It would be better to have the segment of commands as one group when it should first occur. Then, simply call the group each time the group is needed down in the script. In order to do that, the group needs to be given a name.
A function is a named group of commands that is called whenever it is needed, down in the script. The group of commands is not executed when the function is defined at the top of the script. The group is only executed when the function is called.
Article Content
Function Examples
A Function Without Parameters
Consider the following group of commands:
cd myDir
touch myfile.txt
The first command creates the directory, myDir. The second command makes myDir the current directory. The third command creates the file, myFile.txt, in the current directory. If this code segment were to be repeated three times in a long script, then it would be better to put it in a function, giving the function a name. Putting it into a function is defining the function. The function should be defined at the top of the script and then called three different times at different points, down in the script. When you run the script, the group of commands in the function definition is not executed. They are executed when the function is called down in the script. That is, when you run the script, the function definition is established but not executed. The function is executed, when it is called, down in the script.
This function would be defined and called three times as follows:
function aFn
{
mkdir myDir
cd myDir
touch myfile.txt
}
aFn
aFn
aFn
The first line in the script is not part of the function definition or any function call. It makes the cd command work more effectively. The function definition begins with the reserved word, “function”. This is followed by space, then the name of the function. The name of the function is the programmer’s choice. The name of the function must be followed by whitespace before “{“. The first command of the function body must be preceded by whitespace after the “{“. The last command in the body must be separated from the delimiting ”}” by a newline or “;” or “&”.
In the script, the function has been called three times after the function definition, with the name of the function being aFn .
The effect of the script is to create a directory called myDir. Inside myDir, the file myfile.txt is created. Another myDir and nesting myFile.txt are created, nested in the first myDir. Yet, another myDir and nesting myFile.txt are created, nested in the second myDir.
A Function With Parameters
Imagine that there are 3 textbooks and 2 exercise books on a table. The total number of books is 5. The following script shows how this addition and echoing of result can be done:
{
sum=$((${1} + ${2}))
echo $sum ${3}
}
add 3 2 "books"
The function definition begins with the name of the function, “add”, given by the programmer. This is followed by parentheses, preceded with or without space. That is followed by a “{“, preceded by a whitespace. The commands follow; and then a new line or “;”, or “&”; and finally ”}”.
When a function does not take arguments (parameters), its definition should begin with the reserved word, “function”, then the function name, and no parentheses. When it takes arguments, its definition should begin with the function name and followed by parentheses.
The last command in the script, calls the function. Its first argument is 3, its second argument is 2, and its third argument is “books”. If an argument is a number, it should be typed without quotes. If it is a string of one or more words, it should be typed in single or double quotes.
In the function definition, the first argument is obtained with ${1}, the second argument is obtained with ${2}, and the third argument is obtained with ${3}. If there were a fourth argument, it would be obtained with ${4}; and so on.
Bash by default, adds only integers. A special construct is needed to add two float numbers or add an integer and a float number. See the example below:
Positional Parameters
${1}, ${2}, ${3}, etc. as used above, are positional parameters. Normal counting in programming begins with 0. So, what is the use of ${0}? The ${0} holds the name which is preceded by the path of the Bash script. The following code illustrates this:
{
sum=`echo ${1} + ${2} | bc`
echo The sum is $sum for the script ${0} .
}
add 3.5 2.4
The output is:
Where “./temp.com” is the path and name of the author’s script. Note the line and its backticks for adding floating point numbers.
Function Returning
In the above function, note where two integers were added. Instead of echoing the result, the result could have been returned, with the reserved word “return” as the following script shows:
{
sum=$((${1} + ${2}))
return $sum
}
add 3 2
echo $? books
The output is:
In the function definition, the return command returns the sum. This returned value is held in the special variable, “$?”.
Global and Local Scope
Consider the following script:
function fn
{
var=6
echo $var
}
echo $var
The output is 5. This is because the function was not called. The var outside the function is in the global scope, and the var inside the function is in the local scope. Both variables have the same name and should mean the same thing.
When the function is called, its body sees the global scope variable. The following script illustrates this:
function fn
{
var=6
echo $var
}
fn
echo $var
The output is:
6
The function is called before the global variable is echoed at the last two commands in the script. When the function was called, it saw the global variable and changed its value from 5 to 6.
The local variable inside the function body can be independent of the global variable outside the function definition. This is done by declaring the variable inside the function as local, with the reserved word, “local”. The following script illustrates this:
function fn
{
local var=6
echo $var
}
fn
echo $var
The output is:
5
Because of the reserved word, “local”, the local variable with the same name is seen only within the function body. In contrast, the global variable with the same name is seen only outside the function body, not in the function body.
Recursive Function
A recursive function is a function that calls itself repeatedly until a certain condition is met. The first script above, where the function was called 3 times, can be turned into a recursive function. The condition to be met is 3 calls. This can be done with a counter variable. The following script illustrates this:
counter=0
function aFn
{
mkdir myDir
cd myDir
touch myfile.txt
((counter=$counter + 1))
if [ $counter -le 2 ]; then
aFn
fi
}
aFn
Note how the condition to be met has been coded in the if-construct. In the zeroth pass of the function after it has been called, the counter is 1. In the first pass of the function, the counter is 2. In the second pass of the function, the counter is 3. This is a recursive function.
Conclusion
A function is a group of commands that can be called at least once in the script. A function must have a name given to it by the programmer. The positional parameters of a function are ${1}, ${2}, ${3}, etc., according to the order of the arguments. A number as an argument is written without quotes. A string argument of one or more words is written in quotes. A function can return a value. The return value is held in the special variable, “$?”. A variable inside a function body can override a variable outside the function body, with the reserved word, “local”. A function in Bash can be recursive. Meaning, after the first call, it can call itself over and over again. In order to stop recurring, a condition must be met.
from Linux Hint https://ift.tt/3wQAS2I
0 Comments