“Cache” credentials on a bash script

I have created a bash script to run a few js scripts into MongoDB. Basically, what I am doing is, running the bash script and passing a parameter in this case called version, example:

./script.sh 1.0

That will go and execute all scripts for the version 1.0. Now, it is possible that MongoDB requires authentication user/pass, so I have an option in the script execution that will ask the user if it requires authentication.

read -p "Username: " mongo_user; read -s -p "Password: " mongo_pass;

My question is: what would be the best way to kind cache the same credentials to call the script multiple times? For example:

./script.sh 1.0
./script.sh 1.1
./script.sh 1.2 and on.. 

I don’t want to type in the same credentials every time the script runs.

Answer

Caio,

As stated in my comment here’s how I did it:
Thanks for Charles Duffy for the printf solution:

#!/bin/bash

ePass() {
        read -sp "Password: " pass
        echo ""
        printf '%sn' "$pass" | perl -e 'chomp($passwd=<>); chomp($encoded=pack("u",$passwd));print "$encodedn"' > .pswd
        cat .pswd
}

dPass() {
        dPass=`cat .pswd | perl -e 'chomp($encoded=<>); chomp($passwd=unpack("u",$encoded)); print "$passwdn"'`
        echo $dPass
}


ePass  
dPass

You can add these functions to your script. When you want to set the password ePass will do:

[KUBO@home ~]$ ./test.sh 
Password: 

It will mask the input to avoid over-the-shoulder reading. Then it will echo the encoded output (remove after testing):

Password: Hello >>> %2&5L;&`

Then you dPass:

Hello

So when you call your mongo scripts you can use the dPass output as your arg.