Hide

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.

Hide

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:

Hide

News

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.
Hide

Tags

Hide

Functions

Commands tagged bash from sorted by
Terminal - Commands tagged bash - 684 results
for id in `ls -1 ~/.ssh | grep -v "authorized\|known_hosts\|config\|\."` ; do echo -n "$id: " ; ssh-keygen -l -f .ssh/$id ; done
2014-04-16 14:12:20
User: drockney
Functions: echo grep id ssh-keygen
0

Find all private keys and dump their fingerprints.

rename 's/\.sh//' ./*
2014-04-02 16:33:25
User: abhikeny
Functions: rename
0

The 'rename' command with the first argument as "'s/\.//'" and the second argument as "" will remove the specified extension from the filenames.

while read i; do [ ${#i} -gt 72 ] && echo "$i"; done < /path/to/file
2014-03-20 12:27:06
User: flatcap
Functions: echo read
1

Filter out lines of input that contain 72, or fewer, characters.

This uses bash only. ${#i} is the number of characters in variable i.

psql -h <SERVER NAME HERE> -c 'SELECT version();' | grep -v 'version\|---\|row\|^ *$' | sed 's/^\s*//'
2014-03-17 18:36:40
User: pnelsonsr
Functions: grep sed
0

In a multiple PostgreSQL server environment knowing the servers version can be important. Note that psql --version returns just the local psql apps version which may not be what you want. This command dumps the PostgreSQL servers version out to one line. You may need to add more command line options to the psql command for your connection environment.

alias ...='while read line; do echo -n "."; done && echo ""'
for i in `sudo /sbin/fdisk -l |grep Disk |grep dev |awk '{ print $2 }' |sed s/://g` ; do sudo /usr/bin/dd if=/dev/urandom of=$i bs=8M & done
2014-03-16 12:04:59
User: aimana007
Functions: awk grep sed sudo
Tags: bash Linux
0

This command will use the fdisk utility to find all block devices on your system, and overwrite them with data from the /dev/urandom non-blocking random number generator.

CAUTION: This will irrevocably erase EVERY SINGLE physical block storage device visible to the fdisk utility, including plugged USB devices, RAID sets, LVM, etc.

wmic OS get OSArchitecture /value | grep -Eo '[^=]*$'
2014-03-15 02:04:08
User: lowjax
Functions: get grep
0

Using "wmic get * /value" within any Cygwin shell will return lots of Win/Dos newline junk ie "^M$" at the end of found value line, two lines ("$" Unix newline) above, and three below. This makes storing and or evaluating wmic queries as variables a pain. The method i suggest strips the mentioned junk, only returns the value after "OSArchitecture=", and includes only one Unix style newline. Other methods using sed|awk|cut can only handle the output of wmic cleanly when piped or using multiple sed statements.

wmic OS get OSArchitecture /value | sed 's/\r//g;s/^M$//;/^$/d;s/.*=//'

making

wmic OS get OSArchitecture /value | grep -Eo '[^=]*$'

a much cleaner and slightly less costly alternative.

bashrc-reload() { builtin unalias -a; builtin unset -f $(builtin declare -F | sed 's/^.*declare[[:blank:]]\+-f[[:blank:]]\+//'); . ~/.bashrc; }
2014-03-02 14:24:18
User: Xk2c
Functions: sed unalias unset
2

Simply sourcing .bashrc does not function correctly when you edit it and change an alias for a function or the other way round with the *same name*.

I therefor use this function. Prior to re-sourcing .bashrc it unsets all aliases and functions.

tput rmam
2014-02-26 07:06:37
User: kennyld
Functions: tput
Tags: bash ksh
0
tput rmam

will disable line wrapping so that long lines are truncated to width of the terminal ($COLUMNS).

tput smam

will re-enable wrapping.

I've always used tput in bash scripts but I guess it works on the command line too.

Doesn't work in all terminals.

See http://www.gnu.org/software/termutils/manual/termutils-2.0/html_chapter/tput_1.html

printf -- " -e %s" ${ARRAY[*]}
2014-02-25 03:34:12
User: SEJeff
Functions: printf
1
DOCKER_APP_VARS=(DATABASE_USER=dbuserro, DATABASE_PASSWORD=maipass)

[jeff@omniscience container] (master)$ echo docker run $(printf -- " -e %s" ${DOCKER_APP_VARS[*]}) -name 12factorapp mattdm/fedora-small

docker run -e DATABASE_USER=dbuserro, -e DATABASE_PASSWORD=maipass -name 12factorapp mattdm/fedora-small

Note that the printf method by itsself doesn't include a newline (\n), so you'll need to embed it into an echo statement or something that does.

ps -eo pmem,comm | grep application-name
2014-02-23 13:21:29
User: Darkstar
Functions: grep ps
0

Displays memory usage for individual instances of an application that spawns multiple instances of itself. This command also works on single process applications.

(read -r passphrase; b58encode 80$( brainwallet_exponent "$passphrase" )$( brainwallet_checksum "$passphrase" ))
3

A bitcoin "brainwallet" is a secret passphrase you carry in the "wallet" of your brain.

