编程语言
首页 > 编程语言> > 用 python 爬取房价信息

用 python 爬取房价信息

作者:互联网

       这是我们python课程要求我们制作一个项目,用python爬取结果并作数据展示。

我们使用requests的方法对房价的信息做了爬取,一下就是我们所爬取的网页 

我们主要爬取的内容包括了房价的走势,上月的价格,本月的价格,和历史最高的价格和涨幅,等信息做了爬取并用matplotlib 画出了一个折线图并将其保存下来

import matplotlib.pyplot as plt
import datetime
import requests
import pinyin
import re
import os


def oneyear_m():
    x = []
    y = []
    with open("zoushi.txt", 'r', encoding='utf-8') as data1:
        for line in data1.read().split("\n"):
            data1_line = line.split(":")
            x.append(data1_line[0][5:])
            y.append(int(data1_line[1]))
    plt.figure(figsize=(28, 10))
    plt.title('一年变化图')  # 折线图标题
    plt.rcParams['font.sans-serif'] = ['SimHei']  # 显示汉字
    plt.xlabel('时间')  # x轴标题
    plt.ylabel('价格   (元/㎡)')  # y轴标题
    plt.plot(x, y, marker='o', markersize=5)  # 绘制折线图,添加数据点,设置点的大小
    for a, b in zip(x, y):
        plt.text(a, b, b, ha='center', va='bottom', fontsize=10)  # 设置数据标签位置及大小
    plt.legend(['走势'])  # 设置折线名称
    plt.savefig('一年变化图.jpg')
    plt.show()


def paint(x, y,flag):  # 小区上月价格折线图
    plt.figure(figsize=(10, 5))
    plt.title(flag)  # 折线图标题
    plt.rcParams['font.sans-serif'] = ['SimHei']  # 显示汉字
    plt.xlabel('时间')  # x轴标题
    plt.ylabel('价格   (元/㎡)')  # y轴标题
    plt.plot(x, y, marker='o', markersize=5)  # 绘制折线图,添加数据点,设置点的大小
    for a, b in zip(x, y):
        plt.text(a, b, b, ha='center', va='bottom', fontsize=10)  # 设置数据标签位置及大小
    plt.legend(['方案'])  # 设置折线名称
    plt.savefig(flag+'.jpg')
    plt.show()


def getdata_txt(txt):
    name = []
    lastmon = []
    nowmon = []
    history_max = []
    change = []

    for line in txt.read().split("\n"):
        lines = line.split(" ")
        name.append(lines[0])
        lastmon.append(int(lines[1]))
        nowmon.append(int(lines[2]))
        history_max.append(int(lines[3]))
        front = int(lines[1])
        end = int(lines[2])
        if front > end:
            temp = front - end
            change.append(-round(float(temp / front), 4))
        else:
            temp = end - front
            change.append(round(float(temp / front), 4))
    return name, lastmon, nowmon, history_max, change


def main():
    name = []
    lastmon = []
    nowmon = []
    history_max = []
    change = []
    txt = open("data_up.txt", "r", encoding='utf-8')
    name, lastmon, nowmon, history_max, change = getdata_txt(txt)
    txt = open("data_down.txt", "r", encoding='utf-8')
    name, lastmon, nowmon, history_max, change = getdata_txt(txt)
    paint(name, lastmon,"上月房价图")
    paint(name, nowmon,"本月房价图")
    paint(name, history_max,"历史最高分布图")
    paint(name, change,"增率变化图")

def get_first(s):
    # 通过pinyin.get()拿到一个汉字的拼音,利用切片拿到首个字母
    # 原因:url中间某个字段表示该网页是哪个城市
    ans = ''
    for i in s:
        if i == '重':
            ans = ans + 'c'
        else:
            ans = ans + pinyin.get(i)[0]
    return ans


def get_really_time(time):
    your_dt = datetime.datetime.fromtimestamp(int(time) / 1000)
    return your_dt.strftime("%Y-%m-%d")


class reptile:
    def __init__(self):
        self.__city = '天津'
        self.__header = {
            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/96.0.4664.55 Safari/537.36 Edg/96.0.1054.43'
        }

    def up_data(self, city):
        if city != '':
            self.__city = city
        else:
            print('没有得到新的城市名。')

    def write_in(self, data, *, fileName='', title='', time=False):
        # 数据写入
        flag = False
        with open(fileName, 'w', encoding='utf-8') as fp:
            if not title == '':
                fp.write(title + '\n')
            if time:
                for i, j in data:
                    if flag:
                        fp.write('\n')
                    else:
                        flag = True
                    fp.write(str(get_really_time(i)) + ':')
                    fp.write(str(j))
            else:
                for i, j in data.items():
                    if flag:
                        fp.write('\n')
                    else:
                        flag = True
                    fp.write(i + ' ')
                    for k in j:
                        fp.write(k + ' ')


    def show_all(self):
        oneyear_m()
        main()

    def get_photo_data(self):  # 获取目标城市的总体价格走势图的数据
        url = 'http://' + get_first(self.__city) + '.fangjia.com/trend/yearData?'
        param = {
            'defaultCityName': self.__city,
            'districtName': '',
            'region': '',
            'block': '',
            'keyword': ''
        }
        res = requests.get(url=url, params=param, headers=self.__header).json()
        data = res['series']
        d = data[0]['data']
        # 文件写入
        self.write_in(d, fileName='zoushi.txt', time=True)

    def get_which(self, choose='up'):
        url = 'http://' + get_first(self.__city) + '.fangjia.com/zoushi'

        page_txt = requests.get(url=url, headers=self.__header).text
        if choose == 'up':
            ex = '<div class="trend trend03">.*?<tbody>(.*?)<tbody>'
        else:
            ex = '<div class="trend trend03" style="border-bottom:none;">.*?<tbody>(.*?)</tbody>'
        url_list = str(re.findall(ex, page_txt, re.S)[0])
        ex = '<tr class=".*?">(.*?)</tr>'
        all = str(re.findall(ex, url_list, re.S))
        ex_name = '<td class="td02"><a href=".*?">(.*?)</a></td>'
        ex_data = '<td>(.*?)</td>'
        need_name = re.findall(ex_name, all, re.S)
        need_data = re.findall(ex_data, all, re.S)
        need_data = [i for i in need_data if not i == '元/㎡' and not i == '周度']
        d = {}
        i = 1
        for house_name in need_name:
            d[house_name] = need_data[4 * (i - 1):4 * i]
            i += 1
        self.write_in(d, fileName='data_' + choose + '.txt')


if __name__ == '__main__':  # 程序入口

    a = reptile()
    postion = input("请输入城市\n")
    print(get_first(postion))
    if not os.path.exists (postion):
        os.mkdir(postion)
    os.chdir(postion)
    a.up_data(postion)
    a.get_which()
    a.get_which('down')
    a.get_photo_data()
    a.show_all()

标签:plt,name,get,python,self,爬取,房价,txt,data
来源: https://blog.csdn.net/qq_43216483/article/details/121845971