爬取csdn上的归档部分绘制写作历史

爬取了CSDN上的归档部分的内容，并且绘制写作的历史进程，看看自己的水平。

直接上程序：

# -*- coding: utf-8 -*-
import numpy as np
import matplotlib.pyplot as plt
import requests
import re
from matplotlib.font_manager import FontProperties


def get_html_content():
    html = 'https://blog.csdn.net/***'
    url = requests.get(html)
    print(url)
    data = url.text
    return data


def get_what_i_need():
    data = get_html_content()
    zz = '<span class="count float-right">(.*?)</span>'
    need_content_number = re.findall(zz, data)
    for i in range(len(need_content_number)):
        need_content_number[i] = int(need_content_number[i].replace('篇', ''))
    change_order_num = []
    i = len(need_content_number)
    n = 0
    while i >0:
        change_order_num.append(need_content_number[i-1])
        i -= 1
        n += 1
    need_content_number = change_order_num

    zz_1 = '(.*?)<span class="count float-right">.*?</span>'
    need_content_time = re.findall(zz_1, data)
    temp = []
    for i in need_content_time:
        temp.append(i.strip())
    change_order = []
    i = len(temp)
    n = 0
    while i >0:
        change_order.append(temp[i-1])
        i -= 1
        n += 1
    need_content_time = change_order
    return need_content_time, need_content_number


def plot_result():
    x, y = get_what_i_need()
    X = np.arange(len(x))
    plt.figure('***的CSDN写作历史记录', figsize=(10, 5))
    font = FontProperties(fname=r"c:\windows\fonts\simsun.ttc", size=12)
    plt.bar(x, y)
    for i, j in zip(X, y):
        plt.text(i + 0.05, j + 0.05, '%d' % j, ha='center', va='bottom')
    plt.xticks(fontproperties=font, rotation=-30)
    plt.show()


if __name__ == '__main__':
    plot_result()

I'll be here like you were for me.