Commands tagged awk (348)

  • It is not the installed size in files, but the size of RPM packages. Show Sample Output


    0
    rpm -qa --queryformat '%{SIZE}\n' | awk '{sum += $1} END {printf("Total size in packages = %4.1f GB\n", sum/1024**3)}'
    skytux · 2013-12-14 20:22:41 10

  • 0
    while [ 1 ] ;do ps aux|awk '{if ($8 ~ "D") print }'; sleep 1 ;done
    paulp · 2014-01-21 08:20:04 6
  • Grep can search files and directories recursively. Using the -Z option and xargs -0 you can get all results on one line with escaped spaces, suitable for other commands like rm. Show Sample Output


    0
    grep -Rl "pattern" files_or_dir
    N1nsun · 2014-04-06 18:18:07 7

  • 0
    grep URL ~/annex/.git/annex/webapp.html | tr -d '">' | awk -F= '{print $4 "=" $5}'
    kseistrup · 2014-04-20 08:46:37 8

  • 0
    ip route list 0/0
    thrix · 2014-06-09 16:07:38 7

  • 0
    mco ping | head -n -4 | awk '{print $1}' | sort
    mrwulf · 2014-06-24 18:20:16 7
  • Original command: cat "log" | grep "text to grep" | awk '{print $1}' | sort -n | uniq -c | sort -rn | head -n 100 This is a waste of multiple cats and greps, esp when awk is being used


    0
    awk '/text to grep/{print $1}' "log" | sort -n | uniq -c | sort -rn | head -n 100
    kln0thing · 2014-07-09 08:48:06 9
  • The AWK part of the code will "collate" the fields from 2nd to Nth field (this is to handle any svn directories that may have spaces in them - typical when working with code that is interchangeably used with windows environment - for example, documentation teams) - the output is passed to "ls -ld" - the -d option to ls will tell ls to handle directories itself, rather than do ls on the directory. The '-p' option is just for pretty printing directories, links and executables (for added readability). Finally, the entire "constructed" command will be passed onto sh for shell execution. Show Sample Output


    0
    svn status | awk -F" " '{ for (i=2; i<=NF; i++) print "ls -ld \""$i"\""}' | sh
    kln0thing · 2014-07-09 09:41:24 14
  • Gets the Hardware UUID of the current machine using system_profiler. Show Sample Output


    0
    system_profiler SPHardwareDataType | awk '/UUID/ { print $3; }'
    thealanberman · 2014-07-25 06:54:40 8
  • This command makes a small graph with the histogram of size blocks (5MB in this example), not individual files. Fine tune the 4+5*int($1/5) block for your own size jumps : jump-1+jump*($1/jump) Also in the hist=hist-5 part, tune for bigger or smaller graphs Show Sample Output


    0
    du -sm *| sort -nr | awk '{ size=4+5*int($1/5); a[size]++ }; END { print "size(from->to) number graph"; for(i in a){ printf("%d %d ",i,a[i]) ; hist=a[i]; while(hist>0){printf("#") ; hist=hist-5} ; printf("\n")}}'
    higuita · 2014-08-19 14:43:20 8
  • Caution: distructive overwrite of filenames Useful for concatenating pdfs in date order using pdftk


    0
    find . -name "*.pdf" -print0 | xargs -r0 stat -c %y\ %n | sort|awk '{print $4}'|gawk 'BEGIN{ a=1 }{ printf "mv %s %04d.pdf\n", $0, a++ }' | bash
    Randy_Legault · 2014-09-23 06:40:45 9
  • Its possible to user a simple regex to extract de username from the finger command. The final echo its optional, just for remove the initial space Show Sample Output


    0
    finger $(whoami) | egrep -o 'Name: [a-zA-Z0-9 ]{1,}' | cut -d ':' -f 2 | xargs echo
    swebber · 2014-09-24 01:22:07 9
  • Given a hosts list, ssh one by one and echo its name only if 'processname' is not running. Show Sample Output


