Commands using sed (1,319)

  • return external ip

    alternative using 'host' Show Sample Output


    -1
    host -t a dartsclink.com | sed 's/.*has address //'
    dartsclink · 2009-08-14 16:11:18 13
  • Recreate all initrd files

    Do this with caution.


    -1
    for kern in $(grep "initrd " /boot/grub/grub.conf|grep -v ^#|cut -f 2- -d-|sed -e 's/\.img//g'); do mkinitrd -v -f /boot/initrd-$kern.img $kern; done
    oernii2 · 2009-08-19 09:53:29 502
  • Reverse DNS lookups

    Reverse DNS lookups, from a file with list of IP's, here the file is called lookups.txt


    -1
    sed 's/\([0-9]*\)\.\([0-9]*\)\.\([0-9]*\)\.\([0-9]*\).in-addr.arpa domain name pointer\(.*\)\./\4.\3.\2.\1\5/' \ lookups.txt
    hemanth · 2009-08-22 09:37:20 4
  • dig this


    -1
    for dnsREC in $(curl -s http://www.iana.org/assignments/dns-parameters |grep -Eo ^[A-Z\.]+\ |sed 's/TYPE//'); do echo -n "$dnsREC " && dig +short $dnsREC IANA.ORG; done
    commandlinefu · 2009-09-01 03:11:18 3
  • search for a file in PATH

    search argument in PATH accept grep expressions without args, list all binaries found in PATH Show Sample Output


    -1
    function sepath { echo $PATH |tr ":" "\n" |sort -u |while read L ; do cd "$L" 2>/dev/null && find . \( ! -name . -prune \) \( -type f -o -type l \) 2>/dev/null |sed "s@^\./@@" |egrep -i "${*}" |sed "s@^@$L/@" ; done ; }
    mobidyc · 2009-09-11 15:03:22 5
  • Print all members of US House of Representatives

    Uses curl to download page of membership of US Congress. Use sed to strip HTML then perl to print a line starting with two tabs (a line with a representative) Show Sample Output


    -1
    curl "http://www.house.gov/house/MemberWWW.shtml" 2>/dev/null | sed -e :a -e 's/<[^>]*>//g;/</N;//ba' | perl -nle 's/^\t\t(.*$)/ $1/ and print;'
    drewk · 2009-09-24 23:37:36 15
  • get Hong Kong weather infomation from HK Observatory

    From Hong Kong Observatory wap site ;) Show Sample Output


    -1
    wget -q -O - 'http://wap.weather.gov.hk/' | sed -r 's/<[^>]+>//g;/^UV/q' | grep -v '^$'
    twfcc · 2009-09-25 02:21:05 6
  • get Hong Kong weather infomation from HK Observatory

    "get Hong Kong weather infomation from HK Observatory From Hong Kong Observatory wap site ;)" other one showed alot of blank lines for me Show Sample Output


    -1
    wget -q -O - 'http://wap.weather.gov.hk/' | sed -r 's/<[^>]+>//g;/^UV/q' | tail -n4
    dakunesu · 2009-09-25 02:36:46 3
  • Prints per-line contribution per author for a GIT repository

    You'll run into trouble if you have files w/ missing newlines at the end. I tried to use PAGER='sed \$q' git blame and even PAGER='sed \$q' git -p blame to force a newline at the end, but as soon as the output is redirected, git seems to ignore the pager.


    -1
    git ls-files | while read i; do git blame $i | sed -e 's/^[^(]*(//' -e 's/^\([^[:digit:]]*\)[[:space:]]\+[[:digit:]].*/\1/'; done | sort | uniq -ic | sort -nr
    pipping · 2009-10-25 09:40:01 4
  • get ^DJI


    -1
    getdji (){local url sedcmd;url='http://finance.yahoo.com/q?d=t&s=^DJI';sedcmd='/(DJI:.*)/,/Day.*/!d;s/^ *//g;';sedcmd="$sedcmd/Change:/s/Down / -/;/Change:/s/Up / +/;";sedcmd="$sedcmd/Open:/s//& /";lynx -dump "$url" | sed "$sedcmd"; }
    twfcc · 2009-10-26 09:00:18 5
  • Get column names in MySQL

    Useful when you need to write e.g. an INSERT for a table with a large number of columns. This command will retrieve the column names and comma-separate them ready for INSERT INTO(...), removing the last comma.


    -1
    mysql -u <user> --password=<password> -e "SHOW COLUMNS FROM <table>" <database> | awk '{print $1}' | tr "\n" "," | sed 's/,$//g'
    maxmanders · 2009-10-29 13:42:17 3
  • Get your external IP address with the best commandlinefu.com command

    This command uses the top voted "Get your external IP" command from commandlinefu.com to get your external IP address. Use this and you will always be using the communities favourite command. This is a tongue-in-cheek entry and not recommended for actual usage.


