用python爬取历史天气数据的方法示例

def getCity():
  html = reqs.get('https://tianqi.2345.com/js/citySelectData.js').content
  text = html.decode('gbk')
  city = re.findall('([1-5]\d{4})\-[A-Z]\s(.*?)\-\d{5}',text)  #只提取了地级市及以上城市的名称和代码，5以上的是县级市  
  city = list(set(city))                    #去掉重复城市数据
  print('城市列表获取成功')
  return city

def getUrls(cityCode):
  urls = []
  for year in range(2011,2020):
    if year <= 2016:
      for month in range(1, 13):
        urls.append('https://tianqi.数字.com/t/wea_history/js/%s_%s%s.js' % (cityCode,year, month))
    else:
      for month in range(1,13):
        if month<10:
          urls.append('https://tianqi.数字.com/t/wea_history/js/%s0%s/%s_%s0%s.js' %(year,month,cityCode,year,month))          
        else:
          urls.append('https://tianqi.数字.com/t/wea_history/js/%s%s/%s_%s%s.js' %(year,month,cityCode,year,month))
  return urls

def getHtml(url):
  header = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:14.0) Gecko/20100101 Firefox/14.0.1',
       'Referer': '******'}
  request = reqs.get(url,headers = header)
  text = request.content.decode('gbk') #经试解析，这里得用gbk模式
  time.sleep(random.randint(1,3))    #随机暂停，减轻服务器压力
  return text

str1 = "{ymd:'(.*?)',bWendu:'(.*?)℃',yWendu:'(.*?)℃',tianqi:'(.*?)',fengxiang:'(.*?)',fengli:'(.*?)',aqi:'(.*?)',aqiInfo:'(.*?)',aqiLevel:'(.*?)'.*?}"
str2 = "{ymd:'(.*?)',bWendu:'(.*?)℃',yWendu:'(.*?)℃',tianqi:'(.*?)',fengxiang:'(.*?)',fengli:'(.*?)'.*?}"
 
#这个就是偷懒模式，取出来的内容直接存入元组中

def getDf(url):  
  html = getHtml(url)
  pa = re.compile(r'{(ymd.+?)}')           #用'{ymd'打头，把不是每日天气的其它数据忽略掉
  text = re.findall(pa,html)
  list0 = []
  for item in text:
    s = item.split(',')              #分割成每日数据
    d = [i.split(':') for i in s]         #提取冒号前后的数据名称和数据值
    t = {k:v.strip("'").strip('℃') for k,v in d} #用数据名称和数据值构造字典   
    list0.append(t)
  df = pd.DataFrame(list0)              #加入pandas列表中，便于保存
  return df 

def work(city,url):
  con =sql.connect('d:\\天气.db') 
  try:
    df = getDf(url)
    df.insert(0,'城市名称',city)                #新增一列城市名称
    df.to_sql('total', con, if_exists='append', index=False) 
    print(url,'下载完成')    
  except Exception as e:
    print("出现错误:\n",e)
  finally:
    con.commit()
    con.close()