perl that makes this very easy:

rename 's/\d+/sprintf("%05d",$&)/e' *.jpg

The first argument is a Perl expression that is evaluated for each file name.

There is a GUI that will do this called pyRenamer which can be found in the repositories. Install, start, select your directory and just set the 'Renamed file name pattern' to '{num5}'on the 'Patterns' tab and click the 'Rename' button. You can also preview what will happen prior to renaming.

If by any chance your rename command is not perl-rename, you can use this perl code:

foreach (<*.jpg>) { rename $_, sprintf("%03d.jpg", $1) if /^(\d+).jpg/; }

When we say foreach we can refer to the files by $_

But you can install the propper perl-rename with this commands

cpan
cpan1> install File::Rename

Another way of renaming these files easily is opening the folder with ranger file manager, select all files, run :bulkrename and use this vim command:

:%s/^\d\+/\=printf("%03d", submatch(0))

I have also tried using ls + awk

ls -1 | awk -F. '{printf "%s%s%s%03d.%s\n", "mv ", $0, " ", $1, $2}'

Each string or number must be associated on the printf function, the result output will be something like this:

mv 9.jpg 009.jpg

To really perform the renaming with this you have to add | sh at the end of it:

ls -1 | awk -F. '{printf "%s%s%s%03d.%s\n", "mv ", $0, " ", $1, $2}' | sh

The zsh shell has a loadable zmv module, similar to the external mmv command, and it also supports numeric ranges <m-n> in its shell globs. So for instance given:

 % ls *.mp4
'3 - Ariane Rocket Disaster.mp4'           '81 - Proof of Guarantee (Sketch).mp4'
'39 - Conditions Example.mp4'              '82 - Measured vs. Worst-Case Probability.mp4'
'396 - Execution Path of a Program.mp4'    '86 - Coverage of Random Testing.mp4'
'397 - Computation Tree Example.mp4'       '88 - Intro to Automated Test Generation.mp4'
'398 - Existing Approach I.mp4'            '89 - Outline.mp4'
'399 - Existing Approach II.mp4'           '9 - Program Invariants.mp4'
'400 -  Combined Approach.mp4'             '90 - Korat.mp4'
'401 - An Illustrative Example .mp4'       '91 - The Problem.mp4'
'402 - Computation Tree.mp4'               '92 - An Insight.mp4'
'403 - Computation Tree Solution.mp4'      '94 - Scheme for Representing Shapes.mp4'
'7 - Dynamic Program Analysis.mp4'         '95 - Representing Shapes.mp4'
'78 - Concurrency Bug Depth Solution.mp4'  '96 - Representing Shapes Solution.mp4'
'79 - Cuzz Algorithm.mp4'                  '97 - A Simple Algorithm.mp4'
'8 - Static Program Analysis.mp4'          '98 - Enumerating Shapes.mp4'
'80 - Probabilistic Guarantee.mp4'         '99 - Enumerating Shapes Solution.mp4'

then once the module is loaded as

 % autoload -Uz zmv

you can do

 % zmv -n '(<1-99>) - (*.mp4)' '${(l:3::0:)1} - ${2}'
mv -- '3 - Ariane Rocket Disaster.mp4' '003 - Ariane Rocket Disaster.mp4'
mv -- '39 - Conditions Example.mp4' '039 - Conditions Example.mp4'
mv -- '7 - Dynamic Program Analysis.mp4' '007 - Dynamic Program Analysis.mp4'
mv -- '78 - Concurrency Bug Depth Solution.mp4' '078 - Concurrency Bug Depth Solution.mp4'
mv -- '79 - Cuzz Algorithm.mp4' '079 - Cuzz Algorithm.mp4'
mv -- '8 - Static Program Analysis.mp4' '008 - Static Program Analysis.mp4'
mv -- '80 - Probabilistic Guarantee.mp4' '080 - Probabilistic Guarantee.mp4'
mv -- '81 - Proof of Guarantee (Sketch).mp4' '081 - Proof of Guarantee (Sketch).mp4'
mv -- '82 - Measured vs. Worst-Case Probability.mp4' '082 - Measured vs. Worst-Case Probability.mp4'
mv -- '86 - Coverage of Random Testing.mp4' '086 - Coverage of Random Testing.mp4'
mv -- '88 - Intro to Automated Test Generation.mp4' '088 - Intro to Automated Test Generation.mp4'
mv -- '89 - Outline.mp4' '089 - Outline.mp4'
mv -- '9 - Program Invariants.mp4' '009 - Program Invariants.mp4'
mv -- '90 - Korat.mp4' '090 - Korat.mp4'
mv -- '91 - The Problem.mp4' '091 - The Problem.mp4'
mv -- '92 - An Insight.mp4' '092 - An Insight.mp4'
mv -- '94 - Scheme for Representing Shapes.mp4' '094 - Scheme for Representing Shapes.mp4'
mv -- '95 - Representing Shapes.mp4' '095 - Representing Shapes.mp4'
mv -- '96 - Representing Shapes Solution.mp4' '096 - Representing Shapes Solution.mp4'
mv -- '97 - A Simple Algorithm.mp4' '097 - A Simple Algorithm.mp4'
mv -- '98 - Enumerating Shapes.mp4' '098 - Enumerating Shapes.mp4'
mv -- '99 - Enumerating Shapes Solution.mp4' '099 - Enumerating Shapes Solution.mp4'

where (l:3::0:) in the replacement variable expansion is a modifier that left pads to width 3 using string 0.

Remove the -n flag once you are happy with the proposed renamings.

for file in `(find * -maxdepth 1 -type f -name "*.jpg")`     
do    
  mv $file "000"$file    
done

This is crude, since it doesn't do the renaming intelligently, will need to think of a trick for that. But it should help you sort stuff.

Update: Well you have a more complete answer.