'''
   weatherstats.py
   Jeff Ondich, 10/12/09

   This implementation of the Weather Stats program was developed
   with an eye towards keeping the main program easy to understand.
   One result of this design goal is quite a bit of redundant
   computation.
'''

import sys
import math

def loadWeatherData(fileName):
    '''Loads temperature data from the specified file into a
       list of lists of integers, and returns the list.  This
       function assumes that each line of the file consists of
       a year followed by some number of temperatures, all
       separated by commas.  For example, if the file looks
       like this: 

           1939,-10,5,3,14,...
           1940,2,8,-1,4,...
           ...
           2008,7,12,3,5,-2,...

       (which means that on Jan 1, 1939, the high temperature
       was 10, etc.), then loadWeatherData will return a list
       of lists like this:

           [[1939,-10,5,3,14,...], [1940,2,...], ..., [2008,7,12,...]]
    '''
    weatherData = []
    file = open(fileName)
    for line in file:
        yearData = line.strip().split(',')
        yearData = map(int, yearData)
        weatherData.append(yearData)
    file.close()

    return weatherData

def mean(listOfNumbers):
    '''Returns the mean of the specified list of numbers
       if the list contains at least one element.  Otherwise,
       returns 0.  This function assumes that its parameter refers
       to a (possibly empty) list of numbers.'''
    average = 0
    if len(listOfNumbers) > 0:
        average = float(sum(listOfNumbers)) / float(len(listOfNumbers))

    return average

def standardDeviation(listOfNumbers):
    '''Returns the standard deviation of the specified list of
       numbers if the list contains 2 or more elements.  Otherwise,
       returns 0.  This function assumes that its parameter refers
       to a (possibly empty) list of numbers.'''
    if len(listOfNumbers) < 2:
        return 0

    avg = mean(listOfNumbers)
    sumOfSquares = 0
    for x in listOfNumbers:
        sumOfSquares += (x - avg)**2
    
    return math.sqrt(sumOfSquares / float(len(listOfNumbers) - 1))

def getExtremeMeans(data):
    highMean = lowMean = mean(data[0][1:])
    highMeanYear = lowMeanYear = data[0][0]
    for yearData in data:
        avg = mean(yearData[1:])
        if avg > highMean:
            highMean = avg
            highMeanYear = yearData[0]
        if avg < lowMean:
            lowMean = avg
            lowMeanYear = yearData[0]

    return (highMean, highMeanYear, lowMean, lowMeanYear)

def getExtremeStdevs(data):
    highStdev = lowStdev = standardDeviation(data[0][1:])
    highStdevYear = lowStdevYear = data[0][0]
    for yearData in data:
        stdev = standardDeviation(yearData[1:])
        if stdev > highStdev:
            highStdev = stdev
            highStdevYear = yearData[0]
        if stdev < lowStdev:
            lowStdev = stdev
            lowStdevYear = yearData[0]

    return (highStdev, highStdevYear, lowStdev, lowStdevYear)

def getExtremeTemps(data):
    highTemp = lowTemp = data[0][1]
    highTempYear = lowTempYear = data[0][0]
    for yearData in data:
        for temp in yearData[1:]:
            if temp > highTemp:
                highTemp = temp
                highTempYear = yearData[0]
            if temp < lowTemp:
                lowTemp = temp
                lowTempYear = yearData[0]

    return (highTemp, highTempYear, lowTemp, lowTempYear)

# ---- Main program starts here. ---- #

# Check the command line.
if len(sys.argv) != 2:
    sys.stderr.write('Usage: %s weatherDataFile\n' % sys.argv[0])
    sys.exit(1)

# Print the yearly reports.
print 'High temperature data (Fahrenheit)'
weatherData = loadWeatherData(sys.argv[1])
for yearData in weatherData:
    year = yearData[0]
    avg = mean(yearData[1:])
    stdev = standardDeviation(yearData[1:])
    print '%-8d%11.2f%11.2f' % (year, avg, stdev)
print

# Print the all-time reports.
(highMean, highMeanYear, lowMean, lowMeanYear) = getExtremeMeans(weatherData)
print 'Highest mean temperature: %.2f (%d)' % (highMean, highMeanYear)
print 'Lowest mean temperature: %.2f (%d)' % (lowMean, lowMeanYear)

(highStdev, highStdevYear, lowStdev, lowStdevYear) = getExtremeStdevs(weatherData)
print 'Highest standard deviation: %.2f (%d)' % (highStdev, highStdevYear)
print 'Lowest standard deviation: %.2f (%d)' % (lowStdev, lowStdevYear)

(highTemp, highTempYear, lowTemp, lowTempYear) = getExtremeTemps(weatherData)
print 'Highest temperature: %d (%d)' % (highTemp, highTempYear)
print 'Lowest temperature: %d (%d)' % (lowTemp, lowTempYear)