    0
    for i in `cat hosts_list`; do RES=`ssh myusername@${i} "ps -ef " |awk '/[p]rocessname/ {print $2}'`; test "x${RES}" = "x" && echo $i; done
    arlequin · 2014-10-03 14:57:54 9
  • This is useful as a git hook to print out the directories that had files changed on a commit. Each directory is its own package. Show Sample Output


    0
    git log -n 1 --name-only --pretty=oneline | awk -F/ 'NR>=2 {seen[$1]}; END {for (d in seen); print d}'
    Romster · 2014-12-13 10:21:46 9
  • The sample output shows each record/row with the last field zero-padded to 26 digits. For testing, I used (L)ine and field/column numbers.... Line 4, field2 = L42, etc up to the last field where I just used line numbers X 4. I had some whitespace-delimited files with variable-length records/rows (having 4 - 5 fields/columns) which required reformatting by zero-padding the last field to 26 digits. This requires setting NF (Not $NF) as an awk variable, with a simple conditional that assumes that any line where (N)umber of (F)ields does NOT equal 4 has a NF of 5. If needed, more conditional checks can be added, and the "NF" changed to any field ($1, $5, etc). Show Sample Output


    0
    awk '{var = sprintf(NF); if (var == 4) printf "%s %s %s %026d\n" , $1,$2,$3,$4; else printf "%s %s %s %s %026d\n" , $1,$2,$3,$4,$5}' yourfilegoes.here >> yournewfilegoes.here
    genatomics · 2014-12-20 02:53:35 8
  • Use this command to watch apache access logs in real time to see what pages are getting hit. Show Sample Output


    0
    tail -f access_log | awk '{print $1 , $12}'
    tyzbit · 2014-12-24 14:15:52 10

  • 0
    sudo du -kx / |sort -n| awk '{print $1/(1000*1000) " G" ,$2}'
    umiyosh · 2015-01-05 04:49:24 8
  • OSX users as well as linux users with copy/paste buffer commands can remove duplicate items from their copy buffer with this command. I use this often when I have to copy a long list of items that I didn't generate, but I need to paste elsewhere in a list that's unique. If retaining the original order of lines isn't important to you, use the following command which is easier to remember. pbpaste | sort | uniq | pbcopy


    0
    pbpaste | awk ' !x[$0]++' | pbcopy
    dmengelt · 2015-02-05 19:38:38 12
  • us lsof, grep for any pid matching a given name such as "node". Show Sample Output


    0
    lsof -i -n -P | grep -e "$(ps aux | grep node | grep -v grep | awk -F' ' '{print $2}' | xargs | awk -F' ' '{str = $1; for(i = 2; i < NF; i++) {str = str "\\|" $i} print str}')"
    hochmeister · 2015-02-14 23:24:00 10
  • Replace grep | sed with single awk script.


    0
    watch -n10 -d sh -c 'sensors | awk '\''/:.*RPM/ { sub("[^:]*:","") ; print $1 }'\'
    my_username · 2015-04-29 16:50:28 10

  • 0
    pgrep -f /usr/sbin/httpd | awk '{print"-p " $1}' | xargs strace
    savagemike · 2015-06-10 22:55:35 12
  • Removes directories which are less than 1028KB total. This works for systems where blank directories are 4KB. If a directory contains 1 MB (1024KB) or less, it will remove the directory using a path relative to the directory where the command was initially executed (safer than some other options I found). Adjust the 1028 value for your needs. It would be helpful to test the results before proceeding with the removal. Simply run all but the last two commands to see a list of what will be removed: du | awk '{if($1<1028)print;}' | cut -d $'\t' -f 2- If you're unsure what size a blank folder is, test it like this: mkdir test; du test; rmdir test


    0
    du | awk '{if($1<1028)print;}' | cut -d $'\t' -f 2- | tr "\n" "\0" | xargs -0 rm -rf
    i814u2 · 2015-06-25 16:00:48 10
  • Don't want to open up an editor just to view a bunch of XML files in an easy to read format? Now you can do it from the comfort of your own command line! :-) This creates a new function, xmlpager, which shows an XML file in its entirety, but with the actual content (non-tag text) highlighted. It does this by setting the foreground to color #4 (red) after every tag and resets it before the next tag. (Hint: try `tput bold` as an alternative). I use 'xmlindent' to neatly reflow and indent the text, but, of course, that's optional. If you don't have xmlindent, just replace it with 'cat'. Additionally, this example shows piping into the optional 'less' pager; note the -r option which allows raw escape codes to be passed to the terminal. Show Sample Output


    0
    xmlpager() { xmlindent "$@" | awk '{gsub(">",">'`tput setf 4`'"); gsub("<","'`tput sgr0`'<"); print;} END {print "'`tput sgr0`'"}' | less -r; }
    hackerb9 · 2015-07-12 09:22:10 11

  • 0
    eval `cli53 list |grep Name | sed "s/\.$//g" | awk '{printf("echo %s; cli53 export %s > %s;\n", $2, $2, $2);}'`
    cfb · 2015-07-21 14:16:30 10
  • sample.csv: 79.36,94.93,10.92,27.33,95.90 3.57, 20.80,67.06,2.16, 79.23 48.45,27.95,7.66, 56.71,59.97 69.02,89.59,33.88,42.73,22.60 10.15,44.86,70.86,98.45,22.23 Show Sample Output


    0
    awk -F , '{for(i=1;i<=NF;i++)t[NR]+=$i;$0=t[NR]}1' sample.csv
    YardApe · 2015-09-05 13:29:18 13
  • ‹ First  < 9 10 11 12 13 >  Last ›

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

Find usb device in realtime
Using this command you can track a moment when usb device was attached.

Realtime apache hits per second
Change the cut range for hits per 10 sec, minute and so on... Grep can be used to filter on url or source IP.

Synchronize date and time with a server over ssh
Set Remote Server Date using Local Server Time (push)

Copy from host 1 to host 2 through your host

Display rows and columns of random numbers with awk
Displays six rows and five columns of random numbers between 0 and 1. If you need only one column, you can dispense with the "for" loop.

See system users

Find usb device in realtime
Using this command you can track a moment when usb device was attached.

Convert all .flac from a folder subtree in 192Kb mp3
find . -type f -iname '*.flac' # searches from the current folder recursively for .flac audio files | # the output (a .flac audio files with relative path from ./ ) is piped to while read FILE; do FILENAME="${FILE%.*}"; flac -cd "$FILE" | lame -b 192 - "${FILENAME}.mp3"; done # for each line on the list: # FILE gets the file with .flac extension and relative path # FILENAME gets FILE without the .flac extension # run flac for that FILE with output piped to lame conversion to mp3 using 192Kb bitrate

Get the full path of a bash script's Git repository head.
Rather than complicated and fragile paths relative to a script like "../../other", this command will retrieve the full path of the file's repository head. Safe with spaces in directory names. Works within a symlinked directory. Broken down: $cd "$(dirname "${BASH_SOURCE[0]}")" temporarily changes directories within this expansion. Double quoted "$(dirname" and ")" with unquoted ${BASH_SOURCE[0]} allows spaces in the path. $git rev-parse --show-toplevel gets the full path of the repository head of the current working directory, which was temporarily changed by the "cd".

Define Google Chrome urpmi media source for Mandriva/Mageia (works for both 32-bit and 64-bit systems)
This command adds a urpmi media source called "google-chrome" to the urpmi configuration on Mandriva or Mageia. Needs to be run as root. We specify the option "--update" so that when Google provides a newer version of Google Chrome web browser in their download system then running a system update (eg: "urpmi --auto-update") will result in our copy of Google Chrome getting updated (along with any other Mandriva/Mageia pending updates). To install Google Chrome from this source, use: urpmi google-chrome-stable #install Google chrome web browser


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: