Notion API + Python实现阅读、影视数据可视化图表 - 知乎

import numpy as np
import requests,json
import matplotlib.pyplot as plt
import pyecharts.options as opts
from pyecharts.charts import Pie

class Notion_Data:
    def __init__(self):
        print("欢迎来到Notion数据可视化分析！")

    # 获取用户数据库ID及Token密钥
    # 数据处理
    def Notion_Data_deal(self, Database_ID: str, Token_KEY: str):
        base_url = "https://api.notion.com/v1/databases/"
        """
        接口匹配
        """
        headers = {
            "Authorization": "Bearer " + Token_KEY,
            "accept": "application/json",

        }
        query = {"filter": {"property": "出版社", "checkbox":{"equals":True}}}
        # 获取Notion页面下的详细信息 https://developers.notion.com/reference/post-database-query
        response = requests.post(base_url + Database_ID + "/query", headers=headers, data=query)
        jst = json.loads(response.text)
        return jst

    def Json_Data_deal(self, Database_ID: str, Token_KEY: str, Type: str):  # 类型仅限为：书籍、影片
        dict = self.Notion_Data_deal(Database_ID, Token_KEY)
        """获取到数据列"""
        data_len = len(dict['results'])
        # 统计类型数量，进行后续图形比例
        # 书籍以出版社为划分，影片以类别划分
        if Type == "影片":
            """获取到数据列"""
            dic = {}
            dtc = {}
            for i in range(data_len):
                name = dict['results'][i]['properties']['片名']['title'][0]['plain_text']
                select = dict['results'][i]['properties']['类别']['multi_select']
                classify = []
                for j in range(len(select)):
                    classify.append(dict['results'][i]['properties']['类别']['multi_select'][j]['name'])
                dic[name] = classify  # 类别
            ls = list(dic.items())  # 获取数据数量
            for i in range(len(ls)):
                for j in range(len(ls[i][1])):
                    if ls[i][1][j] in "奇幻":
                        ls[i][1][j] = "科幻"
                    if ls[i][1][j] in "惊悚" or ls[i][1][j] in "悬疑":
                        ls[i][1][j] = "恐怖"
                    if ls[i][1][j] in "故事" or ls[i][1][j] in "扫黑" or ls[i][1][j] in "生活":
                        ls[i][1][j] = "剧情"
                    if ls[i][1][j] in "运动":
                        ls[i][1][j] = "冒险"
                    dtc[ls[i][1][j]] = dtc.get(ls[i][1][j], 0) + 1
            lt = list(dtc.items())
            print("有效数据:" + str(len(dic)))
            return lt

        if Type == "书籍":
            dic = {}
            for i in range(data_len):
                    try:
                        name = (dict['results'][i]['properties']['出版社']['select']['name'])
                        dic[name] = dic.get(name, 0) + 1
                    except Exception:
                        pass
                    continue
            ls = list(dic.items())
            return ls

    def Notion_Visualization(self, Database_ID: str, Token_KEY: str, Type: str):    # 交互式可视化图表
        data = self.Json_Data_deal(Database_ID, Token_KEY, Type)
        lenght = len(data)
        sum = 0
        count_num, name = [], []
        for i in range(lenght):
            sum += int(data[i][1])
            name.append(data[i][0])
        for j in range(lenght):
            a = round(int(data[j][1]) / sum * 100, 2)   # 保留为两位小数
            count_num.append(a)
        np.set_printoptions(precision=2)
        data_pair_temp = [list(data) for data in zip(name, count_num)]
        p = (
            Pie()  # 实例化
                .add(
                series_name=Type,  # 系列名称
                data_pair=data_pair_temp,  # 馈入数据
                radius="65%",  # 饼图半径比例
                center=["50%", "50%"],  # 饼图中心坐标
                label_opts=opts.LabelOpts(is_show=False, position="center"),  # 标签位置
            )
                .set_global_opts(legend_opts=opts.LegendOpts(is_show=True))  # 不显示图示
                .set_series_opts(label_opts=opts.LabelOpts(formatter="{b}: {c}"))  # 标签颜色
                .render(Type + ".html")  # 渲染文件及其名称
            # .render_notebook()
        )
        print("文件已保存在当前程序目录!")

        """
        # 静态可视化图表
        plt.rcParams['font.sans-serif'] = ['Microsoft YaHei']  # 显示中文标签,处理中文乱码问题
        plt.rcParams['axes.unicode_minus'] = False  # 坐标轴负号的处理
        plt.pie(x=count_num, labels=name, autopct='%.2f%%')
        plt.legend(loc='center')
        plt.savefig("./Image/Vedio.png")
        plt.show()
        # print(name, count_num)
        """

if __name__ == '__main__':
    text = Notion_Data()
    Database_ID = str(input("请输入数据库ID:\n"))
    Token_KEY = str(input("请输入Token密钥:\n"))
    Type = str(input("请输入类型(仅限书籍、影片):\n"))
    text.Notion_Visualization(Database_ID, Token_KEY, Type)
    # print(text.Json_Data_deal(Database_ID, Token_KEY, Type))

   def Json_Data_deal(self, Database_ID: str, Token_KEY: str, Type: str):  # Type为获取你要处理的数据库为数据还是影片
        dict = self.Notion_Data_deal(Database_ID, Token_KEY)    # 调用前一个函数
        """获取到数据列"""
        data_len = len(dict['results'])    # 统计类型数量，进行后续图形比例

        # 书籍以出版社为划分，影片以类别划分。
        if Type == "影片":
            """获取到数据列"""
            dic = {}  # 字典存放数据
            for i in range(data_len):  # 循环将数据调用出来
                select = dict['results'][i]['properties']['类别']['multi_select']
                for j in range(len(select)):
                    try:    # 捕捉异常数据并跳过
                        if select[j]['name'] == None:    # 如何该行影片类型为空，则跳过
                            continue
                        else:
                            if select[j]['name'] == "奇幻":
                                select[j]['name'] = "科幻"
                            if select[j]['name'] == "惊悚" or select[j]['name'] == "悬疑":
                                select[j]['name'] = "恐怖"
                            if select[j]['name'] == "故事" or select[j]['name'] == "扫黑" or select[j]['name'] == "生活":
                                select[j]['name'] = "剧情"
                            if select[j]['name'] == "运动":
                                select[j]['name'] = "冒险"
                            dic[select[j]['name']] = dic.get(select[j]['name'], 0) + 1
                    except Exception:
                        pass    # 不处理异常
                    continue    # 跳过
            print("有效数据:" + str(len(dic)))
            return dic

        if Type == "书籍":
            dic = {}
            for i in range(data_len):
                    try:    # 捕捉异常数据并跳过
                        name = (dict['results'][i]['properties']['出版社']['select']['name'])
                        dic[name] = dic.get(name, 0) + 1    # 统计每本数据出版社类型数量，为可视化数据使用
                    except Exception:
                        pass    # 不处理异常
                    continue    # 跳过
            ls = list(dic.items())
            return ls

""" 以下代码为测试用例，可任意修改 """
if __name__ == '__main__':
    text = Notion_Data()
    Database_ID = str(input("请输入数据库ID:\n"))
    Token_KEY = str(input("请输入Token密钥:\n"))
    Type = str(input("请输入数据库类型(书籍/影片):\n"))
    # print(text.Notion_Data_deal(Database_ID, Token_KEY))
    print(text.Json_Data_deal(Database_ID, Token_KEY, Type))