-B flag = don't include characters that can be confused for other characters (this helps when you give someone their password for the first time so they don't cause a lockout with, for example, denyhosts or fail2ban) -s flag = make a "secure", or hard-to-crack password -y flag = include special characters (not used in the example because so many people hate it -- however I recommend it) "1 10" = output 1 password, make it 10 characters in length For even more secure passwords please use the -y flag to include special characters like so: pwgen -Bsy 10 1 output>> }&^Y?.>7Wu Show Sample Output

A slightly shorter version. Also doesn't put a return character at the end of the password

./encode.sh [ h264 | xvid | theora | mpeg4 ] Show Sample Output

no need that useless cat.

Detect video size automatically , usually the flv video from youtube , dailymotion are in the /tmp directory . Just launch a video player or an ls -lh on it to find the one you want to convert. Show Sample Output

generate password Show Sample Output

Useful in shell scripts when you're trying to get the shell script name without the full path - and easier than awking or cutting. Bash pattern matching and variable manip is fun. Show Sample Output

This command will use the short username (ie: bsmith) instead of the full (or long) name (ie: Bob Smith), by default in the network authentication dialog, ie: when connecting to servers. Resulting in one less step/time saved. Simply change the boolean value to "NO" (ie: "... -bool NO") to revert to the system's default behavior.

You can use [n]> combined with >(cmd) to attach the various output file descriptors to be the input of different commands.

add overlay png image to video as logo...........

i want to count how many regex code i have used in vim in a long time so i make a directory in svn host and post record to this directory of course i dont want to post manually so i worte a script to do that and this is the core thing to do Show Sample Output

you can just use one awk script to parse the rss feed. No need to pipe so many awk's and sed's. Its ugly and inefficient.

For the record: I didn't build this. Just shared what I found that worked. Apologies to the original author! I decided I should fix the case where http://example.com is not matched for the next time I need this. So I read rfc1035 and formalized the host name regex. If anyone finds any more holes, please comment.

Quick and kludgy rss parser for the recent tracks rss feed from last.fm. Extracts artist and track link. Show Sample Output

don't have to be that complicated

From http://daringfireball.net/2009/11/liberal_regex_for_matching_urls Thought it would be useful to commandlinefuers. Show Sample Output

Tested on Fedora 12. This function will take a man page and convert it to pdf, saving the output to the current working directory. In Gnome, you can then view the output with "gnome-open file.pdf", or your favorite pdf viewer.

if you still get a permissions error using sudo, then nano the file: sudo nano -w /sys/block/sdb/queue/rotational and change 1 to 0 this thread: http://www.ocztechnologyforum.com/forum/showpost.php?p=369836&postcount=15 says that this will "help the block layer to optimize a few decisions"

The iostat command is used for monitoring system input/output device loading by observing the time the devices are active in relation to their average transfer rates. in ubuntu to get the iostat program do this: sudo apt-get install sysstat i found this command here: http://www.ocztechnologyforum.com/forum/showthread.php?t=54379 Show Sample Output

This function displays the latest comic from xkcd.com. One of the best things about xkcd is the title text when you hover over the comic, so this function also displays that after you close the comic. To get a random xkcd comic, I also use the following: xkcdrandom(){ wget -qO- dynamic.xkcd.com/comic/random|tee >(feh $(grep -Po '(?<=")http://imgs[^/]+/comics/[^"]+\.\w{3}'))|grep -Po '(?<=(\w{3})" title=").*(?=" alt)';}

This command lists extended information about files, i.e. whether or not it is a true file or link, who owns it, etc. without having to 'ls' from the specific directory. If you know the filename, but not the location, this helps with finding other information about the file. It can be truncated by creating an alias for 'ls -l'. The sample output shows difference in regular locate vs. ls + locate. Show Sample Output

http://www.reddit.com/r/reddit.com/comments/a7xpt/conspiracy_theories_commence_wikileaks_to_release/ Show Sample Output

By putting ! in front of a command, we are able to run it from an ftp session. Show Sample Output

Barely worth posting because it is so simple, but I use it literally all the time. I was always frustrated by the limitations that a non-gui environment imposes on diff'ing files. This fixes some of those limitations by colourising the output (you'll have to install colordiff, but it is just a wrapper for diff itself), using side-by-side mode for clearer presentation, and of course, the -W parameter, using tput to automatically insert you terminal width. Note that the double quotes aren't necessary if typed into terminal as-is. I included them for safety sake,