How do I change the extension of multiple files?

Question

I would like to change a file extension from *.txt to *.text. I tried using the basename command, but I'm having trouble on changing more than one file.

Here's my code:

files=`ls -1 *.txt`

for x in $files
do
    mv $x "`basename $files .txt`.text"
done

I'm getting this error:

basename: too many arguments Try basename --help' for more information

Answer 1

Straight from Greg's Wiki:

# Rename all *.txt to *.text
for file in *.txt; do 
    mv -- "$file" "${file%.txt}.text"
done

*.txt is a globbing pattern, using * as a wildcard to match any string. *.txt matches all filenames ending with '.txt'.

-- marks the end of the option list. This avoids issues with filenames starting with hyphens.

${file%.txt} is a parameter expansion, replaced by the value of the file variable with .txt removed from the end.

Also see the entry on why you shouldn't parse ls.

If you have to use basename, your syntax would be:

for file in *.txt; do
    mv -- "$file" "$(basename -- "$file" .txt).text"
done

Answer 2

Here's how I change all the file extensions in the current directory on Debian.

rename "s/oldExtension$/newExtension/" *.txt

(This is the Perl rename command, not the util-linux one. See Why is the rename utility on Debian/Ubuntu different than the one on other distributions, like CentOS?)

On MacOS, user Monkpit reports that they were able to use brew install rename to get this to work.

Answer 3

A simple command, rename from util-linux, will do that for you. It replaces every occurrence of "txt" with "text" in all files matching "*.txt":

rename txt text *.txt

Answer 4

rename "s/oldExtension/newExtension/" *.txt

Above works fine but limited to current directory. Try the command below, which is flexible with sub-directories. It will rename all .txt files under directory structure with a new extension.

find . -type f -name "*.txt" -exec rename 's/\.txt$/.newext/' '{}' \;

Answer 5

The answers here referencing s/oldExtension/newExtension/ are wrong. If you use s/txt/text/, you would convert footxt.txt to footext.txt, which is not what you want. Even if you use s/.txt/.text/, that would convert footxt.txt to fo.text.txt.

You have to use \. to match the period (. will match any character). And the trailing $ to match the end of the line. Only this will properly match the extension.

rename 's/\.txt$/.text/' *.txt

rename 's/\.old$/.new/' *.old

Answer 6

Reason #53 to switch to zsh:

zmv '(*).txt' '$1.text'

Answer 7

Based on the @Prince John Wesley answer, here is a simple bash script for changing all extensions of files in the current directory from ext1 to ext2. Also outputs names of the files being renamed.

#!/bin/bash
for f in *.$1
do
    [ -f "$f" ] && mv -v "$f" "${f%$1}$2"
done

Example usage (assuming the name of the script is change-ext):

change-ext ext1 ext2

Answer 8

for f in *.txt
do
    [ -f "$f" ] && mv "$f" "${f%txt}text"
done

Answer 9

let's say your files are scattered in various directory, Assuming that dirx is your parent directory, this can do the job using find:

for f in `find /dirx -iname '*.txt' -type f -print`;do  mv "$f" ${f%.txt}.text; done

Answer 10

On Ubuntu 18.04, the util-linux rename command is available as rename.ul. This worked for me:

rename.ul -o -v .oldext .newext *.oldext

Options:

• -o: don't overwrite preexisting .newext
• -v: verbose
• -n: dry run

For more info, see man rename.ul or rename.ul -h.

Answer 11

When you

do not have an extension for the source files

and target extension is .text you would do it this way -

for f in *; do mv -- "$f" "${f%.\*}.text"; done

Answer 12

This is what works for me:

find . -name '*.txt' -exec rename 's/\.txt$/.text/' \{} \;

Answer 13

Mmv (available in the main distributions repositories) is also very useful for renaming. Give the patterns in quotes and each glob element can be reproduced by #N:

mmv '*.txt' '#1.text'

Some more interesting, neat examples in the manual page:

Rename files ending in .html.en, .html.de, etc. to ending in .en.html, .de.html, etc.:

mmv '*.html.??' '#1.#2#3.html' 

Rename music files from <track no.> - <interpreter> - <song title>.ogg to <interpreter> - <track no.> - <song title>.ogg:

mmv '* - * - *.ogg' '#2 - #1 - #3.ogg' 

Answer 14

In case you want to know what went wrong in your version: You used $files instead of $x in the basename command. So this should work (untested, though):

for x in *.txt
do
  mv "$x" "`basename "$x" .txt`.text"
done

Answer 15

for f in *.old_ext; do
  mv -- "$f" "$(echo "$f" | sed "s/\.old_ext$/.new_ext/")"
done

I am posting this because I have not seen another answer that details this method and why it could potentially be superior to the accepted answer. You can go here to see all the reasons why this method is better than all the other answers, but basically, the accepted answer is not POSIX-compliant since it relies on bash parameter expansion.

I realize OP tagged bash specifically, and for that reason, the accepted answer will work just fine, but if you wanted to use this command in a shell script on a system using sh or dash, you could use the version I posted above, which utilizes any version of sed (GNU/BSD/anything), and will work just fine, while not relying specifically on bash itself.

I actually wrote that command for this specific use case, which I encounter frequently and have been using it as a bash/zsh function for a long time, which I named chext:

chext() {
  old_ext="$1"
  new_ext="$2"
  for file in *.${old_ext}; do
    mv -v -- "$file" "$(echo "$file" | sed "s/\.${old_ext}$/.${new_ext}/")"
  done
}

Usage:

chext [old_extension] [new_extension]

Recently, I had the need to use it on a different computer, and instead of just a one-off copy/paste, I actually added it to my main set of portable commands, phxutils.

After publishing chext in phxutils, I did a quick search to see if there was actually a better way of accomplishing what I set out to do, and after browsing through all of these answers, I still think my way is the best and most universal (at least for current directory extension replacement -- for other methods, I would use one of the find variations)

Answer 16

Nobody has shown using shell parameter expansion, which is the most basic, probably shell compliant, (and I believe most readable) way of doing this.

# by removing suffix 
for f in *.txt; do echo "$f" "${f%txt}text"; done

# by substitution
for f in *.txt; do echo "$f" "${f/%txt/text}"; done

# simply edit mv for echo after testing
# % searched for shortest string, %% is 'greedy' would search for longest
# if you have a basename, use basename*.txt

https://www.debuntu.org/how-to-bash-parameter-expansion-and-string-manipulation/
https://www.gnu.org/software/bash/manual/html_node/Shell-Parameter-Expansion.html

Answer 17

A very simple solution, based on the old good xargs, to add a suffix/extension to filenames corresponding to <pattern>:

ls <pattern> | xargs -i mv {} {}.<suffix>

Answer 18

For a slightly different take, in the case of a directory under git version control, you might consider:

# Rename all *.txt to *.text
for file in *.txt; do 
    git mv -- "$file" "${file%.txt}.text"
done

In my case this avoided having git think the files were simply deleted.