Get a user access token (https://developers.facebook.com/tools/explorer/) and copy it into src/main/resources/token.txt. Set the accounts you want feed to be recovered from in src/main/resources/facebookPages.txt. The account names have to be written as they appear in the browser's address bar, one account per line.

Create the JAR with:

mvn package

The jar file is now located in target/spark-facebook-$VERSION.jar where version is the version number specified in pom.xml.

Copy the dependencies into target/dependency with (only if it is the first time you are building):

mvn dependency:copy-dependencies

Copy the dependencies into target/dependency with (only if it is the first time you are building):

mvn dependency:copy-dependencies

Note: a Cassandra instance must be running. From the $SPARK_HOME folder, run the following:

./bin/spark-submit --jars --jars $(echo $PATH_TO_JAR/dependency/*.jar | tr ' ' ',') --class com.t3g7.spark.facebook.FacebookStreamingApp $PATH_TO_JAR/spark-facebook-$VERSION.jar

$PATH_TO_JAR should be /path_to_project/target where path_to_project is the directory in which the project is located.

You can also store PATH_TO_JAR as a bash variable, so you just have to copy-paste the above command. From the project's root directory do :

export PATH_TO_JAR=./target/

Tip : erase the end of the command, until$VERSION, and hit TAB to automatically get the correct jar version.