Tag cloud taken from http://www.wordle.net/

Bourne shell redirection

0 is stdin;
1 is stdout;
2 is stderr

redirect std err to file
$command 2>file

direct the output from command to stderr
$commamd >& 2

direct std output and std err to a file or /dev/null
$ command >file.log >>file.log
$ command >file.log 2>&1
$ command >/dev/null 2>&1

Some cool one liners..

sum numbers in first column of a file
$ cat filename | awk 'BEGIN { tot=0 } { tot+=$1 } END { print tot }'

only print certain lines in a file, eq 20 - 30
$ sed -n -e 20,30p filename

Search all files in and below for a string xyz:
$ find . -exec egrep xyz {} \; -print

Remove all files name ".log.tmp"
$ find . -name .log.tmp -exec rm {} \;

search a file for any of three strings:
$ egrep 'abc|xyz|or' filename

Sexy Unix Commands

the follwing are the sexy commands.
date;
unzip;
touch;
strip;
finger;
mount;
gasp;
yes;
uptime;

removing duplicate files using shell script

Friends,

One liner script provided below, can be used to identify the duplicate files within the system. If you want to search in a partticular path, just replace it with "/tmp" provided in sample command. It will redirect all the duplicate file names into removal_list.txt which can be used to delete them.

find /tmp "$@" -type f -print0 xargs -0 -n1 md5sum sort --key=1,32 uniq -w 32 -d --all-repeated=separate sed -r 's/^[0-9a-f]*( )*//;s/([^a-zA-Z0-9./_-])/\\\1/g;s/(.+)/#rm \1/' >> removal_list.txt

Performance comparision

Checkout the "time" command.

For checking the time taken for execution of any command do the following:

Say you want to execute "ls -l grep sanju.txt" and measure time taken for this.
Now open a bash shell and on the prompt
$time ls -l grep sanju.txt

it shows you the time taken by the command execution

Unaliasing in BASH

Unaliasing in BASH
How can we unalias a command temporarily without using unalias command ?

just put the escape character before the command i.e. "\"e.g. :-

$\ls

It will show the without alias.