''' Purpose: determine a rough estimate of the geocenter of the
        continental USA
method(s):
compute avg lat and long
compute avg between min and max lat/long
take a random sample and find the avg
find farthest point in all 4 directions and avg
'''

# get access to web-based data support
import url

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

USA_DATASET    = CSV_WEB_FOLDER + 'continental-usa.csv'

# get the dataset of interest using get_dataset() from url module
dataset = url.get_dataset( USA_DATASET )

# determine webfile header and location data
header    = dataset[0]  #first row
locations = dataset[1:] #everything else
#background
# min_value = min(locations)
#the above line won't work because location is a list of lists

# determine geocenter
sum_lat = 0
sum_long = 0
list_lat = []
list_long = []
for location in locations: #we are accessing rows here
    latitude = location[3]
    longitude = location[4]
    # sum_lat = sum_lat + latitude
    # sum_long = sum_long + longitude
    list_lat.append(latitude)
    list_long.append(longitude)

#how do we use our accumulators?
sum_lat = sum(list_lat)
sum_long = sum(list_long) #lines 43-44 do the same as lines 37-38
#how do we get avg lat and long?
avg_lat = sum_lat / len(list_lat)
avg_long = sum_long / len(list_long)
print(avg_lat, avg_long) #here's our first guess
#38.52071282671683, -90.43203752861102
# this was in St.Louis, so not quite

#attempt #2
min_lat = min(list_lat)
max_lat = max(list_lat)
min_long = min(list_long)
max_long = max(list_long)
mid_lat = ( max_lat + min_lat ) / 2
mid_long = ( max_long + min_long ) / 2
print(mid_lat, mid_long)