I am taking a linux class at school and I need some help with the sed command. here is the question. can someone please help.

the receipt data file from the Unit 7 Assignment: receipt (UPC, description, and retail price)
5390:Captain Crunch:4.59
2460:Milk:2.99
8735:Peanut Butter:3.39
4402:Yogurt:0.99
How could you use sed to transform the file to the following format (dollar sign added before cost):

Hi,

here is a good sed-tutorial to get you started:
http://www.grymoire.com/Unix/Sed.html

Once you've looked through the excellent tutorial that was posted above and you still need help, please post what you've tried and where it fails. Please, remember to wrap your code in the code tags.

I came up with this

sed 's/[0-9]*. /&$/' receipt

but the dollar sign is going infront of the description instead of the last column with the dollars. Please give me an idea on what I am doing wrong.

Try checking for the end of line character so it always matches the second numeric occurrence rather than the first.

Ok I guess I am really confused. I don't understand how you would do that. I need a good easy web site that will explain the command very simply, can anyone help?

The website that was suggested in the second post is probably the best one. I can think of 3 ways of accomplishing the task but I'm not going to spoil the fun for you:

I'll just give you some clues:

From your attempt I gather that you know what "&" means in this context
1. Substitute: digit(s) . digit(s) AT THE END with "$&"
2. Substitute the second occurrence of ":" with ":$"
3. Substitute the second occurrence of digit(s) with "$&"

If the rest of the file follows the same pattern all three will work.


edit: Number 2 should actually be: Substitute: digit(s).digit(s)End_of_Line_Character with "$&"

I took a liberty with your data file and got rid of some spaces. I was in a bash shell. <file> is your data file.

cat file |
while read line
do
array0=(`echo $line|sed s'/ //g'`)
array=(`echo $array0|sed s'/:/ /g'`)
echo "${array[0]}:${array[1]}:\$${array[2]}"
done


[tmp]# ./ex
5390:CaptainCrunch:$4.59
2460:Milk:$2.99
8735:PeanutButter:$3.39
4402:Yogurt:$0.99

Imagine. both ways actually work, but I think this is correct.
]$ sed s'/[0-9]*\.[0-9]*$/\$&/' receipt
5390:Captain Crunch:$4.59
2460:Milk:$2.99
8735:Peanut Butter:$3.39
4402:Yogurt:$0.99



Duh!
I agree with sycamorex. Think abut it a few minutes and you can get it down to one line.

[$] sed s'/[0-9]*.[0-9]*$/\$&/' receipt
5390:Captain Crunch:$4.59
2460:Milk:$2.99
8735:Peanut Butter:$3.39
4402:Yogurt:$0.99

================
This is too long but it might be useful for something else.

cat file |
while read line
do
array=(`echo $line | sed s'/ //g' | sed s'/:/ /g'`)
echo "${array[0]}:${array[1]}:\$${array[2]}"
done

[tmp]# ./ex
5390:CaptainCrunch:$4.59
2460:Milk:$2.99
8735:PeanutButter:$3.39
4402:Yogurt:$0.99