Commands using perl (369)

  • modify a file in place with perl

    changes THIS to THAT in all files matching fileglob* without using secondary files


    2
    perl -pi -e 's/THIS/THAT/g' fileglob*
    elofland · 2009-02-05 19:19:52 60
  • The top ten commands you use

    Show Sample Output


    2
    perl -pe 's/.+;//' ~/.zsh_history | sort | uniq -c | sort -r|head -10
    aoiaoi · 2009-02-06 15:24:32 176
  • re-assign line numbers

    Show Sample Output


    2
    perl -pe 's/\d+/++$n/e' file.txt
    mikeda · 2009-02-17 14:51:31 6
  • Comma insertions

    Insert a comma where necessary when counting large numbers. I needed to separate huge amounts of packets and after 12+ hours of looking in a terminal, I wanted it in readable form. Show Sample Output


    2
    perl -pe '$_=reverse;s/\d{3}(?=\d)(?!.*?\.)/$&,/g;$_=reverse'
    sil · 2009-02-18 16:34:18 15
  • Recursive Search and Replace

    Finds the string in every file in an entire directory and all its subdirectories and replaces it with a new string. Especially useful when changing a machine's IP address or hostname - run it on /etc.


    2
    perl -pi -e's/<what to find>/<what to replace it with>/g' `grep -Rl <what to find> /<dir>/*`
    adampbell · 2009-02-26 19:14:39 8
  • make a list of movies(.m3u).


    2
    find $HOME -type f -print | perl -wnlaF'/' -e 'BEGIN{ print "#EXTM3U"; } /.+\.wmv$|.+\.mpg$|.+\.vob$/i and print "#EXTINF:$F[-1]\nfile://$&";' > movies.m3u
    ishiduca · 2009-02-28 12:17:41 4
  • Prints any IP out of a file


    2
    perl -ne 'while (/([0-9]+\.){3}[0-9]+/g) {print "$&\n"};' file.txt
    P17 · 2009-04-01 13:49:46 6
  • Connect to all running screen instances

    There was another line that was dependent on having un-named screen sessions. This just wouldn't do. This one works no matter what the name is. A possible improvement would be removing the perl dependence, but that doesn't effect me.


    2
    for i in `screen -ls | perl -ne'if(/^\s+\d+\.([^\s]+)/){print $1, " "}'`; do gnome-terminal -e "screen -x $i"; done
    hank · 2009-04-25 22:39:24 7
  • Replace Every occurrence of a word in a file

    Replace 'this' with 'that'


    2
    perl -p -i -e 's/this/that/g' filename
    rader5 · 2009-08-09 20:53:50 8
  • show the real times iso of epochs for a given column

    When you have one of those (log)files that only has epoch for time (since no one will ever look at them as a date) this is a way to get the human readable date/time and do further inspection. Mostly perl-fu :-/


    2
    perl -F' ' -MDate::Format -pale 'substr($_, index($_, $F[1]), length($F[1]), time2str("%C", $F[1]))' file.log
    coffeeaddict_nl · 2009-08-13 13:57:33 4
  • convert unixtime to human-readable


    2
    perl -e 'print scalar(gmtime(1234567890)), "\n"'
    andrew112358 · 2009-08-25 15:00:52 4
  • Go get those photos from a Picasa album


    2
    wget 'link of a Picasa WebAlbum' -O - |perl -e'while(<>){while(s/"media":{"content":\[{"url":"(.+?\.JPG)//){print "$1\n"}}' |wget -w1 -i -
    aciancone · 2009-09-27 14:36:27 7
  • A command line calculator in Perl

    Once I wrote a command line calculator program in C, then I found this... and added to it a bit. For ease of use I normally use this in a tiny Perl program (which I call pc for 'Perl Calculator') #!/usr/bin/perl -w die "Usage: $0 MATHS\n" unless(@ARGV);for(@ARGV){s/x/*/g;s/v/sqrt /g;s/\^/**/g}; print eval(join('',@ARGV)),$/; It handles square roots, power, modulus: pc 1+2 (1 plus 2) 3 pc 3x4 (3 times 4) 12 pc 5^6 (5 to the power of 6) 15625 pc v 49 ( square root of 49 ) 7 pc 12/3 (12 divided by 3) 4 pc 19%4 (19 modulus 4) 3 (you can string maths together too) pc 10 x 10 x 10 1000 pc 10 + 10 + 10 / 2 25 pc 7 x v49 49 Show Sample Output


    2
    perl -e 'for(@ARGV){s/x/*/g;s/v/sqrt /g;s/\^/**/g};print eval(join("",@ARGV)),$/;'
    JohnGH · 2009-12-21 21:03:27 9
  • Recursively grep for string and format output for vi(m)

    This is a big time saver for me. I often grep source code and need to edit the findings. A single highlight of the mouse and middle mouse click (in gnome terminal) and I'm editing the exact line I just found. The color highlighting helps interpret the data.


    2
    mgc() { grep --exclude=cscope* --color=always -rni $1 . |perl -pi -e 's/:/ +/' |perl -pi -e 's/^(.+)$/vi $1/g' |perl -pi -e 's/:/ /'; }
    stinkerweed999 · 2010-01-26 17:00:01 4
  • Get your commandlinefu points (upvotes - downvotes)

    Like command #4845, prints score, number of entries, and average score.


    2
    username=bartonski;curl -s http://www.commandlinefu.com/commands/by/$username/json|perl -e 'BEGIN{$s=0;$n=0};END{print "Score: $s\nEntries: $n\nMean: ";printf "%3.2f\n",$s/$n}' -0173 -nae 'foreach $f (@F){if($f =~ /"votes":"(-*\d+)"/){$s += $1; $n++;}}'
    bartonski · 2010-02-16 01:03:29 4
  • Update twitter with Perl

    Requires Net::Twitter. Just replace the double quoted strings with the appropriate info.


    2
    perl -MNet::Twitter -e '$nt = Net::Twitter->new(traits => [qw/API::REST/], username => "YOUR USERNAME", password => "YOUR PASSWORD"); $ud = $nt->update("YOUR TWEET");'
    dbbolton · 2010-06-16 19:46:05 4
  • diff files while disregarding indentation and trailing white space

    **NOTE** Tekhne's alternative is much more succinct and its output conforms to the files actual contents rather than with white space removed My command on the other hand uses bash process substitution (and "Minimal" Perl), instead of files, to first remove leading and trailing white space from lines, before diff'ing the streams. Very useful when differences in indentation, such as in programming source code files, may be irrelevant Show Sample Output


    2
    diff <(perl -wpl -e '$_ =~ s/^\s+|\s+$//g ;' file1) <(perl -wpl -e '$_ =~ s/^\s+|\s+$//g ;' file2)
    jemptymethod · 2010-10-06 19:14:42 5
  • Decode base64-encoded file in one line of Perl

    If you are in an environment where you don't have the base64 executable or MIME tools available, this can be very handy for salvaging email attachments when the headers are mangled but the encoded document itself is intact.


    2
    perl -MMIME::Base64 -ne 'print decode_base64($_)' < file.txt > out
    dmmst19 · 2010-12-13 23:35:20 3
  • Find C/C++ source code comments

    This is a naive way of finding source code comments in source code files that use C-like comments: // and /*...*/


    2
    perl -e 'my $in_comment = 0; while (<>) { $in_comment = 1 if m{\Q/*\E}; print if $in_comment; $in_comment = 0 if m{\Q*/\E}; }' *.cpp
    doherty · 2011-07-08 00:17:27 4
  • Find default gateway (proper at ppp connections too)


    2
    route -n | perl -ne '$ANY="0.0.0.0"; /^$ANY/ and split /\s+/ and print "Gateway to the World: ",($_[1]!=$ANY)?$_[1]:(`ip address show $_[$#_]`=~/peer ([0-9\.]+)/ and $1),", via $_[$#_].\n"'
    bandie91 · 2011-09-13 08:05:58 5
  • FizzBuzz One-liner


    2
    perl -le 'print$_%3?$_%5?$_:"Buzz":$_%5?"Fizz":"FizzBuzz"for 1..100'
    depesz · 2012-01-10 13:10:30 5
  • Remove a range of lines from a file

    Show Sample Output


    2
    perl -i -ne 'print if $. == 3..5' <filename>
    bashrc · 2012-01-27 23:15:10 4
  • Apache CLF access log format to CSV converter

