In this example the fiber is terminated after one yield. I expected the data_rows.each enumerator to provide repeated fiber.yields until the data or events are exhausted. I want to make a single pass through the data while processing all events rather than walking the data file for each event. I have code that does this the slow way. But since I will be running this on thousands of data sets use of Fiber seems appropriate.

This example is simplified from a real world need to repeatedly process tens of millions of data samples for analysis of pre-identified events. The data is sorted by sample # but there are gaps in the sample #s. Thus sample # is NOT the position in the array. Events are sparsely located within the data and do not overlap.

class Example
 def run
    fiber = Fiber.new do
      data_rows.each do |row|
        next unless @onset == row[0]
        p row
        Fiber.yield if row[0] == @completion
      end
    end

.    events.each do |evnt|
      @onset, @completion = *evnt
      fiber.resume
    end
  end

  # [start sample #, completion sample #]
  def events
    [ [2,3], [9,20] ]
  end

  # Test data is sorted by sample # but there are gaps in the sample #s
  # [sample #, datum]
  def data_rows
    [ [0, 'A'], [2, 'B'], [3, 'C'], [7, 'D'], [9, 'F'], [20, 'G'], [30, 'H'] ]
  end
end

Example.new.run```