Python爬虫学习——简单爬虫+可视化

简介: Python爬虫学习——简单爬虫+可视化

1. 用到的技术

爬虫   ——> xpath

数据可视化 ——> matplotlib

2. 爬虫

1. 正常爬取

# -*- coding:UTF-8 -*-
import requests
from lxml import etree
import urllib
"""
正常爬取
爬取17173游戏排行榜前1500名
目的网站:
http://top.17173.com/list-0-0-0-0-0-0-0-0-0-0-1.html
"""
def parse_html(url):
    headers = {
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/89.0.4389.90 Safari/537.36"
    }
    resp = requests.get(url, headers=headers)
    return resp.text
"""
解析主页面 
获取主页面需要信息(排名,子页面链接,游戏名称,票数)
"""
def fetch(home_url):
    home_page = etree.HTML(parse_html(home_url))
    lis = home_page.xpath('//div[@class="main-c1"]/div/div[2]/div/div[2]/ul/li')
    for li in lis:
        # 跳过广告
        if li.xpath('./@data-index') == 4:
            continue
        # 获取主页面需要信息(排名,子页面链接,游戏名称,票数)
        rank = li.xpath('./div/div[1]/em/text()')[0]
        child_url = "https:" + li.xpath('./div/div[2]/div/a/@href')[0]
        name = li.xpath('./div/div[2]/div/a/text()')[0]
        votes = li.xpath('./div/div[3]/text()')[0].strip()
        fetch_child(child_url, rank, name, votes)
        # 测试
        # print(
        #     rank,  # 1
        #     name,  # 暗黑破坏神4
        #     votes  # 93303
        # )
"""
解析子页面 
获取子页面需要信息(标签,类型,语言,开发商,注册,运营商)
"""
def fetch_child(child_url, rank, name, votes):
    child_page = etree.HTML(parse_html(child_url))
    game_label = "暂无"
    game_type = "暂无"
    game_language = "暂无"
    game_developer = "暂无"
    game_registered = "暂无"
    game_operator = "暂无"
    # 判断目标网页是否 因不存在而导致跳转到了其他页面
    if len(child_page.xpath('//ul[@class="list-mater-info"]/li')) != 0:
        game_label = "|".join(child_page.xpath('//div[@class="box-mater-cate"]/a/text()'))
        # print(game_label)  # ['PK', '虚幻引擎', 'TPS', '射击']
        # child_ul = child_page.xpath('//ul[@class="list-mater-info"]')
        # 判断游戏类型部分内容是否为空
        if len(child_page.xpath('//ul[@class="list-mater-info"]/li[1]/a')) == 0:
            game_type = "暂无"
        else:
            game_type = child_page.xpath('//ul[@class="list-mater-info"]/li[1]/a/text()')[0]
        game_language = "".join(child_page.xpath('//ul[@class="list-mater-info"]/li[2]/div/span/text()'))
        # 判断开发商部分内容是否为空
        if len(child_page.xpath('//ul[@class="list-mater-info"]/li[3]/span[2]/text()')) != 0:
            game_developer = child_page.xpath('//ul[@class="list-mater-info"]/li[3]/span[2]/text()')[0]
        else:
            game_developer = "暂无"
        # 判断注册部分内容是否为空
        if len(child_page.xpath('//ul[@class="list-mater-info"]/li[4]/span')) == 2:
            game_registered = child_page.xpath('//ul[@class="list-mater-info"]/li[4]/span[2]/text()')[0]
        else:
            a = child_page.xpath('//ul[@class="list-mater-info"]/li[4]/a/@href')[0]
            game_registered = urllib.parse.unquote(a.rsplit("=")[-1])
        game_operator = child_page.xpath('//ul[@class="list-mater-info"]/li[5]//span[2]/text()')[0]
    # 测试
    # print(
    #     game_label,  # PK,虚幻引擎,TPS,射击
    #     game_type,  # 第三人称射击
    #     game_language,  # 简体中文\英语\葡萄牙语\土耳其语
    #     game_developer,  # PUBG Corporation
    #     game_registered,  # 暂无
    #     game_operator,  # PUBG Corporation(中国)
    # )
    msg = [rank, name, votes, game_label, game_type, game_language, game_developer, game_registered, game_operator]
    writer(msg)
"""
信息写入文件
"""
def writer(msg):
    with open("gameRank.csv", mode="a", encoding="GBK", newline="") as f:
        print(msg[1], "开始")
        f.writelines(msg[0] + "," + msg[1] + "," + msg[2] + "," + msg[3] + "," + msg[4] + "," + msg[5] + "," + msg[6] + "," + msg[7] + "," + msg[8] + "\n")
        print(msg[1], "over!")
if __name__ == '__main__':
    for i in range(70):
        i = i + 1
        url = f"https://top.17173.com/list-0-0-0-0-0-0-0-0-0-0-{i}.html"
        fetch(url)

2. 异步协程爬取

from lxml import etree
import urllib
import aiohttp
import aiofiles
import asyncio
"""
使用异步协程进行爬取
爬取17173游戏排行榜前1500名
目的网站:
http://top.17173.com/list-0-0-0-0-0-0-0-0-0-0-1.html
"""
async def parse_html(url):
    headers = {
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.120 Safari/537.36',
    }
    async with aiohttp.ClientSession() as session:
        # 它的使用方式和requests相似,也是在get()方法中添加一个参数,但此时的参数名为proxy,
        # 参数值是字符串,且字符串中的代理协议,只支持http,写成https会报错。
        async with session.get(url, headers=headers) as resp:
            return await resp.text()
"""
函数目标 : 获取需要的信息
"""
async def fetch(child_url, rank, name, votes):
    child_page = etree.HTML(await parse_html(child_url))
    game_label = "暂无"
    game_type = "暂无"
    game_language = "暂无"
    game_developer = "暂无"
    game_registered = "暂无"
    game_operator = "暂无"
    # 判断目标网页是否 因不存在而导致跳转到了其他页面
    if len(child_page.xpath('//ul[@class="list-mater-info"]/li')) != 0:
        game_label = "|".join(child_page.xpath('//div[@class="box-mater-cate"]/a/text()'))
        # print(game_label)  # ['PK', '虚幻引擎', 'TPS', '射击']
        # child_ul = child_page.xpath('//ul[@class="list-mater-info"]')
        # 判断游戏类型部分内容是否为空
        if len(child_page.xpath('//ul[@class="list-mater-info"]/li[1]/a')) == 0:
            game_type = "暂无"
        else:
            game_type = child_page.xpath('//ul[@class="list-mater-info"]/li[1]/a/text()')[0]
        game_language = "".join(child_page.xpath('//ul[@class="list-mater-info"]/li[2]/div/span/text()'))
        # 判断开发商部分内容是否为空
        if len(child_page.xpath('//ul[@class="list-mater-info"]/li[3]/span[2]/text()')) != 0:
            game_developer = child_page.xpath('//ul[@class="list-mater-info"]/li[3]/span[2]/text()')[0]
        else:
            game_developer = "暂无"
        # 判断注册部分内容是否为空
        if len(child_page.xpath('//ul[@class="list-mater-info"]/li[4]/span')) == 2:
            game_registered = child_page.xpath('//ul[@class="list-mater-info"]/li[4]/span[2]/text()')[0]
        else:
            a = child_page.xpath('//ul[@class="list-mater-info"]/li[4]/a/@href')[0]
            game_registered = urllib.parse.unquote(a.rsplit("=")[-1])
        game_operator = child_page.xpath('//ul[@class="list-mater-info"]/li[5]//span[2]/text()')[0]
    # 测试
    print(
        rank,
        name,
        votes,
        game_label,  # PK,虚幻引擎,TPS,射击
        game_type,  # 第三人称射击
        game_language,  # 简体中文\英语\葡萄牙语\土耳其语
        game_developer,  # PUBG Corporation
        game_registered,  # 暂无
        game_operator,  # PUBG Corporation(中国)
    )
    async with aiofiles.open("gameRank.csv", mode="a", encoding="GBK", newline="") as f:
        print(name, "开始")
        await f.writelines(rank + "," + name + "," + votes + "," + game_label + "," + game_type + "," + game_language + "," + game_developer + "," + game_registered + "," + game_operator + "\n")
        print(name, "over!")
async def main():
    tasks = []
    home_page = etree.HTML(await parse_html(home_url))
    lis = home_page.xpath('//div[@class="main-c1"]/div/div[2]/div/div[2]/ul/li')
    for li in lis:
        # 跳过广告
        if li.xpath('./@data-index') == 4:
            continue
        # 获取主页面需要信息(排名,子页面链接,游戏名称,票数)
        rank = li.xpath('./div/div[1]/em/text()')[0]
        a = "https:" + li.xpath('./div/div[2]/div/a/@href')[0]
        name = li.xpath('./div/div[2]/div/a/text()')[0]
        votes = li.xpath('./div/div[3]/text()')[0].strip()
        tasks.append(asyncio.ensure_future(fetch(a, rank, name, votes)))
    await asyncio.wait(tasks)
if __name__ == '__main__':
    # 创建事件循环
    loop = asyncio.get_event_loop()
    for i in range(10):
        home_url = f"https://top.17173.com/list-0-0-0-0-0-0-0-0-0-0-{i}.html"
        loop.run_until_complete(main())
    loop.close()

