# LinearExpLog10.py, D. Parson, September 13, 2023
# 1. Read CSC223f23CSVassn1.csv
# 2. Sort on NPExponent20, which also sorts NPExp20Log2 which is
#    log2(NPExponent20+1) for each row of data.
# 3. Create derived attribute BinNPExp20A, starting at 0 and incrementing
#    by 1 every time a new, larger value for NPExponent20 is encountered in
#    a row.
# 4. Write a CSV output file LinearExpLog10.csv
#    with BinNPExp20A, NPExponent20, NPExp20Log2.

import csv
from math import log10
from random import Random
rgen = Random(1234542)

if __name__ == '__main__':
    outfile = open('LinearExpLog10.csv', 'w')
    outcsv = csv.writer(outfile, delimiter=',', quotechar='"')
    outcsv.writerow(['LinearA', 'LinearB','Exponential', 'Log10'])
    for rowix in range(0, 10000):
        LinearA = rgen.randint(0,100)
        LinearB = rgen.randint(25,75)
        Exponential = 2 ** (LinearA + LinearB + rgen.randint(0,20))
        # rgen.randint(0,20) is for some random noise in the data
        # expand using base 2, compress with base 10
        Log10 = log10(Exponential+1)    # +1 in case of 0 in real data
        outcsv.writerow([LinearA, LinearB, Exponential, Log10])
    outfile.close()