3.18. システムイベントの確認

この Ruby の例では、ログに記録されたシステムイベントを取得します。

# In order to ensure that no events are lost, it is recommended to write
# the index of the last processed event, in persistent storage.
# Here, it is stored in a file called `index.txt`. In a production environment,
# it will likely be stored in a database.
INDEX_TXT = 'index.txt'.freeze

def write_index(index)
  File.open(INDEX_TXT, 'w') { |f| f.write(index.to_s) }
end

def read_index
  return File.read(INDEX_TXT).to_i if File.exist?(INDEX_TXT)
  nil
end

# This is the function that is called to process the events. It prints
# the identifier and description of each event.
def process_event(event)
  puts("#{event.id} - #{event.description}")
end

# Find the root of the tree of services:
system_service = connection.system_service

# Find the service that manages the collection of events:
events_service = system_service.events_service

# If no index is stored yet, retrieve the last event and start with it.
# Events are ordered by index, in ascending order. `max=1` retrieves only one event,
# the last event.
unless read_index
  events = events_service.list(max: 1)
  unless events.empty?
    first = events.first
    process_event(first)
    write_index(first.id.to_i)
  end
end

# This loop retrieves the events, always starting from the last index. It waits
# before repeating. The `from` parameter specifies that you want to retrieve
# events that are newer than the last index that was processed. Note: the `max`
# parameter is not used, so that all pending events will be retrieved.
loop do
  sleep(5)
  events = events_service.list(from: read_index)
  events.each do |event|
    process_event(event)
    write_index(event.id.to_i)
  end
end

詳細については、EventsService:list を参照してください。