''' Purpose: determine a rough estimate of the geocenter of the
        continental USA
'''

# get access to web-based data support
import url

# specify dataset web location
CSV_WEB_FOLDER = 'http://www.cs.virginia.edu/~cs1112/datasets/csv/'

# dataset name
CONTINENTAL_DATASET = 'usa-continental.csv'

# set link
link    = CSV_WEB_FOLDER + CONTINENTAL_DATASET

# get the dataset of interest
locations = url.get_dataset( link )

# determine geocenter - brainstorming
    # find the median latitude and median longitude
    # find the mean latitude and mean longitude
    # average min() and max() latitude and longitude


# a loop that might be helpful for solving the problem
latitudes = []
longitudes = []
# loop through the dataset to accumulate the columns of interest - latitude and longitude
for location in locations :
    town, state, latitude, longitude = location # we can unpack the rows in this way
    latitudes.append( latitude )
    longitudes.append( longitude )

# now we have a list of latitudes and a list of longitudes

# mean latitude and longitude method:
total_latitude = sum( latitudes )
total_longitude = sum( longitudes )

n = len( locations ) # number of locations is the length of our dataset - number of rows

avg_latitude = total_latitude / n
avg_longitude = total_longitude / n

print( "Average latitude:", avg_latitude )
print( "Average longitude:", avg_longitude )
# these latitudes and longitudes likely do not actually show up in our lists, they were calculated
print()

# median method
sorted_latitude = sorted( latitudes ) # sorted() is a built-in function that does not alter the input list
sorted_longitudes = sorted( longitudes ) # and hands back a NEW sorted list

median_location = n // 2 # needs to be an integer because it's an index

median_latitude = sorted_latitude[ median_location ]
median_longitude = sorted_longitudes[ median_location ]
# these latitudes and longitudes do show up in our list, but there is no guarantee that they correspond to the same
# location

print( "Median latitude:", median_latitude )
print( "Median longitude:", median_longitude )

# look for location corresponding to our guess - we would need conditionals (if statements)
# unlikely that the median latitude and median longitude correspond to the same location in the dataset