lakshmanok

double[] inputx = c.element(); // from input
engine.put("x", inputx);

// run R program, get output from R, send to Dataflow
ListVector result = (ListVector) engine.eval(new InputStreamReader(rprog));

double pvalue = result.getElementAsDouble(1);

lakshmanok

SELECT
  date,
  airline,
  departure_airport,
  departure_schedule,
  arrival_airport,
  arrival_delay
FROM
  [bigquery-samples:airline_ontime_data.flights]
WHERE

lakshmanok

SELECT
  date,
  airline,
  departure_airport,
  departure_schedule,
  arrival_airport,
  arrival_delay
FROM
  [bigquery-samples:airline_ontime_data.flights]
WHERE

lakshmanok

SELECT
  date,
  airline,
  departure_airport,
  departure_schedule,
  arrival_airport,
  arrival_delay
FROM
  [bigquery-samples:airline_ontime_data.flights]
WHERE

lakshmanok

SELECT
  date,
  airline,
  departure_airport,
  departure_schedule,
  arrival_airport,
  arrival_delay
FROM
  [bigquery-samples:airline_ontime_data.flights]
WHERE

lakshmanok

SELECT
  date,
  airline,
  departure_airport,
  departure_schedule,
  arrival_airport,
  arrival_delay,
FROM
  [bigquery-samples:airline_ontime_data.flights]
WHERE

lakshmanok

class SceneInfo:
   def __init__ (self, line):
      try:
        self.SCENE_ID, self.SPACECRAFT_ID, self.SENSOR_ID, self.DATE_ACQUIRED, self.COLLECTION_NUMBER, self.COLLECTION_CATEGORY,self.DATA_TYPE, self.WRS_PATH, self.WRS_ROW, self.CLOUD_COVER, self.NORTH_LAT, self.SOUTH_LAT, self.WEST_LON, self.EAST_LON, self.TOTAL_SIZE, self.BASE_URL = line.split(',')

        self.DATE_ACQUIRED = datetime.datetime.strptime(self.DATE_ACQUIRED, '%Y-%m-%d')
        self.NORTH_LAT = float(self.NORTH_LAT)
        self.SOUTH_LAT = float(self.SOUTH_LAT)
        self.WEST_LON = float(self.WEST_LON)
        self.EAST_LON = float(self.EAST_LON)

lakshmanok

if __name__ == '__main__':
   p = beam.Pipeline(options=beam.utils.options.PipelineOptions())
   index_file = 'gs://gcp-public-data-landsat/index.csv.gz'
   output_file = 'output.txt'

   # center of Reunion Island
   lat =-21.1; lon = 55.50 

   # Read the index file and find the best look
   scenes = (p

lakshmanok

def clearest(scenes):
   if scenes:
      return min(scenes, key=lambda s: s.CLOUD_COVER)
   else:
      return None

lakshmanok

def clearest_look(allscenes, lat, lon, i, j):
   name = 'pt_{}x{}'.format(i,j)
   return (allscenes
     | 'covers_{}'.format(name) >> beam.FlatMap(lambda scene: filterByLocation(scene, lat+0.25*i, lon-0.4*j))
     | 'bymonth_{}'.format(name) >> beam.FlatMap(lambda scene: sceneByMonth(scene) )
     | 'clearest_{}'.format(name) >> beam.CombinePerKey(clearest)
   )

def run():
   # etc.