The Bitcoin Brainwallet Private Key Calculator calculates the standard base58 encoded bitcoin private key from your "brainwallet" passphrase.

The private key is the most important bitcoin number. All other numbers can be derived from it.

This command uses 3 other functions - all 3 are defined on my user page:

1) brainwallet_exponent() - search for Bitcoin Brainwallet Exponent Calculator

2) brainwallet_checksum() - search for Bitcoin Brainwallet Exponent Calculator

3) b58encode() - search for Bitcoin Brainwallet Base58 Encoder

Do make sure you use really strong, unpredictable passphrases (30+ characters)!

http:brainwallet.org can be used to check the accuracy of this calculator.

function b58encode () { local b58_lookup_table=({1..9} {A..H} {J..N} {P..Z} {a..k} {m..z}); bc<<<"obase=58;ibase=16;${1^^}"|(read -a s; for b58_index in "${s[@]}" ; do printf %s ${b58_lookup_table[ 10#"$b58_index" ]}; done); }
3

A bitcoin "brainwallet" is a secret passphrase you carry in your brain.

The Bitcoin Brainwallet Private Key Base58 Encoder is the third of three functions needed to calculate a bitcoin PRIVATE key from your "brainwallet" passphrase.

This base58 encoder uses the obase parameter of the amazing bc utility to convert from ASCII-hex to base58. Tech note: bc inserts line continuation backslashes, but the "read s" command automatically strips them out.

I hope that one day base58 will, like base64, be added to the amazing openssl utility.

find . -mtime +30 -exec mv {} old/ \;
2014-02-09 23:05:41
User: minnmass
Functions: find mv
Tags: bash file
-2

Use find's built-in "exec" option to avoid having to do any weirdness with quoting.

quickscript () { filename="$1"; history | cut -c 8- | sed -e '/^###/{h;d};H;$!d;x' | sed '$d' > ${filename:?No filename given} }
2014-02-09 12:19:29
User: joedhon
Functions: cut sed
1

In order to write bash-scripts, I often do the task manually to see how it works. I type ### at the start of my session.

The function fetches the commands from the last occurrence of '###', excluding the function call. You could prefix this with a here-document to have a proper script-header.

Delete some lines, add a few variables and a loop, and you're ready to go.

This function could probably be much shorter...

for i in $(find . -mtime +30); do mv $i old/; done
2014-02-05 01:24:45
User: valferon
Functions: find mv
Tags: bash file
-3

Will move in that case every file in the current folder older than 30 days to the "old" folder

Replace "mv $i old/" by any command such as rm / echo to do something different.

CMD=chrome ; ps h -o pmem -C $CMD | awk '{sum+=$1} END {print sum}'
rsstail -o -n 1 --f 'RedditQuote: {title}' http://www.reddit.com/r/quotes/new/.rss
2014-01-08 15:33:48
User: Youpfu
-1

The rsstail is the python version (python-rsstail). The final command pipe it on every new line to gammu and a connected phone

rsstail -o -n 1 --f 'RedditQuote: {title}' http://www.reddit.com/r/quotes/new/.rss | while read line; do /usr/bin/gammu --sendsms TEXT +*yournumber* -text "$line"; done
watch -n 10 free -m
2014-01-04 10:10:15
User: Darkstar
Functions: free watch
-1

This command shows a high level overview of system memory and usage refreshed in seconds. Change -n 10 to you desired refresh interval.

ps -eo pmem,comm | grep chrome | cut -d " " -f 2 | paste -sd+ | bc
2014-01-03 15:33:16
User: Darkstar
Functions: cut grep paste ps
0

This command will show the sum total of memory used in gigabytes by a program that spawns multiple instances of itself. Replace chrome with whatever program's memory usage you are investigating. This command is rather useless on software that only spawns a single instance of itself.

alias ?=concalc
2014-01-02 01:46:44
User: boynux
Functions: alias
0

Same functionality without using bash functions.

for i in */; do echo $(find $i -type f -regextype posix-extended -regex ".*\.(mp3|ogg|wav|flac)" | wc -l) $i ; done
find . -type d -maxdepth 1 -print0 | xargs -0 -I{} sh -c 'find "{}" -type f | grep "ogg\|mp3\|wav\|flac$" | wc -l | tr -d "\n"; echo " {}"'
2013-12-22 13:40:29
User: dbrgn
Functions: echo find grep sh tr wc xargs
0

This lists the number of ogg/mp3/wav/flac files in each subdirectory of the current directory. The output can be sorted by piping it into "sort -n".

bind -P | grep -v "is not" | sed -e 's/can be found on/:/' | column -s: -t
2013-12-19 12:30:19
User: leni536
Functions: column grep sed
0

Shows all available keyboard bindings in bash. Pretty printing.

ls -d .??*
2013-12-17 19:06:30
User: pydave
Functions: ls
Tags: bash glob
2

You can omit the -d to see what's inside directories. In that case, you may want -a to see dotfiles inside those directories. (Otherwise you don't need -a since you're explicitly looking at them.)