propagation.py

import numpy as np
import matplotlib.pyplot as plt
import psycopg2
import math
from matplotlib.animation import FuncAnimation
import datetime
import csv
from postgis import Polygon,MultiPolygon
from postgis.psycopg import register
import random
import pandas as pd
import queue
import time
from heapq import heappush, heappop


start_time = time.time()


def polygon_to_points(polygon_string):
    xs, ys = [],[]
    points = polygon_string[9:-2].split(',')
    for point in points:
        (x,y) = point.split()
        xs.append(float(x))
        ys.append(float(y))
    return xs,ys

def points_list_to_points(points_list):
    xs = []
    ys = []

    for point in points_list:
        point_string = point[0]
        point_string = point_string[6:-1]
        (x,y) = point_string.split()
        xs.append(float(x))
        ys.append(float(y))
    return xs, ys

conn = psycopg2.connect("dbname=postgres user=postgres")
register(conn)
cursor_psql = conn.cursor()


# Selecionar os primeiros 10 taxis em serviço com origem no Porto
sql = """
        select taxi, ts
        from tracks as tr, cont_aad_caop2018 as caop
        where caop.concelho='PORTO' and
              st_within(st_pointn(tr.proj_track,1), caop.proj_boundary) and
              tr.state='BUSY' order by ts limit 10;
      """

cursor_psql.execute(sql)
results = cursor_psql.fetchall()

i = random.randint(0,len(results))
taxi_porto = results[i-1][0]
print(taxi_porto)

# Selecionar os primeiros 10 taxis em serviço com origem em Lisboa
sql = """
        select taxi, ts
        from tracks as tr, cont_aad_caop2018 as caop
        where caop.concelho='LISBOA' and
              st_within(st_pointn(tr.proj_track,1), caop.proj_boundary) and
              tr.state='BUSY' order by ts limit 10;
      """

cursor_psql.execute(sql)
results = cursor_psql.fetchall()

i = random.randint(0,len(results))
taxi_lisboa = results[i-1][0]
print(taxi_lisboa)

#Para saber a que coluna corresponde cada taxi no ficheiro offsets
sql = " select distinct taxi from tracks order by 1"

cursor_psql.execute(sql)
results = cursor_psql.fetchall()

infetados = queue.Queue(maxsize=1660)
taxi_id = {}
visitados = [0] * 1660
a = 0


for i in results:
    i = int(i[0])
    taxi_id[i] = a
    a = a+1

print(taxi_id[int(taxi_porto)])

#Obter a coluna do taxi escolhido aleatóriamente do Porto e de Lisboa
id_taxi_porto = taxi_id[int(taxi_porto)]
id_taxi_lisboa = taxi_id[int(taxi_lisboa)]


#Criar queue para saber quais taxis já foram infetados
infetados = queue.Queue(maxsize=1660)
infetados.put(id_taxi_porto)
infetados.put(id_taxi_lisboa)
visitados[id_taxi_porto] = 1
visitados[id_taxi_lisboa] = 1


offsets = []

with open('offsets3.csv', 'r') as csvFile:
    reader = csv.reader(csvFile)
    i = 0
    for row in reader:
        l = []
        for j in row:
            x,y = j.split()
            x = float(x)
            y = float(y)
            l.append([x,y])
        offsets.append(l)

infetados = 0
t_infetados = [0] * len(offsets)
anim = [[0 for count in range(1660)] for i in range(len(offsets))]
heap = []
flag_1 = 0
conta = 2


heappush(heap,(0,id_taxi_porto))
heappush(heap,(0,id_taxi_lisboa))

anim[0][id_taxi_porto] = 1
anim[0][id_taxi_lisboa] = 1

t_infetados[0] = 2

while len(heap)>0:
    print("SIZE")
    print(len(heap))

    a = heappop(heap)
    print("Taxi: " + str(a))
    taxi_1 = a[1]
    ts_i = a[0] # Para saber o ts em que foi infetado, pois só apartir daí é que importa
    for k in range(ts_i,len(offsets)):
        anim[k][taxi_1] = 1

    for i in range(ts_i,len(offsets),6):
        p1 = []
        p1 = offsets[i][taxi_1]
        x_p1 = p1[0]
        y_p1 = p1[1]
        for j in range(1660):
            p2 = []
            p2 = offsets[i][j]
            x_p2 = p2[0]
            y_p2 = p2[1]


            if ((x_p1 == 0) and (y_p1 == 0)) or ((x_p2 == 0) and (y_p2 == 0)) or (math.dist(p1,p2)>50) or (visitados[j]==1) or (taxi_1==j):
                continue;


            else:
                prob = random.randint(1,10)
                if prob == 1:
                    visitados[j] = 1
                    heappush(heap, (i,j))
                    infetados = infetados + 1
                    conta = conta + 1
                    flag_1=1
                    anim[i][j] = 1

        if flag_1==1:
            t_infetados[i] = t_infetados[i] + infetados
            flag_1=0

        infetados = 0


with open('anim.csv', 'w', newline='') as file:
    writer = csv.writer(file)
    writer.writerows(anim)


print("Conta " + str(conta))

total = [0] * 24
flag = 0
j = 0
for i in range (0, len(t_infetados)):
    if flag < 360:
        total[j] = total[j] + t_infetados[i]
        flag = flag + 1

    elif flag == 360:
        j = j + 1
        total[j] = total[j] + t_infetados[i]
        flag = 0


print(total)


df=pd.DataFrame({'x': range(0,24), 'y': total })
plt.plot( 'x', 'y', data=df, color='skyblue', linestyle='-',  marker='o')
plt.savefig('infetados_por_hora.png')
plt.show()

print("Duraçao da execuçao: %s" % (time.time() - start_time))