首页 编程教程正文

python爬虫 爬取58同城上所有城市的租房信息详解

piaodoo 编程教程 2020-02-02 12:00:12 881 0 python教程

这篇文章主要介绍了python爬虫 爬取58同城上所有城市的租房信息详解,文中通过示例代码介绍的非常详细,对大家的学习或者工作具有一定的参考学习价值,需要的朋友可以参考下

代码如下

from fake_useragent import UserAgent
from lxml import etree
import requests, os
import time, re, datetime
import base64, json, pymysql
from fontTools.ttLib import TTFont

ua = UserAgent()


class CustomException(Exception):

  def __init__(self, status, msg):
    self.status = status
    self.msg = msg


class City_58:
  '''
  58同城的爬虫类,目前就写这两个
  出租房url: https://cd.58.com/chuzu/     cd代表成都缩写
  二手房url: https://cd.58.com/ershoufang/
  '''

  font_dict = {
    "glyph00001": "0",
    "glyph00002": "1",
    "glyph00003": "2",
    "glyph00004": "3",
    "glyph00005": "4",
    "glyph00006": "5",
    "glyph00007": "6",
    "glyph00008": "7",
    "glyph00009": "8",
    "glyph00010": "9",
  }
  conn = None

  def __init__(self):
    self.session = requests.Session()
    self.session.headers = {
      "user-agent": ua.random
    }
    self.__init__all_city()

  def __init__all_city(self):
    '''获取所有城市的名字及缩写的对应关系'''
    api = "https://www.58.com/changecity.html"
    headers = self.session.headers.copy()
    response = self.session.get(api, headers=headers)
    html = response.text
    res = re.findall("cityList = (.*?)</script>", html, re.S)[0]
    res = re.sub("\s", "", res)
    dic = json.loads(res)
    for k, v in dic.items():
      for k1, v1 in v.items():
        dic[k][k1] = v1.split("|")[0]
    city_dict = {}

    def traverse_dict(dic: dict):
      for k, v in dic.items():
        if k == "海外" or k == "其他":
          continue
        if isinstance(v, dict):
          traverse_dict(v)
        city_dict[k] = v

    traverse_dict(dic)

    other_city = re.findall("independentCityList = (.*?)var", html, re.S)[0]
    res = re.sub("\s", "", other_city)
    other_city_dic = json.loads(res)

    for k, v in other_city_dic.items():
      other_city_dic[k] = v.split("|")[0]

    city_dict.update(other_city_dic)
    self.all_city_dict = city_dict

  def spider_zufang(self, city: str = "成都", is_get_all: bool = True):
    '''爬取租房信息的爬虫方法'''
    assert self.all_city_dict is not None, "获取所有城市信息失败 !"
    format_city = self.all_city_dict.pop(city, None)
    assert format_city is not None, "{}该城市不在爬取城市之内".format(city)
    while True:
      self.city = city
      # self.file = open("./house_info.json", "a", encoding="utf-8")
      start_url = self.__init_zufang(format_city)

      # 思路是什么,首先进入区域的租房页面,在该页面中先提取出相应的title,比如经纪人,个人房源等等...
      # 我们需要构建出相应的url就可以了
      # start_url的格式为 https://cd.58.com/chuzu/ 我们需要转为这样的格式 https://cd.58.com/jintang/hezu/
      # 我们访问转化后的地址,再拿去到相应的链接,比如经纪人,个人房源等链接
      # 拿到该链接以后,这就是这个分类里的第一页url,我们再对这个链接发生请求,
      # 拿到响应体,这里可以写一个while循环,因为我们不知道有多少页,其实也可以知道有多少页,就是在这个响应体中可拿到
      # 我的思路就是写一个while循环,判断是否有下一页,有的继续,没有的话直接break

