Commands tagged grep (409)

  • Grep for a named process. Show Sample Output


    -4
    psgrep() { if [ ! -z $1 ]; then echo "Grepping for processes matching $1..." ps aux | grep -i $1 | grep -v grep; else echo "!! Need name to grep for"; fi }
    evenme · 2010-02-27 13:47:28 7
  • Suppose you have 11 marbles, 4 of which are red, the rest being blue. The marbles are indistinguishable, apart from colour. How many different ways are there to arrange the marbles in a line? And how many ways are there to arrange them so that no two red marbles are adjacent? There are simple mathematical solutions to these questions, but it's also possible to generate and count all possibilities directly on the command line, using little more than brace expansion, grep and wc! The answer to the question posed above is that there are 330 ways of arranging the marbles in a line, 70 of which have no two red marbles adjacent. See the sample output. To follow the call to marbles 11 4: after c=''; for i in $(seq $1); do c+='{b,r}'; done;, $c equals {b,r}{b,r}{b,r}{b,r}{b,r}{b,r}{b,r}{b,r}{b,r}{b,r}{b,r} After x=$(eval echo $c), and brace expansion, $x equals bbbbbbbbbbb bbbbbbbbbbr ... rrrrrrrrrrb rrrrrrrrrrr, which is all 2^11 = 2048 strings of 11 b's and r's. After p=''; for i in $(seq $2); do p+='b*r'; done;, $p equals b*rb*rb*rb*r Next, after y=$(grep -wo "${p}b*" Finally, grep -vc 'rr' Show Sample Output


    -4
    marbles () { c=''; for i in $(seq $1); do c+='{b,r}'; done; x=$(eval echo $c); p=''; for i in $(seq $2); do p+='b*r'; done; y=$(grep -wo "${p}b*" <<< $x); wc -l <<< "$y"; grep -vc 'rr' <<< "$y"; }
    quintic · 2010-08-27 23:04:33 3
  • Ok so it's rellay useless line and I sorry for that, furthermore that's nothing optimized at all... At the beginning I didn't managed by using netstat -p to print out which process was handling that open port 4444, I realize at the end I was not root and security restrictions applied ;p It's nevertheless a (good ?) way to see how ps(tree) works, as it acts exactly the same way by reading in /proc So for a specific port, this line returns the calling command line of every thread that handle the associated socket


    -5
    p=$(netstat -nate 2>/dev/null | awk '/LISTEN/ {gsub (/.*:/, "", $4); if ($4 == "4444") {print $8}}'); for i in $(ls /proc/|grep "^[1-9]"); do [[ $(ls -l /proc/$i/fd/|grep socket|sed -e 's|.*\[\(.*\)\]|\1|'|grep $p) ]] && cat /proc/$i/cmdline && echo; done
    j0rn · 2009-04-30 12:39:48 639
  • Tail is much faster than sed, awk because it doesn't check for regular expressions. Show Sample Output


    -5
    tail -n +<N> <file> | head -n 1
    qweqq · 2011-09-30 08:30:30 6

  • -6
    grep -A 3 -i "example" demo_text
    techie · 2013-05-09 08:20:31 5
  • I know how hard it is to find an old command running through all the files because you couldn't remember for your life what it was. Heres the solution!! Grep the history for it. depending on how old the command you can head or tail or if you wanted to search all because you cannot think how long ago it was then miss out the middle part of the command. This is a very easy and effective way to find that command you are looking for.


    -9
    cat .bash_history | tail -100 | grep {command}
    techie · 2013-04-10 10:40:52 8
  • Normally, if you just want to see directories you'd use brianmuckian's command 'ls -d *\', but I ran into problems trying to use that command in my script because there are often multiple directories per line. If you need to script something with directories and want to guarantee that there is only one entry per line, this is the fastest way i know Show Sample Output


    -10
    ls -l | grep ^d | sed 's:.*\ ::g'
    LinuxMan · 2011-08-06 23:52:46 9

  • -10
    grep -r <searchterm> *
    totti · 2012-02-08 11:48:08 7
  • David thanks for that grep inside! here is mine version: psgrep() { case ${1} in ( -E | -e ) local EXTENDED_REGEXP=1 shift 1 ;; *) local EXTENDED_REGEXP=0 ;; esac if [[ -z ${*} ]] then echo "psgrep - grep for process(es) by keyword" >&2 echo "Usage: psgrep [-E|-e] ... " >&2 echo "" >&2 echo "option [-E|-e] enables full extended regexp support" >&2 echo "without [-E|-e] plain strings are looked for" >&2 return 1 fi \ps -eo 'user,pid,pcpu,command' w | head -n1 local ARG='' if (( ${EXTENDED_REGEXP} == 0 )) then while (( ${#} > 0 )) do ARG="${1}" shift 1 local STRING=${ARG} local LENGTH=$(expr length ${STRING}) local FIRSCHAR=$(echo $(expr substr ${STRING} 1 1)) local REST=$(echo $(expr substr ${STRING} 2 ${LENGTH})) \ps -eo 'user,pid,pcpu,command' w | grep "[${FIRSCHAR}]${REST}" done else \ps -eo 'user,pid,pcpu,command' w | grep -iE "(${*})" fi }


    -10
    psgrep() ... func to long, please look under "description"
    Xk2c · 2015-01-01 02:58:48 8
  • ‹ First  < 15 16 17

What's this?

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

Share Your Commands


Check These Out

Command line calculator
works with fractions like 1/3.5

Updating the status on identi.ca using curl.
identica is an open source social networking and micro-blogging service. Based on Laconica, a micro-blogging software package built on the OpenMicroBlogging specification. http://identi.ca/

Automagically update grub.conf labels after installing a new kernel
I like to label my grub boot options with the correct kernel version/build. After building and installing a new kernel with "make install" I had to edit my grub.conf by hand. To avoid this, I've decided to write this little command line to: 1. read the version/build part of the filename to which the kernel symlinks point 2. replace the first label lines of grub.conf grub.conf label lines must be in this format: Latest [{name}-{version/build}] Old [{name}-{version/build}] only the {version/build} part is substituted. For instance: title Latest [GNU/Linux-2.6.31-gentoo-r10.201003] would turn to title Latest [GNU/Linux-2.6.32-gentoo-r7.201004]"

Cryptographically secure 32-bit RNG in strict ZSH
ZSH ships a couple random number generators via the $RANDOM environment variable and rand48() via zsh/mathfunc, but neither are cryptographically secure. This code produces a 32-bit random number suitable for cryptography. It's only dependency is /dev/urandom and it does not rely on any shell commands or 3rd party utilities. It assumes ZSH was compiled with 64-bit integer support.

list files recursively by size

Mount a partition from within a complete disk dump
Suppose you made a backup of your hard disk with dd: dd if=/dev/sda of=/mnt/disk/backup.img This command enables you to mount a partition from inside this image, so you can access your files directly. Substitute PARTITION=1 with the number of the partition you want to mount (returned from sfdisk -d yourfile.img).

Convert an image sequence into a video

Make a directory named with the current date
Create a directory named with the current date in ISO 8601 format (yyyy-mm-dd). Useful for storing backups by date. The --iso switch may only work with GNU date, can use format string argument for other date versions.

Show webcam output
Show the webcam output with mplayer.

Let's make screen and ssh-agent friends
When you start screen as `ssh-agent screen`, agent will die after detatch. If you don't want to take care about files when stored agent's pid/socket/etc, you have to use this command.


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: