why? vangav analytics is a vangav backend template covering: service oriented architecture and multi-entry-point api; this is also the analytics service used by all vangav's products

• vangav analytics writer and vangav analytics reader services work together and are generated using vangav backend

• vangav backend tutorials

• splitting the analytics backend into writer and reader services is done because writer/reader loads can vary significantly (i.e.: writer can take the load of millions of users using analyzed services while the reader can be used by few system admins); so it's easier to handle these loads by deploying a lot more writer than reader services
• actions are defined in a flexible/generic json file (as explained later in this section) to make it easily extensible in handling analysis as backend services change/grow
• both of the writer and reader services have actions-config-reload feature to allow modifying actions without having to re-start the services

• handles writing analytics

• handles reading actions in various ways (e.g.: by year, by day, by category, ...)

• actions.json in both services define how analytics are structured

• here's a simple example

  {
    "categories": [
      {
        "category_name": "page_views",
        "category_name_short": "PV"
      }
    ],
    "action_classes": [
      {
        "class_name": "backend_page_views",
        "class_prefix": "BPV",
        "class_wide_categories": [
          "PV"
        ],
        "action_ids": [
          {
            "action_id": "quick_start",
            "action_categories": []
          }
        ]
      }
    ]
  }

• this service is based on vangav backend's vangav analytics template
• this service has the 90+% of the vangav backend's generated code + the 10-% of the logic code needed to complete the service

1. for first timers - follow the steps in the system requirements tutorial
2. for first timers - follow the steps in the workspace initialization tutorial
3. download vos_vangav_analytics_writer.zip and vos_vangav_analytics_reader.zip projects (from the green clone or download button) inside the workspace directory created previously (my_services) and unzip them
4. rename unzipped directories, remove the -master from their names
5. in the terminal cd to vos_vangav_analytics_writer/cassandra/cql/
6. execute ./_start_cassandra.sh to start cassandra
7. cd to vos_vangav_analytics_writer/cassandra/cql/drop_and_create/
8. execute the command ./_execute_cql.sh v_analytics_dev.cql to initialize the services' database tables
9. cd to vos_vangav_analytics_writer and execute ./_run.sh to start the analytics writer service on port 9000
10. cd to vos_vangav_analytics_reader and execute ./_run.sh 7000 to start the analytics reader service on port 7000
11. from your prefered client (we recommned postman) start trying the service; refer to the features and service references sections for reference

• at the end to stop the services: press control + d in the terminal session where each service was started in (9 and 10)
• to stop cassandra: execute ps auwx | grep cassandra to get cassandra's (pid) then kill -9 (pid) to stop cassandra

• follow the following steps to import the downloaded backend service in eclipse

1. cd to the service's directory and execute the ./_eclipsify.sh script
2. file > import > general > existing projects into workspace > next > set "select root directory" to my_services > under projects make sure that vos_calculate_sum is selected > finish
3. double check the java version used for compiling the project: right click the project > properties > java compiler > enable project specific settings > compiler compliance level > 1.7 or 1.8

• generate multiple services (writer + reader) to work together in a service oriented architecture
• generic service design (handles any type of analytics)

• this section lists the 10-% code added after vangav backend generated 90+% of the code

• following are the error codes of whatsapp services