    -1
    eval $(curl -s http://www.commandlinefu.com/commands/matching/external/ZXh0ZXJuYWw=/sort-by-votes/plaintext|sed -n '/^# Get your external IP address$/{n;p;q}')
    jgc · 2009-11-04 16:58:31 6
  • Show Network IP and Subnet

    same thing as the other


    -1
    ipcalc $(ifconfig eth0 | grep "inet addr:" | cut -d':' -f2,4 | sed 's/.+Bcast:/\//g') | awk '/Network/ { print $2 } '
    solarislackware · 2009-12-05 15:00:32 3
  • Extract all urls from the last firefox sessionstore.js file used.


    -1
    sed -e 's/{"url":/\n&/g' ~/.mozilla/firefox/*/sessionstore.js | cut -d\" -f4
    cfajohnson · 2009-12-10 04:31:31 4
  • prints line numbers

    There's too many options to number, My curiosity has forced me to make it using only sed. Maybe useful... or not... :-S


    -1
    sed '/./=' infile | sed '/^/N; s/\n/ /'
    glaudiston · 2009-12-10 16:24:56 6
  • Print line numbers

    Print out contents of file with line numbers. This version will print a number for every line, and separates the numbering from the line with a tab. Show Sample Output


    -1
    sed = <file> | sed 'N;s/\n/\t/'
    jgc · 2009-12-11 14:39:14 3
  • Add a line to crontab using sed

    I needed to add a line to my crontab from within a script and didn't want to have to write my own temporary file. You may find you need to reload the crond after this to make the change take effect. e.g.: if [ -x /sbin/service ] then /sbin/service crond reload else CRON_PID=`ps -furoot | awk '/[^a-z]cron(d)?$/{print $2}'` if [ -n "$CRON_PID" ] then kill -HUP $CRON_PID fi fi The reason I had CRON_HOUR and CRON_MINS instead of numbers is that I wanted to generate a random time between midnight & 6AM to run the job, which I did with: CRON_HOUR=`/usr/bin/perl -e 'printf "%02d\n", int(rand(6))'` CRON_MINS=`/usr/bin/perl -e 'printf "%02d\n", int(rand(60));'`


    -1
    crontab -l | sed -e '$G;$s-$-'"$CRON_MINS $CRON_HOUR"' * * * /usr/bin/command >/dev/null 2>&1-' | crontab -
    JohnGH · 2010-01-07 11:00:05 6
  • wget, tar xzvf, cd, ls

    Combines a few repetitive tasks when compiling source code. Especially useful when a hypen in a file-name breaks tab completion. 1.) wget source.tar.gz 2.) tar xzvf source.tar.gz 3.) cd source 4.) ls From there you can run ./configure, make and etc. Show Sample Output


    -1
    wtzc () { wget "$@"; foo=`echo "$@" | sed 's:.*/::'`; tar xzvf $foo; blah=`echo $foo | sed 's:,*/::'`; bar=`echo $blah | sed -e 's/\(.*\)\..*/\1/' -e 's/\(.*\)\..*/\1/'`; cd $bar; ls; }
    oshazard · 2010-01-17 11:25:47 3
  • keep an eye on system load changes


    -1
    watch -n 7 -d 'uptime | sed s/.*users?, //'
    matthewbauer · 2010-01-17 18:45:52 3
  • Rename all the files in the current directory into their sha1sum


    -1
    find . -maxdepth 1 -type f| xargs sha1sum | sed 's/^\(\w*\)\s*\(.*\)/\2 \1/' | while read LINE; do mv $LINE; done
    foremire · 2010-01-25 20:21:01 11
  • Delete all but latest file in a directory

    xargs deals badly with special characters (such as space, ' and "). To see the problem try this: touch important_file touch 'not important_file' ls not* | xargs rm Parallel https://savannah.nongnu.org/projects/parallel/ does not have this problem.