    - excel date compatible with a separate hour field - added a fixed 1 for easier request counter aggregation - split URL in directory, filename, fileext, query - used with tomcat valve with response bytes replaced by elapsed time Show Sample Output


    2
    #(see sample) $ cat x | perl -pe 'BEGIN{ print "TIME;...\n"; } s!(\S+) - (\S+) - \[(\d\d)/(\S\S\S)/(\S+):(\d\d):(\d\d:\d\d) \S+\] "(\S+) (.*/)(\S+)(?:\.([^?]*)(\?\S*)?) HTTP/\S+" (\d+) (\S+)!$3-$4-$5 $6:$7;$6;$2;$1;$8;$13;1;$14;$11;$10;$9;$12;!' > x.csv
    hute37 · 2012-02-10 16:58:50 3
  • Text graphing ping output filter

    Nasty perl one-liner that provides a sparkline of ping times. If you want a different history than the last 30, just put that value in. It (ab)uses unicode to draw the bars, inspired by https://github.com/joemiller/spark-ping . It's not the most bug-free piece of code, but what it lacks in robustness it makes up for in capability. :) If anyone has any ideas on how to make it more compact or better, I'd love to hear them. I included a ping to google in the command just as an example (and burned up 10 chars doing it!). You should use it with: $ ping example.com | $SPARKLINE_PING_COMMAND Show Sample Output


    2
    ping g.co|perl -ne'$|=/e=(\S+)/||next;(push@_,$1)>30&&shift@_;print"\r",(map{"\xe2\x96".chr(128+7*$_/(sort{$b<=>$a}@_)[0])." "}@_),"$1ms"'
    bartgrantham · 2012-07-06 22:42:06 3
  • MPD + Digitally Imported

    1.- Enter into the playlist path. 2.- Run the command. 3.- Playlists created!


    2
    wget -q -O - http://listen.di.fm/public2 | sed 's/},{/\n/g' | perl -n -e '/"key":"([^"]*)".*"playlist":"([^"]*)"/; print "$1\n"; system("wget -q -O - $2 | grep -E '^File' | cut -d= -f2 > di_$1.m3u")'
    Zort · 2013-02-20 03:37:41 9
  • ‹ First  < 2 3 4 5 6 >  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

recursive search and replace old with new string, inside files
Using -Z with grep and -0 with xargs handles file names with spaces and special characters.

Use find to get around Argument list too long problem
Can be used for other commands as well, replace rm with ls. It is easy to make this shorter but if the filenames involved have spaces, you will need to do use find's "-print0" option in conjunction with xargs's "-0" option. Otherwise the shell that xargs uses to execute the "rm" command line will treat the space as a token separator, thereby treating the name as two (or more) names.

Convert CSV to JSON
Replace 'csv_file.csv' with your filename.

Clear current session history (bash)

Sort all running processes by their memory & CPU usage
you can also pipe it to "tail" command to show 10 most memory using processes.

Backup with versioning

BourneShell: Go to previous directory
cd - would return to the previous directory of your cd command. NB: previous dir is always stored in $OLDPWD variable.

Copy a file using dd and watch its progress
This is a more accurate way to watch the progress of a dd process. The $DDPID=$! is needed so that you don't get the PID of the sleep. The sleep 1 is needed because in my testing at least, if you run kill -USR1 against dd too quickly, it will kill it off instead of display the status. So you need to wait a second, probably so that it can configure itself to trap the USR1 signal.

Create arbitrary big file full of zeroes but done in a second
If you want to create fast a very big file for testing purposes and you do not care about its content, then you can use this command to create a file of arbitrary size within less than a second. Content of file will be all zero bytes. The trick is that the content is just not written to the disk, instead the space for it is somehow reserved on operating system level and file system level. It would be filled when first accessed/written (not sure about the mechanism that lies below, but it makes the file creation super fast). Instead of '1G' as in the example, you could use other modifiers like 200K for kilobytes (1024 bytes), 500M for megabytes (1024 * 1024 bytes), 20G for Gigabytes (1024*1024*1024 bytes), 30T for Terabytes (1024^4 bytes). Also P for Penta, etc... Command tested under Linux.

Convert all Flac in a directory to Mp3 using maximum quality variable bitrate

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: