test_setup.wiki_tarball.yml

---
# This playbooks downloads tarball with a dump of all english wikipedia
# articles enwiki-latest-pages-articles.xml.bz2 on a client machine, so that
# one can try to use it for testing, eg:
#
#  * extract some articles into individual files on gluster volume via
#  wiki-export-split.py script, distributing lot of small files on all bricks
#  so that the files on the bricks will be distributed uniformly
#
#  * copy the tarball into the volume directly (this will be a bit faster than
#  generating equivalent size of random data)
#
# The original source of the tarball is available on dumps.wikimedia.org here:
# https://dumps.wikimedia.org/enwiki/latest/enwiki-latest-pages-articles.xml.bz2
# For the purpose of this playbook, you should setup a local mirror instead of
# using the original url directly, see task about mandatory variables below.

- hosts: usm_client
  remote_user: root
  vars:
    # you are not expected to change these variables
    - wiki_tarball_filename: "enwiki-latest-pages-articles.xml.bz2"
    - wiki_tarball_dest: "/tmp"

  tasks:

    - name: Check that mandatory variables are defined
      assert:
        that:
          - wiki_tarball_url is defined
          - wiki_tarball_checksum is defined
        msg: >
          You need to define *url of the tarball* in wiki_tarball_url
          variable in your inventory file.
          Eg. http://example.com/enwiki-latest-pages-articles.xml.bz2

          You also need to define *checksum of the tarball* in
          wiki_tarball_checksum variable.
          Eg. md5:a538459bfe1bc35e4d8161840eadbcd

    - name: Check that mandatory variables references a local mirror
      assert:
        that:
          - "'wikimedia.org' not in '{{ wiki_tarball_url }}'"
        msg: >
          Don't use original source on wikimedia.org, but host the file on
          your own infrastructure!

    - name: Download the wiki tarball into /tmp directory (wiki_tarball_dest)
      get_url:
        url: "{{ wiki_tarball_url }}"
        checksum: "{{ wiki_tarball_checksum }}"
        dest: "{{ wiki_tarball_dest }}/{{ wiki_tarball_filename }}"

    - name: Extract wiki tarball checksum data
      set_fact:
        wiki_tarball_checksum_type: "{{ wiki_tarball_checksum|regex_replace('([^:]*):.*', '\\1') }}"
        wiki_tarball_checksum_data: "{{ wiki_tarball_checksum|regex_replace('[^:]*:(.*)', '\\1') }}"

    - name: Store wiki tarball checksum in checksum file
      copy:
        content: "{{ wiki_tarball_checksum_data }}  {{ wiki_tarball_filename }}"
        dest: "{{ wiki_tarball_dest }}/{{ wiki_tarball_filename }}.{{ wiki_tarball_checksum_type }}"

    - name: Install wiki-export-split.py script into /usr/local/bin
      copy:
        src: "bin/wiki-export-split.py"
        dest: "/usr/local/bin/wiki-export-split.py"
        mode: "755"