Shell Programming and Scripting

Hello there,

I want to remove duplicate directories of $PATH. I already have the code (found in this forum), which removes duplicate lines in a textfile.

Code:

awk 'x[$0]++==0' filename

By how do I use this for the $PATH-variable?

Code:

echo $PATH | awk -F":" '{for (i=1;i<=NF;i++)print $i}'

PIPE this to your command.

Use the below code.

Code:

PATH=` echo "$PATH" | /usr/xpg4/bin/awk -v RS=':' -v ORS=":" '!a[$1]++'`

or 
PATH=$(echo "$PATH" | /usr/xpg4/bin/awk -v RS=':' -v ORS=":" '!a[$1]++')

these both syntax will work choose one of them

BR

Very much thanks for the fast replies to both of you.

I've have made this from the sample of rakeshawasthi

Code:

echo $PATH | awk -F":" '{for (i=1;i<=NF;i++){if (x[$i]++==0) print $i}}'

But I think the example of you, ahmad.diab, I still a little more elegant, and make nearly exactly, what I want.

-----Post Update-----

I just noticed, that I get a colon at the end of $PATH for everytime I run the command.

-----Post Update-----

I noticed, that when ich use "echo -n" there is only one colon at the end undepending of the number of runs of the command. But I still can't export $PATH, because of the trailing colon.

you can put the last ":" in a new line if you change $1 to $0 in the code:

Code:

/usr/xpg4/bin/awk -v RS=':' -v ORS=":" '!a[$0]++'

Sorry, with echo -n didn't work on solaris. But there I used

Code:

printf "$PATH"

instead and the problem still remains...

Code:

echo -n in solaries is exist the below path:
don't worry :)

/usr/ucb/echo -n
BR