Most (all?) shells have the means of creating new “internal” commands. This is done by creating shell functions. Shell functions are just like those in a programming language. Sets of commands are grouped together and jointly called by a single name.

The format for functions is:

Functions can be defined anywhere, including from the command line. All you need to do is simply type in the lines one at a time, similar to the way shown above. The thing to bear in mind is that if you type a function from a command line, once you exit that shell, the function is gone.

Most of the time functions are defined within a shell script. As you might expect, the function is only valid within the scope of the scipt. However, if you source the shell scripts as we discussed in the section on shell basics, then the function is valid for the current shell.

You can also pass arguments to shell functions, just like commands. A simple example is a script that looks like this:

Hello

Here we need to be careful. The variable $1 is the positional parameter from the call to the display function and not to any script containing this function. We can see how this works if we create a the script to look like this:

This is slightly different than the function by itself in that we first call the function itself and then there is an extra echo at the end. Lets call the script display.sh, change the permissions so that it is executable and start it like this:

The output would then look like this:

The first echo shows us the parameter from the function and the second one shows us the parameter from the command line.

As you can see, although we are using the same positional parameter ($1), what is assigned to it depends on the scope. The parameter command line passed on the command line (“Hi”) is valid for the entires scope of the script. However, it is overwritten within the scope of the function by the new value.