winriver_ascii.py

import pandas as pd
import numpy as np
import argparse

parser = argparse.ArgumentParser()
parser.add_argument("input", nargs="+")

ENSEMBLE_HEADER_LENGTH = 6
COLUMN_NAMES = ["depth", "magnitude", "direction", "east", "north", "up", "error",
	"amplitude1", "amplitude2", "amplitude3", "amplitude4", 
	"pct_good", "discharge"]
VELOCITY_VARIABLES = COLUMN_NAMES[1:6]

def get_metadata(metadata_lines):
	'''
	Parse out relevant metadata from the six metadata lines at the start of
	every ensemble.
	'''
	fields = [x.strip().split() for x in metadata_lines]
	year, month, day, hour, minute, second, hundredth = [int(x) for x in fields[0][0:7]]
	ensemble = int(fields[0][7])
	temp = float(fields[0][12])
	distance = float(fields[2][0])
	lat, lon = [float(x) for x in fields[3][0:2]]
	nbins = int(fields[5][0])
	names = ["year", "month", "day", "hour", "minute", "second", "hundredth", 
		"ensemble", "temp", "lat", "lon", "distance", "nbins"]
	values = [year, month, day, hour, minute, second, hundredth, 
		ensemble, temp, lat, lon, distance, nbins]
	return (names, values)


def read_ascii(filename):
	f = open(filename, "r")
	lines = f.readlines()

	i = 3
	ensembles = []
	while i < len(lines):
		metadata = get_metadata(lines[i : i + ENSEMBLE_HEADER_LENGTH])
		start = i + ENSEMBLE_HEADER_LENGTH
		stop = start + int(metadata[1][-1])
		data = np.array([x.strip().split() for x in lines[start:stop]], dtype=float)
		data = pd.DataFrame(data, columns=COLUMN_NAMES)
		for name, value in zip(metadata[0], metadata[1]):
			data[name] = value
		ensembles.append(data)
		i = stop
	data = pd.concat(ensembles)
	data[data[VELOCITY_VARIABLES] == -32768] = np.nan
	data[data[["discharge"]] == 2147483647] = np.nan
	data[data[["lat", "lon"]] == 30000] = np.nan
	return data.dropna()


if __name__ == '__main__':
	args = parser.parse_args()
	for filename in args.input:
		print filename
		data = read_ascii(filename)
		data.to_csv(filename[0:-4] + ".csv", index=False, na_rep="NA")