Parsing a flat file using shell script

Question

I have a file like in following format

User:blala Pass:blala IP:***.***.**.** 

I tried with IFS but it is not working

input="/path/to/your/input/file.cvs" while IFS=',' read -r f1 f2 f3 f4 f5 f6 f7 do echo "$f1 $f2 $f3 $f4 $f5 $f6 $f7" done < "$input" 

I want to get User, Pass, IP values in my shell script .

Please help.

Answer 1

It may be tempting to use eval, but as passwords can contain arbirary chars (which could/would be executed as code, it makes it risky; bordering on a plain-and-simple: "Don't do it!".
This works - using arrays.

Contents of test file

User:blala Pass:blala with spaces: and colons: and $PATH IP:***.***.**.** 

---

set -f # disable globbing IFS=$'\n' # 'word' split only at newline char value=($(sed 's/:/\n./' file)) # '.' to cater for empty value fields echo user="${value[1]/./}" echo pass="${value[3]/./}" echo ip="${value[5]/./}" 

Output:

user=blala pass=blala with spaces: and colons: and $PATH ip=***.***.**.** 

Answer 2

I'm not really sure if I get the point of your question.

At first: IFS is a variable, that contains a separator, like a tab, a space or something. By default it contains a space, tab and newline.

EDIT: from a for-loop loop to a while-loop, suggested by terdon

while read line do echo $line|cut -d\: -f2 done < /path/to/file.csv 

This generates this output:

blala blala ***.***.**.** 

You can do this with a while-loop as well, in personal I just like the for-loop here more.

What is the loop doing? At first, the loop reads the first line of the file.csv and saves it to the variable "line".

Inside the loop we echo the variable $line and cut it into peaces. We cut it with the delimiter (-d) ":", because in youre file.csv there is always a ":" between the identifier and the value. with the field-indicator (-f) we only want to show us the field 2 (-f2). The for-loop ends when the file.csv has no more lines.

For more informations see the man page of cut(1).