      for url_info_list in self.__get_url(start_url):
        # 这里的话,最好进行判断一下,因为每个title(值个人房源,品牌公寓等..)不一样的话,可能爬取的策略也不太一样
        title = url_info_list[1]
        if title in ["个人房源", "安选房源", "经纪人", "热租房源"] or "出租" in title:
          self.__spiders_v1(url_info_list)
          # pass
        elif title == "品牌公寓":
          self.__spiders_v2(url_info_list)
          pass
        elif title == "房屋求租":
          # 房屋求租不太想写,数据也不是很多
          pass
        else:
          # 这种情况不在范围内,直接pass掉
          continue
      if not is_get_all:
        return
      try:
        city = list(self.all_city_dict.keys()).pop()
        format_city = self.all_city_dict.pop(city)
      except IndexError:
        print('全国出租房信息,爬取完毕')
        return

  def spider_ershoufang(self, city: str = "cd"):
    '''爬取二手房信息的爬虫方法'''
    pass

  def __spiders_v1(self, url_info_list):
    "负责处理个人房源,安选房源等等页面的方法"
    url = url_info_list[2]
    page_num = 1
    while True:
      time.sleep(2)
      print("正在爬取{}-{}--第{}页数据".format(url_info_list[0], url_info_list[1], page_num))
      response = self.__get_html_source(url)
      # 从html源码中获取到想要的数据
      for house_info_list in self.__deal_with_html_source_v1(response):
        self.__save_to_mysql(house_info_list, url_info_list)
      # 判断是否还有下一页
      next_page_url = self.__is_exist_next_page(response)
      if not next_page_url:
        print("{}-{}爬取完毕".format(url_info_list[0], url_info_list[1]))
        return
      url = next_page_url
      page_num += 1

  def __spiders_v2(self, url_info_list):
    '''处理品牌公寓的爬虫信息'''
    base_url = url_info_list[2]
    format_url = self.__format_url_v2(base_url)
    page_num = 1
    params = None
    while True:
      print("正在爬取{}--第{}页数据...".format(url_info_list[1], page_num))
      time.sleep(2)
      url = format_url.format(page_num)
      response = self.__get_html_source(url, params)
      # 获取到有用的数据 deal_with_html_source_v2
      for house_info_list in self.__deal_with_html_source_v2(response):
        # self.__save_to_file_v2(house_info_list)
        self.__save_to_mysql(house_info_list)

      # 获取到下一页的encryptData
      encryptData = self.__get_html_encryptData(response)

      # 判断是否还有下一页,通过<div class="tip">信息不足,为您推荐附近房源</div>
      if not self.__is_exist_next_page_v2(response):
        print("{}爬取完毕".format(url_info_list[1]))
        return
      page_num += 1
      params = {
        "encryptData": encryptData or "",
        "segment": "true"
      }

  def __save_to_file_v2(self, house_info_list):
    '''
    :param house_info_list: 关于房子的信息的列表
    :param url_info_list: [区域,类型(个人房源,经纪人等等...),url]
    :return:
    '''

    print("房间图片地址>>:", file=self.file)
    print(json.dumps(house_info_list[0], ensure_ascii=False), file=self.file)
    print("房间描述>>:", file=self.file)
    print(json.dumps(house_info_list[1], ensure_ascii=False), file=self.file)
    print("房间详情>>:", file=self.file)
    print(json.dumps(house_info_list[2], ensure_ascii=False), file=self.file)
    print("房间地理位置>>:", file=self.file)
    print(json.dumps(house_info_list[3], ensure_ascii=False), file=self.file)
    print("获取房间的标签>>:", file=self.file)
    print(json.dumps(house_info_list[4], ensure_ascii=False), file=self.file)
    print("获取房间的价格>>:", file=self.file)
    print(json.dumps(house_info_list[5], ensure_ascii=False), file=self.file)
    print(file=self.file)

  def __save_to_mysql(self, house_info_list, url_info_list=None):
    '''保存到数据库'''
    if not self.conn:
      self.conn = pymysql.connect(host="127.0.0.1",
                    port=3306,
                    user="root",
                    password="root",
                    db="city_58")
      self.conn.cursor = self.conn.cursor(cursor=pymysql.cursors.DictCursor)
    if not url_info_list:
      sql = "insert into zu_house_copy (house_img_url,house_title,house_details,house_address,house_tags,hoouse_price,house_type,city) values (%s,%s,%s,%s,%s,%s,%s,%s)"
      house_info_list.append("品牌公寓")
    else:
      sql = "insert into zu_house_copy (house_img_url,house_title,house_details,house_address,house_tags,hoouse_price,area,house_type,city) values (%s,%s,%s,%s,%s,%s,%s,%s,%s)"
      house_info_list.append(url_info_list[0])
      house_info_list.append(url_info_list[1])
    house_info_list.append(self.city)
    row = self.conn.cursor.execute(sql, house_info_list)
    if not row:
      print("插入失败")
    else:
      self.conn.commit()

  def __deal_with_html_source_v1(self, response):
    html = response.text
    self.__get_font_file(html)
    html = self.__format_html_source(html)
    for house_info_list in self.__parse_html_v1(html):
      yield house_info_list

  def __deal_with_html_source_v2(self, response):

    html = response.text
    # 源码里的关于数字0123456789都是进行处理过的,我们需要先获取到字体文件
    # 我们先获取到字体文件并且保存
    self.__get_font_file(html)

    # 对源码中的字体进行处理,得到浏览器显示的数据
    html = self.__format_html_source(html)

    # 开始从页面中提取出想要的数据
    for house_info_list in self.__parse_html_v2(html):
      yield house_info_list

  def __parse_html_v1(self, html):
    xml = etree.HTML(html)

    li_xpath_list = xml.xpath("//ul[@class='listUl']/li[@logr]")

    for li_xpath in li_xpath_list:
      house_info_list = []
      try:
        house_img_url = li_xpath.xpath("div[@class='img_list']/a/img/@lazy_src")[0]
      except IndexError:
        house_img_url = li_xpath.xpath("div[@class='img_list']/a/img/@src")[0]
      house_info_list.append(house_img_url)
      # 房间描述
      house_title = re.sub("\s", "", li_xpath.xpath("div[@class='des']/h2/a/text()")[0])
      house_info_list.append(house_title)
      # 房间详情
      house_details = re.sub("\s", "",
                  li_xpath.xpath("div[@class='des']/p[@class='room strongbox']/text()")[0].strip())
      house_info_list.append(house_details)
      # 房间地理位置
      house_address = re.sub("\s", "",
                  li_xpath.xpath("div[@class='des']/p[@class='add']")[0].xpath("string(.)"))
      house_info_list.append(house_address)
      # 获取房间的标签
      house_tags = "暂无标签"
      house_info_list.append(house_tags)
      # 获取房间的价格
      hoouse_price = re.sub("\s", "",
                 li_xpath.xpath("div[@class='listliright']/div[@class='money']")[0].xpath("string(.)"))
      house_info_list.append(hoouse_price)

      yield house_info_list

  def __parse_html_v2(self, html):
    '''解析页面,拿到数据'''
    xml = etree.HTML(html)
    li_xpath_list = xml.xpath("//ul[@class='list']/li")
    for li_xpath in li_xpath_list:
      house_info_list = []
      # 房间图片地址,这里只获取了一张,我在想要不要获取多张
      # 先空着。。。。。。。。。。。。。
      house_img_url = li_xpath.xpath("a/div[@class='img']/img/@lazy_src")[0]
      house_info_list.append(house_img_url)
      # 房间描述
      house_title = li_xpath.xpath("a/div[@class='des strongbox']/h2/text()")[0].strip()
      house_info_list.append(house_title)
      # 房间详情
      house_details = re.sub("\s", "", li_xpath.xpath("a/div[@class='des strongbox']/p[@class='room']/text()")[0])
      # house_details = li_xpath.xpath("a/div[@class='des strongbox']/p[@class='room']/text()")[0]
      house_info_list.append(house_details)
      # 房间地理位置
      house_address = re.sub("\s", "", li_xpath.xpath(
        "a/div[@class='des strongbox']/p[@class='dist']")[0].xpath("string(.)")) or "暂无地址"
      # house_address = li_xpath.xpath( "a/div[@class='des strongbox']/p[@class='dist']/text()")[0]
      house_info_list.append(house_address)
      # 获取房间的标签
      house_tags = ",".join(li_xpath.xpath("a/div[@class='des strongbox']/p[@class='spec']/span/text()"))
      house_info_list.append(house_tags)
      # 获取房间的价格
      hoouse_price = re.sub("\s", "", li_xpath.xpath("a/div[@class='money']/span[@class='strongbox']")[0].xpath(
        "string(.)")) or "暂无价格"
      house_info_list.append(hoouse_price)

      yield house_info_list

  def __get_font_file(self, html):
    '''从源码中获取到字体文件,并且转为保存,转为TTFont对象'''
    try:
      b64 = re.findall(r"base64,(.*?)\'", html, re.S)[0]
      res = base64.b64decode(b64)
      with open("./online_font.ttf", "wb") as f:
        f.write(res)
      self.online_font = TTFont("./online_font.ttf")
      self.online_font.saveXML("./online.xml")
    except IndexError:
      return

  def __format_html_source(self, html):
    assert self.online_font, "必须创建字体对象"
    assert os.path.exists("./online.xml"), "请先获取到字体文件。"

    with open("./online.xml", "rb") as f:
      file_data = f.read()

    online_uni_list = self.online_font.getGlyphOrder()[1:]
    file_selector = etree.HTML(file_data)
    for uni2 in online_uni_list:
      code = file_selector.xpath("//cmap//map[@name='{}']/@code".format(uni2))[0]
      dd = "&#x" + code[2:].lower() + ";"
      if dd in html:
        html = html.replace(dd, self.font_dict[uni2])
    return html

  def __format_url_v2(self, url):
    '''
    :param url: https://cd.58.com/pinpaigongyu/?from=58_pc_zf_list_ppgy_tab_ppgy
    :return: https://cd.58.com/pinpaigongyu/pn/{}/?from=58_pc_zf_list_ppgy_tab_ppgy
    '''
    a = url.split("?")
    a[0] = a[0] + "pn/{}"
    format_url = "?".join(a)
    return format_url

  def __is_exist_next_page_v2(self, response):
    xml = self.__response_to_xml(response)
    try:
      _ = xml.xpath("//div[@class='tip']")[0]
      return False
    except IndexError:
      return True

  def __get_html_encryptData(self, response):
    html = response.text
    encryptData = re.findall(r"encryptData\":\"(.*?)\"", html, re.S)[0]
    return encryptData

  def __get_url(self, start_url: str):
    url_set = set()
    for area, v in self.area_dict.items():
      url = self.__conversion_url(start_url, v)
      response = self.__get_html_source(url)
      title_dict = self.__get_title_info(response)
      for title_name, v in title_dict.items():
        # 对于求租、品牌公寓这个url,它是重复的,在这里进行判断判断就好了
        if v in url_set:
          continue
        else:
          url_set.add(v)
          yield [area, title_name, v]

  def __conversion_url(self, url: str, area: str):
    '''
    :param url: https://cd.58.com/chuzu/
    :param area:
    :return: https://cd.58.com/区域缩写/chuzu/
    '''
    lis = url.split("/")
    lis.insert(3, area)
    return "/".join(lis)

  def __init_zufang(self, format_city):
    '''首先将所需要的数据的获取到'''
    start_url = "https://{}.58.com/chuzu/".format(format_city)
    headers = self.session.headers.copy()
    response = self.session.get(url=start_url, headers=headers)
    self.__get_area_info(response)
    return start_url

  def __get_html_source(self, url, params=None):
    '''通过get方式获取到网页的源码'''
    time.sleep(1)
    headers = self.session.headers.copy()
    try:
      if not params:
        params = {}
      response = self.session.get(url=url, headers=headers, params=params)
      return response
    except Exception as e:
      with open("./url_log_error.txt", "a", encoding="utf-8") as f:
        f.write(str(datetime.datetime.now()) + "\n")
        f.write(str(e) + "\n")
        f.write("error_url>>:{}".format(url) + "\n")

  def __response_to_xml(self, response):
    try:
      xml = etree.HTML(response.text)
      return xml
    except AttributeError:
      raise CustomException(10000, "response对象转换为xml失败,错误的链接地址为>>:{}".format(response))

  def __is_exist_next_page(self, response):
    '''判断是否存在下一页,存在拿到下一页的链接,不存在返回False'''
    xml = self.__response_to_xml(response)
    try:
      next_page_url = xml.xpath("//a[@class='next']/@href")[0]
      return next_page_url
    except IndexError:
      return False

  def __get_area_info(self, response):
    '''获取到当前城市的区域'''
    xml = self.__response_to_xml(response)
    a_xpath_list = xml.xpath("//dl[@class='secitem secitem_fist']//a[not(@class)]")
    area_key_list = []
    area_value_list = []
    for a_xpath in a_xpath_list:
      area_key_list.append(a_xpath.xpath("text()")[0])
      area_value_list.append(re.findall("com/(.*?)/", a_xpath.xpath("@href")[0])[0])
    assert len(area_key_list) == len(area_value_list), "数据不完整"

    self.area_dict = {k: v for k, v in zip(area_key_list, area_value_list)}

  def __get_title_info(self, response):
    '''获取房屋的分类,比如个人房源,合租房,经纪人,热选房源...'''
    "listTitle"
    xml = self.__response_to_xml(response)
    a_xpath_list = xml.xpath("//div[@class='listTitle']//a[not(@class)]")
    title_key_list = []
    title_value_list = []
    for a_xpath in a_xpath_list:
      title_key_list.append(a_xpath.xpath("span/text()")[0])
      title_value_list.append(a_xpath.xpath("@href")[0])
    assert len(title_key_list) == len(title_value_list), "数据不完整"
    return {k: v for k, v in zip(title_key_list, title_value_list)}
if __name__ == '__main__':
  city_58 = City_58()
  city_58.spider_zufang("重庆")

附上数据库爬取的结果

嗨学网

嗨学网

以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持脚本之家。

版权声明:

本站所有资源均为站长或网友整理自互联网或站长购买自互联网,站长无法分辨资源版权出自何处,所以不承担任何版权以及其他问题带来的法律责任,如有侵权或者其他问题请联系站长删除!站长QQ754403226 谢谢。

有关影视版权:本站只供百度云网盘资源,版权均属于影片公司所有,请在下载后24小时删除,切勿用于商业用途。本站所有资源信息均从互联网搜索而来,本站不对显示的内容承担责任,如您认为本站页面信息侵犯了您的权益,请附上版权证明邮件告知【754403226@qq.com】,在收到邮件后72小时内删除。本文链接:https://www.piaodoo.com/855.html

评论

搜索

游戏网站源码,织梦网站源码,wordpress,wordpress主题,wordpress下载,wordpress插件,wordpress.com,wordpress模板,wordpress教程,wordpress 主题,wordpress安装,wordpress 模板,wordpress 插件,wordpress主题下载,wordpress企业主题,wordpress seo,wordpress主题开发,wordpress theme,wordpress论坛,wordpress 企业主题,wordpress主机,wordpress中文主题,wordpress cms主题,wordpress plugin,wordpress 主题下载,wordpress 主机,wordpress空间,wordpress mu,wordpress 模版,wordpress汉化主题,wordpress淘宝客主题,wordpress 空间,wordpress代码,WORDPRESS HOSTING,wordpress优点,wordpress安卓客户端,wordpress技巧,wordpress换空间,wordpress themes,网站模板,ppt模板网站,模板网站,企业网站模板,网站设计模板,免费网站模板,个人网站模板,ppt模板下载网站,网站模板下载,公司网站模板,门户网站模板,学校网站模板,网站首页模板,网站模板免费下载,旅游网站模板,网站后台模板,免费网站模板下载,传奇网站模板,网站建设模板,外贸网站模板,网站 模板,个人主页网站模板,个人网站模板下载,政府网站模板,音乐网站模板,导航网站模板,免费企业网站模板,企业网站模板下载,手表网站模板,韩国网站模板,汽车网站模板,教育网站模板,网站后台管理模板,班级网站模板,新闻网站模板,房产中介网站模板,旅游网站模板下载,工艺品网站模板,电子商务网站模板,旅游网站设计模板,团购网站模板,flash网站模板,个人网站设计模板,婚庆网站模板,广告公司网站模板,商业网站模板,手机网站模板,免费模板网站推荐,ppt免费模板网站推荐,织梦网站模板,html网站模板建站,网站html模板,免费个人网站模板,公司网站源码,sns源码,彩票网站源码,周易网站源码,源码基地,交友源码,学校网站源码,asp.net 源码,源码天下,jsp网站源码,论坛源码下载,广告联盟源码,建站源码,delphi源码,源码爱好者,酷源码,net源码,源码超市,医疗网站源码,flash源码,搜源码,源码程序,dede源码,新闻网站源码,易语言源码大全,旅游网站源码下载,flash 源码,免费源码论坛,android游戏源码,电脑维修网站源码,30源码网,股票软件源码,卖源码,源码教程,安居客 源码,vip源码,家教源码,.net源码下载,Web源码,网络公司源码,佛教网站源码,android源码学习,房产源码,钓鱼网站源码,775源码屋,web游戏源码,成品网站 源码78w78不用下载,h5游戏网站源码,asp网站源码下载,webgame源码,电子商务网站源码,vb.net源码,乐嘿源码,8a商业源码论坛,fbreader源码,在线客服系统 源码,google源码,.net网站源码,快递查询源码,源码搜藏网,dede整站源码,周易 源码,52源码论坛,财经网站源码,织梦下载站源码,qq钓鱼网站源码,flash游戏源码,房产网源码,源码搜搜,电子商务源码,团购网站源码,团购网源码,jsp源码下载,jsp源码,h站源码,8a源码,婚纱摄影网站源码,易语言盗号源码,x站源码,qq空间psd源码,免费商业源码,笑话网站源码,源码集合,源码家园,啊哦源码,星期六源码,源码熊,阿奇源码,百分百源码网,一手日源码资源,旅行网站源码,b站工程源码泄露,新站长源码,8a商业源码,asp论坛源码,flash源码下载,404源码社区,创业网站源码,php网页源码,易支付源码,成品网站w灬源码,免费CMS成品网站源码,成品网站W灬源码1688仙踪林,成品APP短视频源码下载网站,成品网站源码1688可靠吗,免费B2B网站源码,成品APP直播源码下载,国外儿童网站源码在线,成品网站W灬源码1688,源码,成品网站w灬 源码1688,免费源码网站都有哪些,成品网站源码78W78隐藏通道1,网站源码,源码网,源码网站,源码时代,源码之家,源码下载,php源码,易语言源码,源码论坛,源码是什么,商城源码,论坛源码,源码交易,源码站,源码库,免费源码,免费网站ja**源码大全,ja**源码,成品网站w灬源码1377,a5源码,站长源码,成品网站源码78W78隐藏通道1APP,源码分享,网站源码下载,源码中国,asp源码,源码社区,企业网站源码,php源码下载,成品app直播源码搭建,在线观看视频网站源码2021,旅游网站源码,安卓源码,通达信选股公式源码,神马影院php源码,c#源码,成品网站w灬源码1688网页,php 源码,网页游戏源码,android源码下载,源码吧,视频源码大全,成品短视频APP源码搭建,asp源码下载,私服源码,电脑维修源码,个人主页源码,源码出售,php网站源码,刀客源码,网址导航源码,导航网站源码,源码天空,asp 源码,软件源码,精品源码,成品网站源码1688自动跳转,个人网站源码,源码哥,在线考试系统源码,cms源码,c# 源码,商业源码,vb源码,门户网站源码,音乐网站源码,中国源码,安卓源码下载,asp网站源码,在线客服源码,电影网站源码,免费源码下载,整站源码,源码交易网,易语言源码网,.net源码,在线客服系统源码,淘客源码,卡盟源码,网站源码出售,vb源码下载,莎莎源码,熊猫烧香源码,asp.net源码,商业源码网,外贸网站源码,61源码网,zblog模板,zblog企业模板,帝国cms模板,帝国cms插件,discuz模板