Write average weather data from a year given by user to a text file

Question

I have a text file with weather data from over 60 years that looks like this: (if its messed up just copy into text file!)

Stnr    Dato        DD06  DD12  DD18  FFM   FXM   POM     TAM   UUM
50540   07.01.1957  150   170   170   6.2   8.8   1010.6  6.3   94
50540   08.01.1957  160   160   200   7.2   9.8   1001.8  8.0   99
50540   09.01.1957  290   200   160   8.1   13.3  990.2   5.7   91

I want to write a function that reads data from this file for one year, where the year is determined by the user, how do I do this? And then write the avg value for FFM and TAM for each month in a new file.

Answer 1

This looks like a .csv file or more specifically a tab delimited file .

I would recommend reading the file into some python data format, list or dictionary perhaps, and then doing a search across the date column.

The process is:

1. Figuring out how to store the data in an accessible fashion.
2. Providing a method to create such a data set.
3. Provide a method to search through your data.

I answered these steps with a dictionary, the method data_to_python below, and the method search_by_year , respectively.

Here is my rough attempt, test output below...

def data_to_python(data_file_name):
    with open(data_file_name,'r') as f:
        data = []
        first = True
        for line in f:
            if first:
                first = False
                datanames = line.split('\t')
            else:
                temp = {}
                for i,item in enumerate(line.split('\t')):
                     temp[datanames[i]] = item
                data.append(temp)
    return data

def searchByYear(data,year):
    temp = []
    for entry in data:
        if entry['Dato'].endswith(str(year)):
            temp.append(entry)
    return temp

I put the input you provided into a txt file called test.txt

Sample output:

>>> data = dataToPython('test.txt')
>>> searchByYear(data,1957)
    [{'FFM': '6.2', 'DD18': '170', 'DD06': '150', 'Stnr': '50540', 'DD12': '170', 'FXM':'8.8', 'Dato': '07.01.1957', 'POM': '1010.6', 'UUM\n': '94\n', 'TAM': '6.3'}, {'FFM': '7.2', 'DD18': '200', 'DD06': '160', 'Stnr': '50540', 'DD12': '160', 'FXM': '9.8', 'Dato': '08.01.1957', 'POM': '1001.8', 'UUM\n': '99\n', 'TAM': '8.0'}, {'FFM': '8.1', 'DD18': '160', 'DD06': '290', 'Stnr': '50540', 'DD12': '200', 'FXM': '13.3', 'Dato': '09.01.1957', 'POM': '990.2', 'UUM\n': '91', 'TAM': '5.7'}]
>>> searchByYear(data,1956)
[]

Answer 2

I would suggest treating this as a csv file using csv.DictReader() You would specify delimiter='\\t' (if the text file shown has tabs as delimiter and not space). You would loop through the rows, determine the appropriate dates using (datetime.datetime) strptime() with a format of '%m.%d.%Y'

The input row gets the dictionary keys from the first line and you accumulate the data and use numpy to get the mean for each set of data. Your output file could use cvs.DictWriter() with the keys month, FFM, TAM for the averages. You can use matplotlib.pyplot to create a map of the values by month to see how they chanege. A bar graph of two different colors , for example.