    -1
    ls -t1 | sed 1d | parallel -X rm
    unixmonkey8046 · 2010-01-28 12:28:18 3
  • Weather on the Command line

    Will return temperature in Fahrenheit of a location (New York City in example). Uses a Google API. Show Sample Output


    -1
    curl -s "http://www.google.com/ig/api?weather=New%20York" | sed 's|.*<temp_f data="\([^"]*\)"/>.*|\1|'
    matthewbauer · 2010-02-08 23:06:48 5
  • Get Google Reader unread count

    Get Google Reader unread count from the command line. You'll have to define your auth token with $auth Or use: curl -s -H "Authorization: GoogleLogin auth=$(curl -sd "Email=$email&Passwd=$password&service=reader" https://www.google.com/accounts/ClientLogin | grep Auth | sed 's/Auth=\(.*\)/\1/')" "http://www.google.com/reader/api/0/unread-count?output=json" | tr '{' '\n' | sed 's/.*"count":\([0-9]*\),".*/\1/' | grep -E ^[0-9]+$ | tr '\n' '+' | sed 's/\(.*\)+/\1\n/' | bc Show Sample Output


    -1
    curl -s -H "Authorization: GoogleLogin auth=$auth" "http://www.google.com/reader/api/0/unread-count?output=json" | tr '{' '\n' | sed 's/.*"count":\([0-9]*\),".*/\1/' | grep -E ^[0-9]+$ | tr '\n' '+' | sed 's/\(.*\)+/\1\n/' | bc
    matthewbauer · 2010-02-11 00:42:57 7
  • Bold matching string without skipping others


    -1
    sed 's/pattern/^[[1m&^[[0m/g'
    rmcb · 2010-02-12 14:05:34 3
  • Move itens from subdirectories to current directory

    You WILL have problems if the files have the same name. Use cases: consolidate music library and unify photos (especially if your camera separates images by dates). After running the command and verifying if there was no name issues, you can use ls -d */ | sed -e 's/^/\"/g' -e 's/$/\"/g' | xargs rm -r to remove now empty subdirectories.


    -1
    ls -d */* | sed -e 's/^/\"/g' -e 's/$/\"/g' | xargs mv -t $(pwd)
    leovailati · 2010-03-01 23:43:26 6
  • ‹ First  < 42 43 44 45 46 >  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

Lists unambigously names of all xml elements used in files in current directory
This set of commands was very convenient for me when I was preparing some xml files for typesetting a book. I wanted to check what styles I had to prepare but coudn't remember all tags that I used. This one saved me from error-prone browsing of all my files. It should be also useful if one tries to process xml files with xsl, when using own xml application.

Find files that have been modified on your system in the past 60 minutes
Useful mainly for debugging or troubleshooting an application or system, such as X11, Apache, Bind, DHCP and others. Another useful switch that can be combined with -mmin, -mtime and so forth is -daystart. For example, to find files that were modified in the /etc directory only yesterday: $ sudo find /etc -daystart -mtime 1 -type f

Which processes are listening on a specific port (e.g. port 80)
swap out "80" for your port of interest. Can use port number or named ports e.g. "http"

recursively change file name from uppercase to lowercase (or viceversa)
easier way to recursively change files to lowercase using rename instead

how many pages will my text files print on?
This gives a very rough estimate of how many pages your text files will print on. Assumes 60 lines per page, and does not take long lines into account.

Find and display most recent files using find and perl
This pipeline will find, sort and display all files based on mtime. This could be done with find | xargs, but the find | xargs pipeline will not produce correct results if the results of find are greater than xargs command line buffer. If the xargs buffer fills, xargs processes the find results in more than one batch which is not compatible with sorting. Note the "-print0" on find and "-0" switch for perl. This is the equivalent of using xargs. Don't you love perl? Note that this pipeline can be easily modified to any data produced by perl's stat operator. eg, you could sort on size, hard links, creation time, etc. Look at stat and just change the '9' to what you want. Changing the '9' to a '7' for example will sort by file size. A '3' sorts by number of links.... Use head and tail at the end of the pipeline to get oldest files or most recent. Use awk or perl -wnla for further processing. Since there is a tab between the two fields, it is very easy to process.

list files recursively by size

check open ports without netstat or lsof

Scans for open ports using telnet

Efficient count files in directory (no recursion)
$ time perl -e 'if(opendir D,"."){@a=readdir D;print $#a - 1,"\n"}' 205413 real 0m0.497s user 0m0.220s sys 0m0.268s $ time { ls |wc -l; } 205413 real 0m3.776s user 0m3.340s sys 0m0.424s ********* ** EDIT: turns out this perl liner is mostly masturbation. this is slightly faster: $ find . -maxdepth 1 | wc -l sh-3.2$ time { find . -maxdepth 1|wc -l; } 205414 real 0m0.456s user 0m0.116s sys 0m0.328s ** EDIT: now a slightly faster perl version $ perl -e 'if(opendir D,"."){++$c foreach readdir D}print $c-1,"\n"' sh-3.2$ time perl -e 'if(opendir D,"."){++$c foreach readdir D}print $c-1,"\n"' 205414 real 0m0.415s user 0m0.176s sys 0m0.232s

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: