[Python知识库]python58同城二手房爬取 xpath练习

这大公司的大网站果然这反爬都太厉害了呜呜

这篇文章是xpath的练习项目调了我3个小时，

有很多技巧还是不会，比如是不是有直接获取这个标签下的所有标签，就不需要用嵌套try...except来碰到错误停止了；

是不是能把所要检索的html通过标签单独挑出来来减少检索的时间，导致代码运行有点慢

import pprint
import re
import os
import time
import requests
from lxml import etree

"""
这篇文章是xpath的练习项目调了我3个小时，

有很多技巧还是不会，比如是不是有直接获取这个标签下的所有标签，就不需要用嵌套`try...except`来碰到错误停止了；

是不是能把所要检索的html通过标签单独挑出来来减少检索的时间，导致代码运行有点慢

"""


def get_html(url):
    headers = {
        'User-Agent': "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.77 Safari/537.36",
        'Connection': 'close'
    }

    r = requests.get(url=url, headers=headers)
    r.encoding = r.apparent_encoding
    time.sleep(0.5)
    return r.text


def text_save(filename, text):  # filename为写入CSV文件的路径，data为要写入数据列表.
    file = open(filename, 'w', encoding='UTF-8')
    file.write(text)
    file.close()
    print("保存文件成功")


url = "https://linyi.58.com/ershoufang/"

if not os.path.getsize('./58.html'):
    html = get_html(url)
    text_save('./58.html', html)

h5 = etree.parse('./58.html', etree.HTMLParser())
result = etree.tostring(h5)

# print(html.xpath('//p'))

div_list = []
house = []

xpath_div = '//*[@id="__layout"]/div/section/section[3]/section[1]/section[2]/div[{}]/a/div[2]/'

for i in range(1, 120):
    xpath = xpath_div.format(i)
    div_list.append(xpath)

# pprint.pprint(div_list)
try:
    i = 1
    while 1:
        xpath = xpath_div.format(i)
        info = {}

        # 标题
        title_xpath = xpath + 'div[1]/div[1]/h3/text()'
        title = h5.xpath(title_xpath)[0].encode("ISO-8859-1").decode('utf-8') if len(h5.xpath(title_xpath)) else ''
        info['title'] = title.replace(' ', ',')


        # 房型
        house_shape_xpath = xpath + 'div[1]/section/div[1]/p[1]/span[{}]/text()'
        s = h5.xpath(house_shape_xpath.format(1))[0].encode("ISO-8859-1").decode('utf-8')
        # print(type(s))
        house_shape = s
        try:
            cnt = 2
            while 1:
                s = h5.xpath(house_shape_xpath.format(cnt))[0].encode("ISO-8859-1").decode('utf-8')
                house_shape = house_shape + s
                cnt = cnt + 1
        except:
            pass
        info['house_shape'] = house_shape


        # 面积
        area_xpath = xpath + 'div[1]/section/div[1]/p[2]/text()'
        area = h5.xpath(area_xpath)[0].encode("ISO-8859-1").decode('utf-8') if len(h5.xpath(area_xpath)) else ''
        info['area'] = area.split()[0]
        # pprint.pprint(area.split())


        # 户型朝向
        direction_xpath = xpath + 'div[1]/section/div[1]/p[3]/text()'
        direction = h5.xpath(direction_xpath)[0].encode("ISO-8859-1").decode('utf-8') if len(h5.xpath(direction_xpath)) else ''
        info['direction'] = direction.split()[0]


        # 楼层状况
        floor_info_xpath = xpath + 'div[1]/section/div[1]/p[4]/text()'
        floor_info = h5.xpath(floor_info_xpath)[0].encode("ISO-8859-1").decode('utf-8') if len(h5.xpath(floor_info_xpath)) else ''
        info['floor_info'] = floor_info.split()[0]


        # 建造时间
        build_time_xpath = xpath + 'div[1]/section/div[1]/p[5]/text()'
        # print((h5.xpath(build_time_xpath)))
        build_time = h5.xpath(build_time_xpath)[0].encode("ISO-8859-1").decode('utf-8') if len(h5.xpath(build_time_xpath)) else ''
        info['build_time'] = build_time.split()[0].replace('建造', '') if len(build_time.split()) else ''


        # 小区
        community_xpath = xpath + 'div[1]/section/div[2]/p[1]/text()'
        community = h5.xpath(community_xpath)[0].encode("ISO-8859-1").decode('utf-8') if len(h5.xpath(community_xpath)) else ''
        info['community'] = community.split()[0]


        # 地址
        address_xpath = xpath + 'div[1]/section/div[2]/p[2]/span[{}]/text()'
        _ = []
        try:
            cnt = 1
            while 1:
                s = h5.xpath(address_xpath.format(cnt))[0].encode("ISO-8859-1").decode('utf-8')
                _.append(s.split()[0])
                cnt = cnt + 1
        except:
            pass
        info['address'] = _

        house.append(info)

        i = i + 1
except:
    pass

# pprint.pprint(house)
print(house)