parse_wiki_xml.rb

# encoding: utf-8

require 'xmlsimple'
require 'csv'

hash = XmlSimple.xml_in('m.xml')

start_date = DateTime.parse "2013-12-28"
end_data   = Date.today

CSV.open("out.csv", "w") do |csv|
  csv << ['revision_id', 'page_title', 'username', 'timestamp', 'bytes', 'last_revision_id']
  hash['page'].each do |page|
    page_title = page['title'].first
    if first_revision = page['revision'].first
      revision_id   = first_revision['id'].first
      username      = first_revision['contributor'].first.values.first.first
      timestamp     = first_revision['timestamp'].first

      last_revision = page['revision'].last
      bytes         = last_revision['text'].first['bytes']
      last_revision_id  = last_revision['id'].first

      date        = DateTime.parse timestamp

      if date >= start_date && date <= end_data
        if page_title.include?('താൾ:')
          csv << [revision_id, page_title, username, timestamp, bytes, last_revision_id]
        end
      end
    end
  end
end


# Checked ID with filtered file using this little shell script
# cat out.csv| while read line; do id=`echo $line | cut -f1 -d','`; (cat ~/Downloads/out\ 5200.csv| cut -f1 -d',' | grep $id > /dev/null) && (echo $line >> data.csv);done