Format historical weather data exported from the NCDC

ncdc_processor.py

#!/usr/bin/env python
'''
ncdc_processor.py
imports a csv of historical weather data from the NCDC 
exports a somewhat more sanely formatted csv suitable for graphing

data from http://www7.ncdc.noaa.gov/CDO/cdo

usage:
    $ python ncdc_processor.py /path/to/ncdc_data.txt /path/to/export.csv
'''
import csv
import sys
import time

class NCDC():
    def __init__(self):
        # more easily understandable headers
        self.labels = ['Station', 'WBAN', 'date', 'temp (F)', 'temp count'
            , 'dewpoint', 'dwp count', 'sea level pressure', 'slp count'
            , 'station pressure', 'stp count', 'visibility (miles)'
            , 'vis count', 'windspeed (knots)', 'windspeed count'
            , 'max windspeed (knots)', 'gust (knots)', 'max temp (F)'
            , 'min temp (F)', 'precipitation (in)', 'snow_depth (in)'
            , 'events']
        self.reader = csv.DictReader(open(sys.argv[1], 'rb')
            , fieldnames=self.labels)
       
        # setting extrasaction to prevent missing fieldname error
        # ..though there shouldn't be any missing fields
        self.writer = csv.DictWriter(open(sys.argv[2], 'wb')
            , fieldnames=self.labels, extrasaction='ignore')
        self.writer.writeheader()


    def process(self):
        for row in self.reader:
            # skip the headers from the original
            if row['Station'] == 'STN---':
                continue
            
            for label in self.labels:
                # remove whitespace from all values
                row[label] = row[label].strip()

            # an asterisk denotes the origin of derived max temp
            max_label = 'max temp (F)'
            if row[max_label][-1] == '*':
                row[max_label] = row[max_label][:-1]
            
            # an asterisk denotes the origin of derived min temp
            min_label = 'min temp (F)'
            if row[min_label][-1] == '*':
                row[min_label] = row[min_label][:-1]

            # letters in the precip data denotes the length of a storm
            precip_label = 'precipitation (in)'
            if row[precip_label][-1] in 'ABCDEFGHI':
                row[precip_label] = row[precip_label][:-1]

            # format the date
            date_label = 'date'
            row[date_label] = time.strftime('%Y-%m-%d'
                , time.strptime(row[date_label], '%Y%m%d'))

            self.writer.writerow(row)


if __name__ == '__main__':
    ncdc = NCDC()
    ncdc.process()