用python爬疫情数据--basemap

首先我百度到了一份代码：

#%%
 
import time, json, requests
from datetime import datetime
import matplotlib
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
from matplotlib.font_manager import FontProperties
from mpl_toolkits.basemap import Basemap
from matplotlib.patches import Polygon
import numpy as np
import jsonpath
 
plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
 
#%%
 
# 全国疫情地区分布(省级确诊病例)
def catch_cn_disease_dis():
 timestamp = '%d'%int(time.time()*1000)
 url_area = ('https://view.inews.qq.com/g2/getOnsInfo?name=disease_h5'
    '&callback=&_=') + timestamp
 world_data = json.loads(requests.get(url=url_area).json()['data'])
 china_data = jsonpath.jsonpath(world_data, 
         expr='$.areaTree[0].children[*]')
 list_province = jsonpath.jsonpath(china_data, expr='$[*].name')
 list_province_confirm = jsonpath.jsonpath(china_data, expr='$[*].total.confirm')
 dic_province_confirm = dict(zip(list_province, list_province_confirm)) 
 return dic_province_confirm
 
area_data = catch_cn_disease_dis()
print(area_data)
 
#%%
 
# 抓取全国疫情按日期分布
'''
数据源：
"chinaDayList": [{
        "date": "01.13",
        "confirm": "41",
        "suspect": "0",
        "dead": "1",
        "heal": "0"
    }, {
        "date": "01.14",
        "confirm": "41",
        "suspect": "0",
        "dead": "1",
        "heal": "0"
    }
'''
def catch_cn_daily_dis():
 timestamp = '%d'%int(time.time()*1000)
 url_area = ('https://view.inews.qq.com/g2/getOnsInfo?name=disease_h5'
    '&callback=&_=') + timestamp
 world_data = json.loads(requests.get(url=url_area).json()['data'])
 china_daily_data = jsonpath.jsonpath(world_data, 
         expr='$.chinaDayList[*]')
 
 # 其实没必要单独用list存储，json可读性已经很好了；这里这样写仅是为了少该点老版本的代码  
 list_dates = list() # 日期
 list_confirms = list() # 确诊
 list_suspects = list() # 疑似
 list_deads = list() # 死亡
 list_heals = list() # 治愈  
 for item in china_daily_data:
  month, day = item['date'].split('.')
  list_dates.append(datetime.strptime('2020-%s-%s'%(month, day), '%Y-%m-%d'))
  list_confirms.append(int(item['confirm']))
  list_suspects.append(int(item['suspect']))
  list_deads.append(int(item['dead']))
  list_heals.append(int(item['heal']))  
 return list_dates, list_confirms, list_suspects, list_deads, list_heals  
 
list_date, list_confirm, list_suspect, list_dead, list_heal = catch_cn_daily_dis() 
print(list_date)
  
 
#%%
 
# 绘制每日确诊和死亡数据
def plot_cn_daily():
 # list_date, list_confirm, list_suspect, list_dead, list_heal = catch_cn_daily_dis() 
  
 plt.figure('novel coronavirus', facecolor='#f4f4f4', figsize=(10, 8))
 plt.title('全国新型冠状病毒疫情曲线', fontsize=20)
 print('日期元素数：', len(list_date), "\n确诊元素数：", len(list_confirm))
 plt.plot(list_date, list_confirm, label='确诊')
 plt.plot(list_date, list_suspect, label='疑似')
 plt.plot(list_date, list_dead, label='死亡')
 plt.plot(list_date, list_heal, label='治愈')
 xaxis = plt.gca().xaxis 
 # x轴刻度为1天
 xaxis.set_major_locator(matplotlib.dates.DayLocator(bymonthday=None, interval=1, tz=None))
 xaxis.set_major_formatter(mdates.DateFormatter('%m月%d日'))
 plt.gcf().autofmt_xdate() # 优化标注（自动倾斜）
 plt.grid(linestyle=':') # 显示网格
 plt.xlabel('日期',fontsize=16)
 plt.ylabel('人数',fontsize=16)
 plt.legend(loc='best')
  
plot_cn_daily()
 
#%%
 
# 绘制全国省级行政区域确诊分布图
count_iter = 0
def plot_cn_disease_dis():
 # area_data = catch_area_distribution()
 font = FontProperties(fname='res/coure.fon', size=14)
  
