voycey

	$subdomain = substr( env("HTTP_HOST"), 0, strpos(env("HTTP_HOST"), ".") );
	Configure::write('captacion', '');
	if( strlen($subdomain)>0 && $subdomain != "m" ) { 
    Router::connect('*',array('controller'=>'private','action'=>'promote'));
    // Router::connect('/foo', array('controller'=>'mobiles','action'=>'foo'));
    Configure::write('captacion', $subdomain);
	}

voycey

%%time
import dask_cudf as dc

ddf = dc.read_csv('/data/Data Files/Vegas/datafiles/csv/*.csv.gz', compression='gzip')

voycey

%%time
ddf = ddf.repartition(npartitions=3000)

voycey

%%time
ddf.to_orc('/data/Data Files/Vegas/datafiles/orc/')

voycey

SELECT m.id, d.name, d.brand, d.wkt from dealerships d
LEFT JOIN movement m ON ST_CONTAINS(d.wkt, ST_POINT(m.lng, m.lat))
WHERE d.brand IN ('AUDI', 'BMW')

voycey

int pnpoly(int nvert, float *vertx, float *verty, float testx, float testy)
{
  int i, j, c = 0;
  for (i = 0, j = nvert-1; i < nvert; j = i++) {
    if ( ((verty[i]>testy) != (verty[j]>testy)) &&
	 (testx < (vertx[j]-vertx[i]) * (testy-verty[i]) / (verty[j]-verty[i]) + vertx[i]) )
       c = !c;
  }
  return c;
}

voycey

SELECT count(distinct(id)) from datatable d
LEFT JOIN polygons p ON ST_DWITHIN(ST_MAKEPOINT(d.lon,d.lat), p.geom, 0.1)
WHERE p.place_cat IN ('beauty', 'sports')

voycey

mapped_readings = ApplyMapping.apply(frame=datasource, mappings=[
    ("meter_id", "string", "meter_id", "string"),
    ("period_start", "string", "period_start", "string"),
    ("period_end", "string", "period_end", "string"),
    ("e1", "double", "export_interval.e1", "double"),
    ("e2", "double", "export_interval.e2", "double"),
    ("e3", "double", "export_interval.e3", "double"), #NB: this can technically have many more circuits
    ("quality_method", "string", "quality_method", "string"),
    ("event", "string", "event", "string"),
    ("b1", "double", "import_interval.b1", "double"),

voycey

mapped_readings_df = mapped_readings_df \
    .withColumn("week_of_year", F.weekofyear(reading_time)) \
    .withColumn("date_str", F.regexp_replace(F.col("period_end").substr(1,10), "-", "")) \
    .withColumn("day_of_month", F.dayofmonth(reading_time)) \
    .withColumn("month", F.month(reading_time)) \
    .withColumn("year", F.year(reading_time)) \
    .withColumn("hour", F.hour(reading_time)) \
    .withColumn("minute", F.minute(reading_time)) \
    .withColumn("reading_date_time", reading_time) \
    .withColumn("ingestion_date", F.current_timestamp()) \

voycey

def write_to_quest(df_row):
    HOST = args['questdb_host']
    PORT = int(args['questdb_port'])
    sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)

    try:
        sock.connect((HOST, PORT))
        sock.sendall(_row_to_line_protocol(df_row).encode())

    except socket.error as e: