Functions in Bash

We don’t name function parameters in the function declaration. Instead we have an implied index of arguments: $1, $2, $3,.... When the function is called, the first value after the function name becomes $1 by default, then the subsequent arguments.

function expandRange() {
    declare -a expandedRange=()
    for (( i=$1; i<=$2; i++ )); do
        expandedRange+=($i)
    done
    echo "${expandedRange[@]}"
}
expandedRange=$(expandRange 1 4)
echo $expandedRange
# 1 2 3 4

Get all arguments as an array

We can access all the arguments passed to a function using the $@ syntax we encountered before when Passing_arguments_to_Bash_scripts. (Here a function is a kind of script in miniature so the process is the same.)

function numberThings() {
  i=1
  for f in "$@"; do
    echo $i: "$f"
    (( i++ ))
  done
}

Local variables

var1="I'm variable 1"

function myfunction() {
  var2="I'm variable 2"
  local var3="I'm variable 3"
}

myfunction
echo $var1
echo $var2
echo $var3

# I'm variable 1
# I'm variable 2

The convention is to put functions at the top of the script, after the shebang and after the global variables

logo