Python实战之实现获取动态图表

前言

利用Python实现获取动态图表，废话不多说~

让我们愉快地开始吧~

开发工具

Python版本： 3.6.4

环境搭建

安装Python并添加到环境变量，pip安装需要的相关模块即可。

看一下B站2019年「数据可视化」版块的情况，第一个视频超2百万的播放量，4万+的弹幕

百度指数

获取百度指数，首先需要登陆你的百度账号

以关键词「王者荣耀」为例，时间自定义为2020-10-01～2020-10-10

通过开发者工具，我们就能看到曲线图的数据接口

然而一看请求得到的结果，发现并没有数据，原因是这里使用了JS加密

找到解决方法，成功实现爬取，代码实现

import time
import json
import execjs
import datetime
import requests
from urllib.parse import urlencode
 
 
def get_data(keywords, startDate, endDate, area):
    """
    获取加密的参数数据
    """
    # data_url = "http://index.baidu.com/api/SearchApi/index?area=0&word=[[%7B%22name%22:%22%E7%8E%8B%E8%80%85%E8%8D%A3%E8%80%80%22,%22wordType%22:1%7D]]&startDate=2020-10-01&endDate=2020-10-10"
    params = {
        'word': json.dumps([[{'name': keyword, 'wordType': 1}] for keyword in keywords]),
        'startDate': startDate,
        'endDate': endDate,
        'area': area
    }
    data_url = 'http://index.baidu.com/api/SearchApi/index?' + urlencode(params)
    # print(data_url)
    headers = {
        # 复制登录后的cookie
        "Cookie": '你的cookie',
        "Referer": "http://index.baidu.com/v2/main/index.html",
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.90 Safari/537.36"
    }
 
    # 获取data和uniqid
    res = requests.get(url=data_url, headers=headers).json()
    data = res["data"]["userIndexes"][0]["all"]["data"]
    uniqid = res["data"]["uniqid"]
 
    # 获取js函数中的参数t = "ev-fxk9T8V1lwAL6,51348+.9270-%"
    t_url = "http://index.baidu.com/Interface/ptbk?uniqid={}".format(uniqid)
    rep = requests.get(url=t_url, headers=headers).json()
    t = rep["data"]
    return {"data": data, "t": t}
 
 
def get_search_index(word, startDate, endDate, area):
    """
    获取最终数据
    """
    word = word
    startDate = startDate
    endDate = endDate
    # 调用get_data获取data和uniqid
    res = get_data(word, startDate, endDate, area)
    e = res["data"]
    t = res["t"]
 
    # 读取js文件
    with open('parsing_data_function.js', encoding='utf-8') as f:
        js = f.read()
 
    # 通过compile命令转成一个js对象
    docjs = execjs.compile(js)
 
    # 调用function方法,得到指数数值
    res = docjs.call('decrypt', t, e)
    # print(res)
    return res
 
 
def get_date_list(begin_date, end_date):
    """
    获取时间列表
    """
    dates = []
    dt = datetime.datetime.strptime(begin_date, "%Y-%m-%d")
    date = begin_date[:]
    while date <= end_date:
        dates.append(date)
        dt += datetime.timedelta(days=1)
        date = dt.strftime("%Y-%m-%d")
    return dates
 
 
def get_area():
    areas = {"901": "山东", "902": "贵州", "903": "江西", "904": "重庆", "905": "内蒙古", "906": "湖北", "907": "辽宁", "908": "湖南", "909": "福建", "910": "上海", "911": "北京", "912": "广西", "913": "广东", "914": "四川", "915": "云南", "916": "江苏", "917": "浙江", "918": "青海", "919": "宁夏", "920": "河北", "921": "黑龙江", "922": "吉林", "923": "天津", "924": "陕西", "925": "甘肃", "926": "新疆", "927": "河南", "928": "安徽", "929": "山西", "930": "海南", "931": "台湾", "932": "西藏", "933": "香港", "934": "澳门"}
    for value in areas.keys():
        try:
            word = ['王者荣耀']
            time.sleep(1)
            startDate = '2020-10-01'
            endDate = '2020-10-10'
            area = value
            res = get_search_index(word, startDate, endDate, area)
            result = res.split(',')
            dates = get_date_list(startDate, endDate)
            for num, date in zip(result, dates):
                print(areas[value], num, date)
                with open('area.csv', 'a+', encoding='utf-8') as f:
                    f.write(areas[value] + ',' + str(num) + ',' + date + '\n')
        except:
            pass
 
 
def get_word():
    words = ['诸葛大力', '张伟', '胡一菲', '吕子乔', '陈美嘉', '赵海棠', '咖喱酱', '曾小贤', '秦羽墨']
    for word in words:
        try:
            time.sleep(2)
            startDate = '2020-10-01'
            endDate = '2020-10-10'
            area = 0
            res = get_search_index(word, startDate, endDate, area)
            result = res.split(',')
            dates = get_date_list(startDate, endDate)
            for num, date in zip(result, dates):
                print(word, num, date)
                with open('word.csv', 'a+', encoding='utf-8') as f:
                    f.write(word + ',' + str(num) + ',' + date + '\n')
        except:
            pass
 
 
get_area()
get_word()

