前面有同学参加市场调查大赛，需要西安市的小区信息数据，一个小爬虫程序完美解决。

百度地图开放平台

详情访问：百度PlaceAPI 这里用到了矩形区域检索，具体参数说明如下：

调取详情

所需库： import MySQLdb, time, urllib2, json

定义属性：定义属性：

def __init__(self, location, city, into_db):self.location = locationself.city = cityself.ziduan = ['name', 'address', 'city', 'district', 'scope', 'crawler_time', 'location']self.seq = ['"', '"', '"', '"', '"', '', '"']self.into_db = into_db

抓取过程：

def spider(self):dif = [self.location[0] - self.location[2], self.location[3] - self.location[1]] # 经纬度范围b = [x / 100.0 + self.location[2] for x in xrange(int(dif[0] * 50))]c = [x / 100.0 + self.location[1] for x in xrange(int(dif[1] * 50))]d = [[x, y] for x in b for y in c] # 构建小矩形框cnxn = MySQLdb.connect(host=self.into_db[0], user=self.into_db[1],passwd=self.into_db[2], charset=self.into_db[3]) # 连接到MySQLcursor = cnxn.cursor()# 定义一个MySQL游标sql = "select name from munity_info where city = '{}' ".format(self.city)cursor.execute(sql) # 执行SQL语句url_database = [item[0] for item in cursor.fetchall()] # 取出当前城市已有小区的名字mit()# 提交事务！不然python无法执行sql操作dict_data_list = [] # 字典列表i = 0for x in d: # 遍历当前城市所有划分出来的小矩形html = urllib2.urlopen(r'http://api./place/v2/search?query=小区&bounds={},{},{},{}4&page_size=20&output=json&ak=byEXq8wAVGXWKqOI4sA99ErDAtqlP6Fk'.format(x[0], x[1], x[0] + 0.01, x[1] + 0.01))b = html.read()c = json.loads(b) # json#print 'josn->>', c, '<<--'if not c['results']:continuefor x in c['results']:dict_data = {}dict_data['city'] = self.citydict_data['name'] = x['name'].encode('utf-8', 'ignore')dict_data['address'] = x['address'].encode('utf-8', 'ignore')try:lng_lat = str(x['location']['lng']) + ',' + str(x['location']['lat'])except KeyError:lng_lat = '0.0,0.0'dict_data['location'] = lng_latlng_lat = ','.join(lng_lat.split(',')[::-1])html = urllib2.urlopen(r'http://api./geocoder/v2/?callback=renderReverse&location={}&output=json&pois=1&ak=你的AK码'.format(lng_lat))b = html.read()b = b.split('renderReverse&&renderReverse(')[1][:-1]c = json.loads(b) # jsondict_data['scope'] = c['result']['business'].split(',')[0].encode('utf-8', 'ignore')dict_data['crawler_time'] = str(int(time.time())).encode('utf-8', 'ignore')if not dict_data['scope']:dict_data['scope'] = '其他'dict_data['district'] = c['result']['addressComponent']['district'].encode('utf-8', 'ignore')if not dict_data['district']:dict_data['district'] = '其他'dict_data_list.append(dict_data)cnxn = MySQLdb.connect(host=self.into_db[0], user=self.into_db[1], passwd=self.into_db[2],charset=self.into_db[3])cursor = cnxn.cursor()#print len(dict_data_list)for x in dict_data_list :# 遍历if not x['name'].decode('utf-8') in url_database: # 判断小区是否已经存在# 插入数据sql = "insert into munity_info ({}) values ({})".format(",".join([item for item in self.ziduan]),",".join([j + x[i] + j for j, i in zip(self.seq, self.ziduan)]))cursor.execute(sql)mit() # 同理，提交事务cnxn.close() # 断开连接

主程序入口：

# 主程序入口if __name__ == '__main__':a = [('西安', [34.4438130000, 108.8006150000, 34.1830240000, 109.1546790000])] # 设定西安市区经纬度范围into_db=("localhost","username","user_password","utf8")for x, y in a:example1 = community_info(y, x, into_db)

最后MySQL中数据样例如下：

附上原文作者链接：/qq_22073849/article/details/54429979