#!/usr/bin/env python3

import csv
import matplotlib.pyplot as plt
import numpy as np

targets = []
hashvalues = []
diff = []
distrib=[]
ratio=[]

with open("headers.csv", "r") as f:
    reader = csv.reader(f, delimiter=";")
    for row in reader:
        if row[0][0] != '#':
            targets.append(int(row[3]))
            hashvalues.append(int(row[5]))
            diff.append(int(row[3]) - int(row[5]))
            ratio.append(int(row[5])/int(row[3]))

plt.plot(hashvalues, '+', label="hash")
plt.plot(targets, 'o', label="target")
plt.legend(loc="upper right")
plt.savefig("target_hash.pdf")

plt.clf()
plt.plot(diff, '+', label="difference")
plt.legend()
plt.savefig("diff.pdf")

sorted_diff = np.sort(diff)
p = 1. * np.arange(len(sorted_diff)) / (len(sorted_diff) - 1)


plt.plot(sorted_diff, p, label="proba de dépassement de k 0")
plt.legend()
plt.xlabel("k")
plt.savefig("distribution.pdf")

ratio=[]
for v in range(len(targets)):
    ratio.append(hashvalues[v]/targets[v])

plt.clf()
plt.plot(ratio)

sorted_ratio = np.sort(ratio)
p = 1. * np.arange(len(sorted_ratio))/(len(sorted_ratio) -1)
plt.clf()
plt.plot(range(len(p)), p, label="$\mathfrak{diff}(b)/\mathfrak{target}(b)$")
plt.legend(loc="lower right")
plt.savefig("distribution_ratio.pdf")