 # 经纬度范围
 lat_min = 10 # 纬度
 lat_max = 60
 lon_min = 70 # 经度
 lon_max = 140
   
 # 标签颜色和文本 
 legend_handles = [
    matplotlib.patches.Patch(color='#7FFFAA', alpha=1, linewidth=0),
    matplotlib.patches.Patch(color='#ffaa85', alpha=1, linewidth=0),
    matplotlib.patches.Patch(color='#ff7b69', alpha=1, linewidth=0),
    matplotlib.patches.Patch(color='#bf2121', alpha=1, linewidth=0),
    matplotlib.patches.Patch(color='#7f1818', alpha=1, linewidth=0),
 ]
 legend_labels = ['0人', '1-10人', '11-100人', '101-1000人', '>1000人']
 
 fig = plt.figure(facecolor='#f4f4f4', figsize=(10, 8)) 
 # 新建区域
 axes = fig.add_axes((0.1, 0.1, 0.8, 0.8)) # left, bottom, width, height, figure的百分比,从figure 10%的位置开始绘制, 宽高是figure的80%
 axes.set_title('全国新型冠状病毒疫情地图（确诊）', fontsize=20) # fontproperties=font 设置失败 
 # bbox_to_anchor(num1, num2), num1用于控制legend的左右移动，值越大越向右边移动，num2用于控制legend的上下移动，值越大，越向上移动。
 axes.legend(legend_handles, legend_labels, bbox_to_anchor=(0.5, -0.11), loc='lower center', ncol=5) # prop=font
  
 china_map = Basemap(llcrnrlon=lon_min, urcrnrlon=lon_max, llcrnrlat=lat_min, urcrnrlat=lat_max, resolution='l', ax=axes)
 # labels=[True,False,False,False] 分别代表 [left,right,top,bottom]
 china_map.drawparallels(np.arange(lat_min,lat_max,10), labels=[1,0,0,0]) # 画经度线
 china_map.drawmeridians(np.arange(lon_min,lon_max,10), labels=[0,0,0,1]) # 画纬度线
 china_map.drawcoastlines(color='black') # 洲际线
 china_map.drawcountries(color='red') # 国界线
 china_map.drawmapboundary(fill_color = 'aqua')
 # 画中国国内省界和九段线
 china_map.readshapefile('res/china-shapefiles-master/china', 'province', drawbounds=True)
 china_map.readshapefile('res/china-shapefiles-master/china_nine_dotted_line', 'section', drawbounds=True)
  
 global count_iter
 count_iter = 0
  
 # 内外循环不能对调，地图中每个省的数据有多条(绘制每一个shape，可以去查一下第一条“台湾省”的数据)
 for info, shape in zip(china_map.province_info, china_map.province):
  pname = info['OWNER'].strip('\x00')
  fcname = info['FCNAME'].strip('\x00')
  if pname != fcname: # 不绘制海岛
   continue
  is_reported = False # 西藏没有疫情，数据源就不取不到其数据 
  for prov_name in area_data.keys():    
   count_iter += 1
   if prov_name in pname:
    is_reported = True
    if area_data[prov_name] == 0:
     color = '#f0f0f0'
    elif area_data[prov_name] <= 10:
     color = '#ffaa85'
    elif area_data[prov_name] <= 100:
     color = '#ff7b69'
    elif area_data[prov_name] <= 1000:
     color = '#bf2121'
    else:
     color = '#7f1818'
    break
    
  if not is_reported:
   color = '#7FFFAA'
    
  poly = Polygon(shape, facecolor=color, edgecolor=color)
  axes.add_patch(poly)
   
 
plot_cn_disease_dis()
print('迭代次数', count_iter)

 在这份代码中，先就是import各种库，在mpl_toolkits.basemap上费了不少功夫，解决方案就是basemap的安装参考这位博主的操作（https://blog.csdn.net/moxigandashu/article/details/68945845）。即在http://www.lfd.uci.edu/%7Egohlke/pythonlibs/ 下载basemap和pyproj两个库（注意在下载的时候要看whl的cp版本，我是python3.7，就只能安装cp37以下的版本）安装完成之后可能会遇见奇奇怪怪的版本问题，我遇见了

ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
Traceback (most recent call last):

  File "<ipython-input-1-4fc84cbcc854>", line 2, in <module>
    from mpl_toolkits.basemap import Basemap

  File "D:\anaconda3\lib\site-packages\mpl_toolkits\basemap\__init__.py", line 45, in <module>
    import _geoslib

SystemError: execution of module _geoslib raised unreported exception

 

我用pip install --user --upgrade numpy解决了

 

但是又遇见了：

ModuleNotFoundError: No module named 'mpl_toolkits.basemap'

conda install basemap//等一会输入y,解决上述问题

 KeyError:'PROJ_LIB'

1）找到“epsg”的相应位置，复制文件所在路径（例如我的电脑中“epsg”的位置为C:\ProgramData\Anaconda3\Library\share）
2）设置系统环境变量PROJ_LIB

ModuleNotFoundError: No module named 'jsonpath'

pip install jsonpath

 

'bool' object is not iterable