dataset.yml

---

- name: Deploy the dataset
  tags: data
  hosts: hadoopnodes
  vars:
    download_dir: "/tmp"
    db_dir: "/tmp/nist"
    db4_path: "{{ db_dir }}/{{ databases.4.url | splitext | first }}"
    databases:
      4:
        url: "https://s3.amazonaws.com/nist-srd/SD4/NISTSpecialDatabase4GrayScaleImagesofFIGS.zip"
        sha512: "8fb35d27c4bbb1e15008fbdad0fae0fc64fa0b5cff12259753ca039223cf1b789ff2362ca2889c4a9777b71f85a4101c3338155e3c6fc5f8f22fa146d21e941b"

  tasks:

    - name: prepare the data directory
      file:
        path: "{{ item }}"
        state: directory
      with_items:
        - "{{ download_dir }}"
        - "{{ db_dir }}"

    - name: download
      get_url:
        url: "{{ databases.4.url }}"
        checksum: "sha512:{{ databases.4.sha512 }}"
        dest: "{{ download_dir }}"
        timeout: 60

    - name: extract
      become: yes
      unarchive:
        src: "{{ download_dir }}/{{ databases.4.url | basename }}"
        dest: "{{ db_dir }}"
        owner: hadoop
        group: hadoop
        creates: "{{ db4_path }}"
        copy: false

    - name: fix permissions
      become: yes
      file:
        path: "{{ db_dir }}"
        owner: hadoop
        group: hadoop
        recurse: yes