The code is exactly the same as with the previous recipe so we will not be repeating it here; the only major difference is that we do not have the build stage in this script, and the _spark_source variable is different. 

As in the previous recipe, we start by specifying the location of Spark's source code, which is in _spark_source. The _spark_archive contains the name of the archive; we use awk to extract the last element. The _spark_dir contains the name of the directory our archive will unpack into; in our current case, this will be spark-2.3.1. Finally, we specify our destination folder where we will be moving the binaries to: it will either be /opt/spark (default) or your home directory if you use the -ns (or --nosudo) switch when calling the ./installFromBinary.sh script.

Next, we check the OS name. Depending on whether you work in a Linux or macOS environment, we will use different tools to download the archive from the internet (check the downloadThePackage function). Also, when setting up the environment variables, we will output to different bash profile files: the .bash_profile on macOS and the .bashrc on Linux (check the setEnvironmentVariables function). 

Following the OS check, we download the package: on macOS, we use curl and on Linux, we use wget tools to attain this goal. Once the package is downloaded, we unpack it using the tar tool, and then move it to its destination folder. If you are running with sudo privileges (without the -ns or --nosudo parameters), the binaries will be moved to the /opt/spark folder; if not—they will end up in the ~/spark folder.

Finally, we add environment variables to the appropriate bash profile files: check the previous recipe for an explanation of what is being added and for what reason. Also, follow the steps at the end of the previous recipe to test if your environment is working properly.