making it "sound" more "natural" language like -- additionally sorting the longest words alphabetically:

this approach is using:

* to get at all lines of input

* post-"for" structure

* short-circuit-or in sort: if the lengths are the same, then sort alphabetically otherwise don't even evaluate the right hand side of the or

* -C sets all input and ouput channels to utf8

dirrrty: use -p to chomp automatically, substitute all newlines away and then replace the "---" by a newline ? bingo!

s/// => s/// is just a cooler way to write s///, s/// which is just the small brother of s///; s/// (comma is an operator!)

have fun!

the output of svn log is annoying to grep, since it spreads the useful info over multiple lines. This compacts the output down to one line so eg you can grep for a comment and see the rev, date & committer straight away.

Updated: MUCH shorter, easier to remember. Now it just replaces newlines with spaces, except on '---' lines.

Require: perl -MCPAN -e 'install Acme::POE::Tree'

From: http://www.catonmat.net/blog/christmas-tree-in-the-shell/

with a semicolon text file map, apply multiple replace to a single file

from http://stackoverflow.com/questions/1030787/multiline-search-replace-with-perl

added greedy trick in wildcard match (.*?) from

http://www.troubleshooters.com/codecorn/littperl/perlreg.htm#Greedy

Calls the POSIX strerror() function to look up the meaning of integer ERRNOs set by some functions.

say only processes a complete file, at eof, so following a file isn't possible. Quick and dirty perl oneliner to feed each line from the tail -f to say. Yes, expensive to lauch a new process each line.

This little ditty was prompted by a discussion on how horrible it is to use VoiceOver on ncurses programs such as irssi.

Find which directories on your system contain a lot of files.

Edit: much shorter and betterer with -n switch.

PmWiki stores wiki pages as Group.Name. Simply split the directory listing and count frequency of group occurances.

Perl version - just for completeness sake ;)

This version now adds a header with consecutive numbering.

Before running, do:

curl -sO http://world.std.com/%7Ereinhold/diceware.wordlist.asc

Lets say you have a file with the following layout:

LINUX,DIR,FILE1,FILE2,FILE3

You want the file to look like this:

LINUX,DIR,FILE1

LINUX,DIR,FILE2

LINUX,DIR,FILE3

This perl command does it for you.

I use this in my bashrc to expand hosts defined in ~/.ssh/config:

function _ssh_completion() {

perl -ne 'print "$1 " if /^Host (.+)$/' ~/.ssh/config

}

complete -W "$(_ssh_completion)" ssh

Here's a great article on how to setup your own ~/.ssh/config:

http://blogs.perl.org/users/smylers/2011/08/ssh-productivity-tips.html

Find all files in /var/spool/mqueue older than 7 days, pass to perl to efficiently delete them (faster than xargs or -exec when you've got millions or hundreds of thousands to delete). Naturally the type, directory, and file age vars can be adjusted to meet your specific needs.

Can be run as a script `ftrace` if my_command is substrituted with "$@"

It is useful when running a command that fails and you have the feeling it is accessing a file you are not aware of.

If you've ever tried "grep -P" you know how terrible it is. Even the man page describes it as "highly experimental". This function will let you 'grep' pipes and files using Perl syntax for regular expressions.

The first argument is the pattern, e.g. '/foo/'. The second argument is a filename (optional).

this command example converts to 25 fps subtitles that were originally created for 24 fps movie

more idiomatic version of the same, using the flip-flop-operator; also printing lines with '//'-style comments