What's this?

commandlinefu.com is the place to record those command-line gems that you return to again and again.

Delete that bloated snippets file you've been using and share your personal repository with the world. That way others can gain from your CLI wisdom and you from theirs too. All commands can be commented on, discussed and voted up or down.

If you have a new feature suggestion or find a bug, please get in touch via http://commandlinefu.uservoice.com/

Get involved!

You can sign-in using OpenID credentials, or register a traditional username and password.

First-time OpenID users will be automatically assigned a username which can be changed after signing in.

Universal configuration monitoring and system of record for IT.

Stay in the loop…

Follow the Tweets.

Every new command is wrapped in a tweet and posted to Twitter. Following the stream is a great way of staying abreast of the latest commands. For the more discerning, there are Twitter accounts for commands that get a minimum of 3 and 10 votes - that way only the great commands get tweeted.

» http://twitter.com/commandlinefu
» http://twitter.com/commandlinefu3
» http://twitter.com/commandlinefu10

Subscribe to the feeds.

Use your favourite RSS aggregator to stay in touch with the latest commands. There are feeds mirroring the 3 Twitter streams as well as for virtually every other subset (users, tags, functions,…):

Subscribe to the feed for:



2011-03-12 - Confoo 2011 presentation
Slides are available from the commandlinefu presentation at Confoo 2011: http://presentations.codeinthehole.com/confoo2011/
2011-01-04 - Moderation now required for new commands
To try and put and end to the spamming, new commands require moderation before they will appear on the site.
2010-12-27 - Apologies for not banning the trolls sooner
Have been away from the interwebs over Christmas. Will be more vigilant henceforth.
2010-09-24 - OAuth and pagination problems fixed
Apologies for the delay in getting Twitter's OAuth supported. Annoying pagination gremlin also fixed.




Commands using rm from sorted by
Terminal - Commands using rm - 259 results
find . | sort | awk 'NR%2==0' | xargs rm $1
2013-07-11 07:36:18
User: sucotronic
Functions: awk find rm sort xargs

If you have a directory with lot of backups (full backups I mean), when it gets to some size, you could want to empty some space. With this command you'll remove half of the files. The command assumes that your backup files starts with YYYYMMDD or that they go some alphabetical order.

find . -name ".DS_Store" -print0 | xargs -0 rm -rf
find . -type f ! -path \*CVS\* -exec rm {} \; -exec cvs remove {} \;
2013-06-28 20:17:40
User: jasonsydes
Functions: cvs find rm
Tags: bash cvs delete rm

This command removes and then cvs removes all files in the current directory recursively.

mkdir -p temp && for f in *.pdf ; do qpdf --password=YOURPASSWORDHERE --decrypt "$f" "temp/$f"; done && mv temp/* . && rm -rf temp
2013-06-25 18:41:51
Functions: mkdir mv rm

Replace YOURPASSWORDHERE with the pdf password. [qpdf needed]

sudo apt-get install git gcc make libx11-dev libxtst-dev pkg-config -y && git clone https://github.com/hanschen/ksuperkey.git && cd ksuperkey && make && sudo mv ksuperkey /usr/bin/ksuperkey && cd ~ && rm -rf ksuperkey
2013-04-17 07:12:46
User: FadeMind
Functions: cd gcc install make mv rm sudo

Install Ksuperkey one command in Kubuntu.

You must manually add ksuperkey to autostart in System Settings KDE.

rm index.html | wget www.google.com;cat index.html | sed 's/<script>/\n\n\<script>\n\n/g' | sed 's/<\/script>/>\n\n/g'
2013-04-10 04:05:30
User: lbhack
Functions: cat rm sed wget

remove old index.html if you download it again and organiaz the java script tag on the file index.html

find . -type f -exec echo echo rm {} '|' batch ';'|bash
2013-03-01 15:14:08
User: Ztyx
Functions: batch echo find rm

While `echo rm * | batch` might seem to work, it might still raise the load of the system since `rm` will be _started_ when the load is low, but run for a long time. My proposed command executes a new `rm` execution once every minute when the load is small.

Obviously, load could also be lower using `ionice`, but I still think this is a useful example for sequential batch jobs.

find ./ -type f -mtime +365 -exec rm -f {} \;
find . -type f -name "*.txt" | while read; do (($(cat $THISFILE | wc -l) < 10)) && rm -vf "$THISFILE"; done
dd if=/dev/zero of=T bs=1024 count=10240;mkfs.ext3 -q T;E=$(echo 'read O;mount -o loop,offset=$O F /mnt;'|base64|tr -d '\n');echo "E=\$(echo $E|base64 -d);eval \$E;exit;">F;cat <(dd if=/dev/zero bs=$(echo 9191-$(stat -c%s F)|bc) count=1) <(cat T;rm T)>>F
2013-01-31 01:38:30
User: rodolfoap

This is just a proof of concept: A FILE WHICH CAN AUTOMOUNT ITSELF through a SIMPLY ENCODED script. It takes advantage of the OFFSET option of mount, and uses it as a password (see that 9191? just change it to something similar, around 9k). It works fine, mounts, gets modified, updated, and can be moved by just copying it.


The file is composed of three parts:

a) The legible script (about 242 bytes)

b) A random text fill to reach the OFFSET size (equals PASSWORD minus 242)

c) The actual filesystem

Logically, (a)+(b) = PASSWORD, that means OFFSET, and mount uses that option.

PLEASE NOTE: THIS IS NOT AN ENCRYPTED FILESYSTEM. To improve it, it can be mounted with a better encryption script and used with encfs or cryptfs. The idea was just to test the concept... with one line :)

It applies the original idea of http://www.commandlinefu.com/commands/view/7382/command-for-john-cons for encrypting the file.

The embedded bash script can be grown, of course, and the offset recalculation goes fine. I have my own version with bash --init-file to startup a bashrc with a well-defined environment, aliases, variables.

echo '#!/bin/bash' > junk.sh ; find . -iname *.pdf -type f -printf \p\s\2\a\s\c\i\i\ \"%p\"\ \ \"%p\.\t\x\u\"\;\ \p\a\r\ \<\"%p\.\t\x\u\"\ \>\"%p\.\t\x\t\"\ \;\ \r\m\ \"%p\.\t\x\u\"\ \\n >>junk.sh; chmod 766 junk.sh; ./junk.sh ; rm junk.sh
2013-01-27 21:29:08
User: p0g0
Functions: chmod echo find rm

Linux users wanting to extract text from PDF files in the current directory and its sub-directories can use this command. It requires "bash", "ps2ascii" and "par", and the PARINIT environment variable sanely set (see man par). WARNING: the file "junk.sh" will be created, run, and destroyed in the current directory, so you _must_ have sufficient rights. Edit the command if you need to avoid using the file name "junk.sh"

scrotit(){ echo "Screenshot in $1 seconds...";scrot -d $1 '%Y%m%d%h.png' -e 'curl -sF file1=@- http://ompldr.org/upload < $f | grep -P -o "(?<=File:).*(http://ompldr.org/.*)\<\/a\>";rm $f'| sed -r 's@.*(http://ompldr.org/\w{1,7}).*@\1@';}
2012-12-03 01:21:19
User: dzup
Functions: echo grep rm sed

Take a screenshot, give $1 seconds pause to choose what to screenshot, then upload and get URI of post in ompdlr.org

find . -size 0c -print -exec rm -f {} \;
for I in $(find . -depth -type d -not -path "*/.svn*" -print) ; do N="$(ls -1A ${I} | wc -l)"; if [[ "${N}" -eq 0 || "${N}" -eq 1 && -n $(ls -1A | grep .svn) ]] ; then svn rm --force "${I}"; fi ; done
edit-notime () { FILE=$1; TMP=`mktemp /tmp/file-XXXXXX`; cp -p $FILE $TMP; $EDITOR $TMP; touch -r $FILE $TMP; cp -p $TMP $FILE; rm -f $TMP; }
2012-10-31 00:54:19
User: jecxjoopenid
Functions: cp rm touch

