Changed the order of steps for word count in Kafka

Author: Petter Graff

labs/06-Streaming/wordcount-kafka-streaming.md

@@ -30,21 +30,9 @@ our `Topology` and hook it up to Kafka via our Kafka client properties, then `st
 
 Lastly, for convenience, we wait for the user to hit enter to `close` the stream and exit.
 
-### Build the streaming app
-
-Now that we have coded our app, we need to build it. Fortunately, we can use `docker` for this so that we don't have to
-have `maven` and its prerequisites installed locally. Open a terminal in the lab's `wordcount-kafka-streaming` directory and issue
-the following command:
-
-```shell
-$ docker run -it --rm -v "$(cd "$PWD/../.."; pwd)":/course-root -w "/course-root/$(basename $(cd "$PWD/.."; pwd))/$(basename "$PWD")" -v "$HOME/.m2/repository":/root/.m2/repository maven:3-jdk-11 ./mvnw clean package
-```
+## Start Kafka
 
-On a windows machine, you have to replace the `$PWD` with the current directory and the `$HOME` with a directory where you have the `.m2` folder.
-
-The command above will build and package our uber jar with the application and all of its dependencies.
-
-## Run Kafka
+### Start containers
 
 In order to run our app, we first need to run Kafka. First, ensure that you've shut down any prior docker containers.
 
@@ -58,33 +46,56 @@ $ docker-compose -f kafka-streaming.yaml up
 You will see logs from all the containers that are launched as part of the solution. Once the terminal stops reflecting
 new output, the infrastructure is initialized.
 
-## Start our streaming application
+### Create topics and start the consumer
 
-Now, let's start our streaming application connecting to Kafka running in our Docker environment:
+In yet another terminal, change into the lab's root directory again and this time, start a bash session in the _kafka_
+container:
 
 ```shell
-$ docker run --network "$(cd .. && basename "$(pwd)" | tr '[:upper:]' '[:lower:]')_default" --rm -it -v "$PWD:/pwd" -w /pwd openjdk:11 java -jar target/wordcount-kafka-solution-*.jar
+$ docker-compose -f kafka-streaming.yaml exec kafka bash
 ```
 
-On a windows machine, you have to replace the `$PWD` with the current directory and the `$HOME` with a directory where you have the `.m2` folder.
+At the subsequent prompt, create the topics & start a Kafka console consumer:
+
+```shell
+I have no name!@c07eea6aed61:/$ for it in input output; do kafka-topics.sh --create --bootstrap-server kafka:9092 --topic stream-$it; done
+I have no name!@c07eea6aed61:/$ kafka-console-consumer.sh --bootstrap-server kafka:9092 --topic stream-output --from-beginning --property print.key=true
+```
 
 
-Now that our Kafka streaming application is running, it's time to feed it some input lines via the Kafka topic.
+## Streaming Application
 
-In yet another terminal, change into the lab's root directory again and this time, start a bash session in the _kafka_
-container:
+### Build the streaming app
+
+Now that we have coded our app, we need to build it. Fortunately, we can use `docker` for this so that we don't have to
+have `maven` and its prerequisites installed locally. Open a terminal in the lab's `wordcount-kafka-streaming` directory and issue
+the following command:
 
 ```shell
-$ docker-compose -f kafka-streaming.yaml exec kafka bash
+$ docker run -it --rm -v "$(cd "$PWD/../.."; pwd)":/course-root -w "/course-root/$(basename $(cd "$PWD/.."; pwd))/$(basename "$PWD")" -v "$HOME/.m2/repository":/root/.m2/repository maven:3-jdk-11 ./mvnw clean package
 ```
 
-At the subsequent prompt, create the topics & start a Kafka console consumer:
+On a windows machine, you have to replace the `$PWD` with the current directory and the `$HOME` with a directory where you have the `.m2` folder.
+
+The command above will build and package our uber jar with the application and all of its dependencies.
+
+### Start our streaming application
+
+Now, let's start our streaming application connecting to Kafka running in our Docker environment:
 
 ```shell
-I have no name!@c07eea6aed61:/$ for it in input output; do kafka-topics.sh --create --bootstrap-server kafka:9092 --topic stream-$it; done
-I have no name!@c07eea6aed61:/$ kafka-console-consumer.sh --bootstrap-server kafka:9092 --topic stream-output --from-beginning --property print.key=true
+$ docker run --network "$(cd .. && basename "$(pwd)" | tr '[:upper:]' '[:lower:]')_default" --rm -it -v "$PWD:/pwd" -w /pwd openjdk:11 java -jar target/wordcount-kafka-solution-*.jar
 ```
 
+On a windows machine, you have to replace the `$PWD` with the current directory and the `$HOME` with a directory where you have the `.m2` folder.
+
+## Stream data
+
+### Run a kafka-console-producer
+
+Now that our Kafka streaming application is running, it's time to feed it some input lines via the Kafka topic.
+
+
 In another terminal, start a bash session and pump some lines into the kafka console producer:
 
 ```shell
@@ -107,12 +118,19 @@ me	20
 too	7
 ```
 
+### A larger data set
+
 For fun, you can submit the full text of Leo Tolstoy's "War & Peace"!
 
 ```shell
 I have no name!@c07eea6aed61:/$ cat /lab-root/war-and-peace.txt | kafka-console-producer.sh --bootstrap-server kafka:9092 --topic stream-input
 ```
 
+
+## Eplilogue 
+
+### Shutdown
+
 In the lab's root directory, you can now bring down the cluster with the command
 
 ```shell