ObiLab - BASH Scripting

Functions

Information can be passed to functions through arguments. An argument is just like a variable. Arguments are specified after the function name.

#!/bin/bash
function printTwoStrings {
               echo $1 $2    #parameters aren't listed
                             #you can call them with $1 $2
                             #You must know, before to call the function, how it works.
           }

printTwoStrings Hello Obistudends!

We had just replaced the "echo" function, with "printTwoStrings"
Apparently really useless! But...

#!/bin/bash
function sum{
		sum=$(( $1 + $2 ))
		echo $sum	
}

sum 3 4

#!/bin/bash
function sum {
                sum=$(( $1 + $2 ))
                echo $sum  #is used to return value in standard output or to the main program
}

sum 3 4
sum 4 5
sum 7 9

# or better

read -p "write the first number " a
read -p "write the second number " b
sum $a $b

# or better

        for  i in {0..9}
        do
                sum $i 1
        done

So, how we can write the function, called pow for for

#!/bin/bash
function pow {
        res=$(( $1*$1))
        echo $res
}

read -p "insert a number : " a
pow a

Best way to obtain the function "pow" is: for

#!/bin/bash
function pow {
        res=1
        for (( i=0; i<$2; i++ ))
        do
                res=$(( $res * $1 ))
                # echo $res
        done
        echo $res
}

read -p "insert a number : " base
read -p "insert exp : " exp
pow $base $exp

Functions must be as atomic as possible and as reusable as possible. Our goal is to write functions that we will be able to use at work, daily. So it is useless to write a function to obtain the sum of 2+3, or the pow of 2, but the sum of variables.
Variables, as we saw, can be integer or strings, so I'm pretty confident that you are already thinking about some useful idea...

Arguments or standard input?

We can send information to a script via standard input (using the read function), or passing it as argument once you are running it over the shell.

	./runScript Argument1 Argument2

As for the functions, we can re-call arguments, into the script, with $n

./runScript Ciao Obilab

#!/bin/bash
echo $1 $2 #this will print the 2 arguments we've sent

Let's do some tests: array iteration with function

#!/bin/bash

function printArray()
{
        for (( i=0; i<$1; i++ ))
        do
                echo ${vet[$i]}
        done
}

vet=("17" "9" "14") # let's declare a simple array of 3 elements, INT


n=3; # Number of elements of Array, as noobs does!
tot=${#vet[@]} #Number of elements of Array, as shady does!

#Print the Array
#       for (( i=0; i<tot; i++ ))  #we can use "n" or "tot"
#       do
#               echo ${vet[$i]}
#       done

clear
echo now we will print the array
        printArray ${#vet[@]} ${vet[@]}

echo "that's all!!"

So... are you ready for the final test? Let's figure it out!

Two other tips: I do love math, but... do not write codes as a math book. Variable, functions, must be readable, reusable with mnemonic names.

function x{
      x=y/z*k-r
	echo x
}
x a b c d

## or worse than worse

function x{x=y/z*k-r; echo x;}

It is the WORST method to implement a function, especially if you are working with other people, and if you will have to read your codes of 10.000 lines, after... 2 years!!!