3. 数据可视化

import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
import pandas as pd
# 汉字字体,优先使用楷体,如果找不到楷体,则使用黑体
mpl.rcParams['font.sans-serif'] = ['KaiTi', 'SimHei', 'FangSong']
# 字体大小
mpl.rcParams['font.size'] = 12
# 正常显示负号
mpl.rcParams['axes.unicode_minus'] = False
# 使用ggplot的绘图风格,这个类似于美化
plt.style.use('ggplot')
# 读取csv文件
csv_data = pd.read_csv("../reptile/gameRank.csv", encoding="GBK")
# 1-游戏类型分布-柱状图
def picture01():
    # 获取 (类型)MMORPG   (该类游戏数量)489
    group_type = csv_data.groupby("type", as_index=False).size()
    # 获取 所有类型对应的游戏的 数量 一个数组 [489  56  14  86 ……]
    number_types = group_type["size"].values
    # 游戏类型
    x = group_type["type"].values
    # 类型对应的游戏的数量
    y = number_types
    # 画图:
    plt.bar(x, +y, facecolor='#ff9999', edgecolor='white')
    # 数据录入:zip指把x,y结合为一个整体,一次可以读取一个x和一个y
    for x, y in zip(x, y):
        plt.text(x, y, y, ha='center', va='bottom')  # 指字体在中间和柱最顶的顶部
    # 轴坐标 竖着显示
    plt.xticks(rotation=270)
    # 设置轴标签以及标题
    plt.xlabel("type")
    plt.ylabel("number")
    plt.title("1-各游戏类型条形图")
    plt.show()
# 2-游戏类型分布-饼图
def picture02():
    plt.axes(aspect='equal')  # 将横、纵坐标轴标准化处理,确保饼图是一个正圆,否则为椭圆
    # 以type为标准 分组 (type size)
    group_type = csv_data.groupby("type", as_index=False).size()
    # 游戏类型 数组
    type_types = group_type["type"].values
    # 游戏类型对应游戏的数量
    number_types = group_type["size"].values
    # 每种类型游戏的个数占比 数组
    pro_types = number_types / number_types.sum()
    plt.pie(
        x=pro_types,  # 绘图数据
        labels=type_types,  # 添加游戏类型标签
        autopct='%.2f%%',  # 设置百分比的格式,这里保留两位小数
        pctdistance=0.8,  # 设置百分比标签与圆心的距离
        labeldistance=1.05,  # 设置游戏类型标签与圆心的距离
        startangle=180,  # 设置饼图的初始角度
        radius=1.1,  # 设置饼图的半径
        counterclock=False,  # 是否逆时针,这里设置为顺时针方向
        wedgeprops={'linewidth': 1.5, 'edgecolor': 'green'},  # 设置饼图内外边界的属性值
        textprops={'fontsize': 5, 'color': 'black'},  # 设置文本标签的属性值
    )
    # 添加图标题
    plt.title('2-各大游戏类型分布')
    # 显示图形
    plt.show()
# 3-游戏隶属厂商TOP20-折线图
def picture03():
    # 数据准备
    group_dev = csv_data.groupby("developer", as_index=False).size()
    group_dev = group_dev.sort_values(by='size', axis=0, ascending=False).drop(index=1046, axis=0)[0:20]
    type_dev = group_dev["developer"].values
    number_dev = group_dev["size"].values
    x = type_dev
    y = number_dev
    plt.xticks(rotation=270)
    plt.title("3-各游戏隶属厂商TOP20")
    plt.plot(x, y)
    plt.show()
# 4-最火游戏类型top10-雷达图
def picture04():
    # 以type为标准 分组 (type size)
    group_type = csv_data.groupby("type", as_index=False).size()
    group_type = group_type.sort_values(by='size', axis=0, ascending=False)[0:10]
    # 游戏类型 数组
    type_types = group_type["type"].values
    # 游数戏类型对应游戏的量
    number_types = group_type["size"].values
    labels = type_types
    values = number_types
    # 设置每个数据点的显示位置,在雷达图上用角度表示
    angles = np.linspace(0, 2 * np.pi, len(values), endpoint=False)
    # 拼接数据首尾,使图形中线条封闭
    values = np.concatenate((values, [values[0]]))
    angles = np.concatenate((angles, [angles[0]]))
    labels = np.concatenate((labels, [labels[0]]))
    # 绘图
    fig = plt.figure()
    # 设置为极坐标格式
    ax = fig.add_subplot(111, polar=True)
    # 绘制折线图
    ax.plot(angles, values, 'o-', linewidth=2)
    # 填充颜色
    ax.fill(angles, values, alpha=0.25)
    # 设置图标上的角度划分刻度,为每个数据点处添加标签
    ax.set_thetagrids(angles * 180 / np.pi, labels)
    # 设置雷达图的范围
    ax.set_ylim(0, 500)
    # 添加标题
    plt.title('4-最火游戏类型top10')
    # 添加网格线
    ax.grid(True)
    plt.show()
# 5-最火游戏top20-横向柱状图
def picture05():
    data = csv_data.sort_values(by='popularity', axis=0, ascending=False)[0:20]
    # 游戏类型
    y = data["game_name"].values[::-1]
    # 类型对应的游戏的数量
    x = data["popularity"].values[::-1]
    # 图像绘制
    fig, ax = plt.subplots()
    b = ax.barh(range(len(y)), x, color='#ff9999')
    # 添加数据标签
    for rect in b:
        w = rect.get_width()
        ax.text(w, rect.get_y() + rect.get_height() / 2, '%d' % int(w), ha='left', va='center')
    # 设置Y轴刻度线标签
    ax.set_yticks(range(len(y)))
    ax.set_yticklabels(y)
    plt.title("5-最火游戏top20")
    plt.show()
# 6-公司游戏版权数量TOP20-横向柱状图
def picture06():
    # 数据准备
    group_dev = csv_data.groupby("developer", as_index=False).size()
    group_dev = group_dev.sort_values(by='size', axis=0, ascending=False).drop(index=1046, axis=0)[0:20]
    y = group_dev["developer"].values[::-1]
    # 类型对应的游戏的数量
    x = group_dev["size"].values[::-1]
    # 图像绘制
    fig, ax = plt.subplots()
    b = ax.barh(range(len(y)), x, color='#ff9999')
    # 添加数据标签
    for rect in b:
        w = rect.get_width()
        ax.text(w, rect.get_y() + rect.get_height() / 2, '%d' % int(w), ha='left', va='center')
    # 设置Y轴刻度线标签
    ax.set_yticks(range(len(y)))
    ax.set_yticklabels(y)
    plt.title("6-公司游戏版权数量TOP20")
    plt.show()
if __name__ == '__main__':
    picture01()
    # picture02()
    # picture03()
    # picture04()
    # picture05()
    # picture06()
相关文章
|
1天前
|
Linux 开发者 iOS开发
|
1天前
|
Shell iOS开发 MacOS
|
1天前
|
数据安全/隐私保护 iOS开发 MacOS
|
1天前
|
数据采集 存储 开发者
Python爬虫实战:打造高效数据采集工具
本文将介绍如何利用Python编写一个高效的网络爬虫,实现对特定网站数据的快速抓取与处理,帮助开发者更好地应对大规模数据采集的需求。
|
2天前
|
数据采集 Web App开发 前端开发
Python爬虫之Ajax分析方法与结果提取#6
Ajax分析方法、Ajax结果提取
9 0
Python爬虫之Ajax分析方法与结果提取#6
|
2天前
|
数据采集 前端开发 JavaScript
|
2天前
|
存储 NoSQL MongoDB
|
2天前
|
SQL 关系型数据库 MySQL
Python爬虫之关系型数据库存储#5
python MySQL 增删改查操作
9 1
|
2天前
|
JSON JavaScript 文件存储
Python爬虫之文件存储#5
TXT文件存储、JSON文件存储、CSV文件存储
23 2
|
2天前
|
数据采集 前端开发 JavaScript
Python爬虫 pyquery库详解#4
pyquery 初始化,基本CSS选择器,查找节点,遍历,节点操作,伪类选择器
11 1

热门文章

最新文章