Copies file to a temporary location, edit and set to real file's time stamp then copy back. Assumes access to /tmp and has $EDITOR, but can be replaced with better values.

rm ^*.(tex|pdf)(.)
2012-10-17 16:36:56
User: fooacad
Functions: rm

Only zsh supports it. This removes all the regular files in the current directory except for any .tex and .pdf files.

diff ../source-dir.orig/ ../source-dir.post/ | grep "Only in" | sed -e 's/^.*\:.\(\<.*\>\)/\1/g' | xargs rm -r
2012-10-17 14:12:32
User: bigc00p
Functions: diff grep rm sed xargs

Good for when your working on building a clean source install for RPM packaging or what have you. After testing, run this command to compare the original extracted source to your working source directory and it will remove the differences that are created when running './configure' and 'make'.

rm -v *.(log|toc|aux|nav|snm|out|tex.backup|bbl|blg|bib.backup|vrb|lof|lot|hd|idx)(.e/'[[ -f ${REPLY:r}.tex ]]'/)
2012-09-18 20:49:28
User: xro
Functions: rm
Tags: rm zsh latex test

Uses zsh globbing syntax to safely remove all the files known to be generated by LaTeX, but only if there is actually a .tex source file with the same basename present. So we don't accidentally delete a .nav .log or .out file that has nothing to do with LaTeX, e/'[[ -f ${REPLY:r}.tex ]]'/ actually checks for the existance of a .tex file of the same name, beforehand.

A different way to do this, would be to glob all *.tex files and generate a globbing pattern from them:

TEXTEMPFILES=(*.tex(.N:s/%tex/'(log|toc|aux|nav|snm|out|tex.backup|bbl|blg|bib.backup|vrb|lof|lot|hd|idx)(.N)'/)) ; rm -v ${~TEXTEMPFILES}

or, you could use purge() from grml-etc-core ( http://github.com/grml/grml-etc-core/blob/master/usr_share_grml/zsh/functions/purge )

rm **/*(-@)
2012-09-18 20:18:57
User: xro
Functions: rm
Tags: rm symlinks zsh

recursively deletes all broken symlinks using zsh globbing syntax.

cp foo.txt foo.txt.tmp; sed '$ d' foo.txt.tmp > foo.txt; rm -f foo.txt.tmp
2012-09-13 20:57:40
User: kaushalmehra
Functions: cp rm sed
Tags: sed unix

sed '$ d' foo.txt.tmp

...deletes last line from the file

find . -name vmware-*.log -exec rm -i {} \;
touch -t 201208211200 first ; touch -t 201208220100 last ; find /path/to/files/ -newer first ! -newer last | xargs -ifile mv -fv file /path/to/destination/ ; rm first; rm last;
2012-08-22 09:51:40
User: ktopaz
Functions: file find last mv rm touch xargs

touch -t 201208211200 first ; touch -t 201208220100 last ;

creates 2 files: first & last, with timestamps that the find command should look between:

201208211200 = 2012-08-21 12:00

201208220100 = 2012-08-22 01:00

then we run find command with "-newer" switch, that finds by comparing timestamp against a reference file:

find /path/to/files/ -newer first ! -newer last

meaning: find any files in /path/to/files that are newer than file "first" and not newer than file "last"

pipe the output of this find command through xargs to a move command:

| xargs -ifile mv -fv file /path/to/destination/

and finally, remove the reference files we created for this operation:

rm first; rm last;

find . -name ".svn" -type d -exec rm -rf {} \;
2012-08-16 13:04:47
User: gigo6000
Functions: find rm
Tags: svn git rm

This is useful when you are uploading svn project files to a new git repo.

ls -il; find * \( -type d -prune \) -o -inum <NUM> -exec rm -i {} \;
2012-08-04 00:22:33
User: yingw
Functions: find ls rm

Code to delete file with gremlins/special characters/unicode in file name.

Use ls -i to find the INODE number corresponding to the file and then delete it using that find statement.

detailed here:


find /backup/directory -name "FILENAME_*" -mtime +15 -exec rm -vf {} +