得到的CSV文件结果如下，有两种形式的数据

一种是多个关键词每日指数数据，另一种是一个关键词各省市每日指数数据

有了数据就可以用Python制作动图

import pandas as pd
import bar_chart_race as bcr
 
# 读取数据
# df = pd.read_csv('word.csv', encoding='utf-8', header=None, names=['name', 'number', 'day'])
df = pd.read_csv('area.csv', encoding='utf-8', header=None, names=['name', 'number', 'day'])
 
# 数据处理，数据透视表
df_result = pd.pivot_table(df, values='number', index=['day'], columns=['name'], fill_value=0)
 
# 生成GIF
# bcr.bar_chart_race(df_result, filename='word.gif', title='爱情公寓5演职人员热度排行')
bcr.bar_chart_race(df_result, filename='area.gif', title='国内各省市王者荣耀热度排行')

5行Python代码，看看实现的效果

微博指数

百度搜索新浪的微博指数，打开网站一看，发现网页版无法使用

我们只需打开开发者工具，将你的浏览器模拟为手机端，刷新网页即可

可以看到，微指数的界面出来了

添加关键词，查看指数的数据接口

请求是Post方法，并且不需要登陆微博账号

import re
import time
import json
import requests
import datetime
 
 
# 请求头信息
headers = """accept: application/json
accept-encoding: gzip, deflate, br
accept-language: zh-CN,zh;q=0.9
content-length: 50
content-type: application/x-www-form-urlencoded
cookie: '你的cookie'
origin: https://data.weibo.com
referer: https://data.weibo.com/index/newindex?visit_type=trend&wid=1011224685661
sec-fetch-mode: cors
sec-fetch-site: same-origin
user-agent: Mozilla/5.0 (iPhone; CPU iPhone OS 11_0 like Mac OS X) AppleWebKit/604.1.38 (KHTML, like Gecko) Version/11.0 Mobile/15A372 Safari/604.1
x-requested-with: XMLHttpRequest"""
 
# 将请求头字符串转化为字典
headers = dict([line.split(": ",1) for line in headers.split("\n")])
print(headers)
 
# 数据接口
url = 'https://data.weibo.com/index/ajax/newindex/getchartdata'
 
 
# 获取时间列表
def get_date_list(begin_date, end_date):
    dates = []
    dt = datetime.datetime.strptime(begin_date, "%Y-%m-%d")
    date = begin_date[:]
    while date <= end_date:
        dates.append(date)
        dt += datetime.timedelta(days=1)
        date = dt.strftime("%Y-%m-%d")
    return dates
 
 
# 相关信息
names = ['汤唯', '朱亚文', '邓家佳', '乔振宇', '王学圻', '张艺兴', '俞灏明', '吴越', '梁冠华', '李昕亮', '苏可', '孙骁骁', '赵韩樱子', '孙耀琦', '魏巍']
 
 
# 获取微指数数据
for name in names:
    try:
        # 获取关键词ID
        url_id = 'https://data.weibo.com/index/ajax/newindex/searchword'
        data_id = {
            'word': name
        }
        html_id = requests.post(url=url_id, data=data_id, headers=headers)
        pattern = re.compile(r'li wid=\\\"(.*?)\\\" word')
        id = pattern.findall(html_id.text)[0]
 
        # 接口参数
        data = {
            'wid': id,
            'dateGroup': '1month'
        }
        time.sleep(2)
        # 请求数据
        html = requests.post(url=url, data=data, headers=headers)
        result = json.loads(html.text)
        # 处理数据
        if result['data']:
            values = result['data'][0]['trend']['s']
            startDate = '2019-01-01'
            endDate = '2020-01-01'
            dates = result['data'][0]['trend']['x']
            # 保存数据
            for value, date in zip(values, dates):
                print(name, value, date)
                with open('weibo.csv', 'a+', encoding='utf-8') as f:
                    f.write(name + ',' + str(value) + ',' + date + '\n')
    except:
        pass

获取到的信息

也来生成一个动态图表

import pandas as pd
import bar_chart_race as bcr
 
# 读取数据
df = pd.read_csv('weibo.csv', encoding='utf-8', header=None, names=['name', 'number', 'day'])
 
# 数据处理，数据透视表
df_result = pd.pivot_table(df, values='number', index=['day'], columns=['name'], fill_value=0)
# print(df_result[:10])
 
# 生成GIF
bcr.bar_chart_race(df_result[:10], filename='weibo.gif', title='大明风华演职人员热度排行')

结果展示

有喜欢可以尝试动手试试哦~

频道导航