亲宝软件园·资讯

展开

Python获取房价信息

Python 集中营 人气:0

整个数据获取的信息是通过房源平台获取的,通过下载网页元素并进行数据提取分析完成整个过程

导入相关的网页下载、数据解析、数据处理库

from fake_useragent import UserAgent  # 身份信息生成库

from bs4 import BeautifulSoup  # 网页元素解析库
import numpy as np  # 科学计算库
import requests  # 网页下载库
from requests.exceptions import RequestException  # 网络请求异常库
import pandas as pd  # 数据处理库

然后,在开始之前初始化一个身份信息生成的对象,用于后面随机生成网页下载时的身份信息。

user_agent = UserAgent()

编写一个网页下载函数get_html_txt,从相应的url地址下载网页的html文本。

def get_html_txt(url, page_index):
    '''
    获取网页html文本信息
    :param url: 爬取地址
    :param page_index:当前页数
    :return:
    '''
    try:
        headers = {
            'user-agent': user_agent.random
        }
        response = requests.request("GET", url, headers=headers, timeout=10)
        html_txt = response.text
        return html_txt
    except RequestException as e:
        print('获取第{0}页网页元素失败!'.format(page_index))
        return ''

编写网页元素处理函数catch_html_data,用于解析网页元素,并将解析后的数据元素保存到csv文件中。

def catch_html_data(url, page_index):
    '''
    处理网页元素数据
    :param url: 爬虫地址
    :param page_index:
    :return:
    '''

    # 下载网页元素
    html_txt = str(get_html_txt(url, page_index))

    if html_txt.strip() != '':

        # 初始化网页元素对象
        beautifulSoup = BeautifulSoup(html_txt, 'lxml')

        # 解析房源列表
        h_list = beautifulSoup.select('.resblock-list-wrapper li')

        # 遍历当前房源的详细信息
        for n in range(len(h_list)):
            h_detail = h_list[n]

            # 提取房源名称
            h_detail_name = h_detail.select('.resblock-name a.name')
            h_detail_name = [m.get_text() for m in h_detail_name]
            h_detail_name = ' '.join(map(str, h_detail_name))

            # 提取房源类型
            h_detail_type = h_detail.select('.resblock-name span.resblock-type')
            h_detail_type = [m.get_text() for m in h_detail_type]
            h_detail_type = ' '.join(map(str, h_detail_type))

            # 提取房源销售状态
            h_detail_status = h_detail.select('.resblock-name span.sale-status')
            h_detail_status = [m.get_text() for m in h_detail_status]
            h_detail_status = ' '.join(map(str, h_detail_status))

            # 提取房源单价信息
            h_detail_price = h_detail.select('.resblock-price .main-price .number')
            h_detail_price = [m.get_text() for m in h_detail_price]
            h_detail_price = ' '.join(map(str, h_detail_price))

            # 提取房源总价信息
            h_detail_total_price = h_detail.select('.resblock-price .second')
            h_detail_total_price = [m.get_text() for m in h_detail_total_price]
            h_detail_total_price = ' '.join(map(str, h_detail_total_price))

            h_info = [h_detail_name, h_detail_type, h_detail_status, h_detail_price, h_detail_total_price]
            h_info = np.array(h_info)
            h_info = h_info.reshape(-1, 5)
            h_info = pd.DataFrame(h_info, columns=['房源名称', '房源类型', '房源状态', '房源均价', '房源总价'])
            h_info.to_csv('北京房源信息.csv', mode='a+', index=False, header=False)

        print('第{0}页房源信息数据存储成功!'.format(page_index))
    else:
        print('网页元素解析失败!')

编写多线程处理函数,初始化网络网页下载地址,并使用多线程启动调用业务处理函数catch_html_data,启动线程完成整个业务流程。

import threading  # 导入线程处理模块


def thread_catch():
    '''
    线程处理函数
    :return:
    '''
    for num in range(1, 50, 3):
        url_pre = "https://bj.fang.lianjia.com/loupan/pg{0}/".format(str(num))
        url_cur = "https://bj.fang.lianjia.com/loupan/pg{0}/".format(str(num + 1))
        url_aft = "https://bj.fang.lianjia.com/loupan/pg{0}/".format(str(num + 2))

        thread_pre = threading.Thread(target=catch_html_data, args=(url_pre, num))
        thread_cur = threading.Thread(target=catch_html_data, args=(url_cur, num + 1))
        thread_aft = threading.Thread(target=catch_html_data, args=(url_aft, num + 2))
        thread_pre.start()
        thread_cur.start()
        thread_aft.start()


thread_catch()

数据存储结果展示效果

加载全部内容

相关教程
猜你喜欢
用户评论