statistic.py

import matplotlib.pyplot as plt
import requests
import json
import time
import copy
import pandas as pd
from lxml import etree
from data import dead
from data import suspected
from data import cured
from data import confirmed

# Define color
confirmed_color = '#F74C31'
suspected_color = '#F78207'
dead_color = '#5D7092'
cured_color = '#28B7A3'
confirmed_and_suspected_color = '#2196F3'

# Setup parameters
plt.rcParams['font.family'] = ['Microsoft YaHei']
plt.rcParams['savefig.format'] = 'png'
plt.rcParams['figure.figsize'] = (16.0, 8.0)
plt.rcParams['savefig.dpi'] = 100
plt.rcParams['figure.dpi'] = 100
url = "https://3g.dxy.cn/newh5/view/pneumonia_peopleapp?from=timeline&isappinstalled=0"
request_headers = {
    'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) '
                  'Chrome/79.0.3945.130 Safari/537.36'}

# Send request
response = requests.get(url, headers=request_headers)
response.encoding = 'utf-8'
html = etree.HTML(response.text)

# Pick data
area_stat = html.xpath('//*[@id="getAreaStat"]')
statistics = html.xpath('//*[@id="getStatisticsService"]')
_time = time.strftime('20%y-%m-%d', time.localtime())

# Save data into files
with open('./data/statistics (' + _time + ').json', 'w') as f:
    f.write(statistics[0].text[36:-11])
    f.close()
with open('./data/statistics (' + _time + ').json', 'r') as f:
    data_1 = json.load(f)

# Append data
confirmed[_time[5:]] = data_1['confirmedCount']
suspected[_time[5:]] = data_1['suspectedCount']
cured[_time[5:]] = data_1['curedCount']
dead[_time[5:]] = data_1['deadCount']

# Merge data
confirmed_and_suspected = copy.deepcopy(confirmed)
for key, value in suspected.items():
    if key in confirmed:
        confirmed_and_suspected[key] += value

# Draw graph
f1 = plt.subplot(2, 1, 1)
f2 = plt.subplot(2, 1, 2)
plt.sca(f1)
plt.plot(list(confirmed.keys()), list(confirmed.values()),
         marker='o', label='确诊', color=confirmed_color, linewidth=3)
plt.plot(list(suspected.keys()), list(suspected.values()),
         marker='v', label='疑似', color=suspected_color, linewidth=3)
plt.plot(list(confirmed_and_suspected.keys()), list(confirmed_and_suspected.values()),
         marker='P', label='疑似与疑似', color=confirmed_and_suspected_color, linewidth=3)
plt.title('2019-nCov 全国疫情新增趋势图 2019-01-11 至 ' + _time)
plt.legend()
plt.sca(f2)
plt.plot(list(cured.keys()), list(cured.values()), marker='s',
         label='治愈', color=cured_color, linewidth=3)
plt.plot(list(dead.keys()), list(dead.values()), marker='x',
         label='死亡', color=dead_color, linewidth=3)
title1 = '2019-nCov 全国趋势图 ' + _time
plt.title('2019-nCov 全国疫情死亡/治愈累计趋势图 2019-01-11 至 ' + _time)
plt.legend()
plt.savefig('./view/{0}.png'.format(title1))

# Save data into files
with open('./data/area_stat (' + _time + ').json', 'w') as f:
    f.write(area_stat[0].text[27:-11])
    f.close()
data_2 = json.loads(area_stat[0].text[27:-11])

# Draw graph
provinceShortName = []
confirmedCount = []
curedCount = []
deadCount = []
for i in data_2:
    provinceShortName.append(i['provinceShortName'])
    confirmedCount.append(i['confirmedCount'])
    curedCount.append(i['curedCount'])
    deadCount.append(i['deadCount'])
provinceShortName.reverse()
confirmedCount.reverse()
curedCount.reverse()
deadCount.reverse()
df = pd.DataFrame({
    '确诊': pd.Series(confirmedCount, index=provinceShortName),
    '治愈': pd.Series(curedCount, index=provinceShortName),
    '死亡': pd.Series(deadCount, index=provinceShortName)
})
df.plot.barh(stacked=True, color=[confirmed_color, cured_color, dead_color])
title2 = '2019-nCov 省际统计图 ' + _time
plt.title(title2)
plt.savefig('./view/{0}.png'.format(title2))
plt.show()