爬取斗鱼并制图
from selenium import webdriver
import time
import matplotlib.pyplot as plt
import jieba
from wordcloud import WordCloud
from pyecharts.charts import Bar
from pyecharts import options as opts
class DouYuSpider():
def __init__(self):
self.url = 'https://www.douyu.com/g_LOL'
self.driver = webdriver.Chrome(executable_path='F:\chromedriver.exe')
self.dict = {"大于100w": 0, "50w-100w": 0, "30w-50w": 0, "10w-30w": 0, "1w-10w": 0, "小于1w": 0}
self.titles = ''
self.page = 1
def send_request(self, url):
self.driver.get(url)
def parse_content(self):
titles = self.driver.find_elements_by_css_selector('.DyListCover-intro') # 取出标题
hots = self.driver.find_elements_by_css_selector('.DyListCover-hot') # 热度
authors = self.driver.find_elements_by_css_selector('.DyListCover-userName') # 热度
for t in zip(titles, hots, authors):
print(t[0].text, t[1].text, t[2].text)
title = t[0].text
hot = t[1].text
if "万" in hot:
hot = float(hot.split('万')[0]) * 10000
else:
hot = float(hot)
self.cal_hot(hot)
self.titles += title
def create_word_cloud(self, word):
word_list = list(jieba.cut(word))
word_list = [word for word in word_list if len(word) > 1] # 清洗数据
wc = WordCloud(width=800, height=800, font_path='Fonts/font1.ttf')
wc.generate(" ".join(word_list))
wc.to_file('douyu.png')
def cal_hot(self, hot):
if hot >= 1000000:
self.dict['大于100w'] = self.dict['大于100w'] + 1
elif hot >= 500000 and hot < 1000000:
self.dict['50w-100w'] = self.dict['50w-100w'] + 1
elif hot >= 300000 and hot < 500000:
self.dict['30w-50w'] = self.dict['30w-50w'] + 1
elif hot >= 100000 and hot < 300000:
self.dict['10w-30w'] = self.dict['10w-30w'] + 1
elif hot >= 10000 and hot < 100000:
self.dict['1w-10w'] = self.dict['1w-10w'] + 1
else:
self.dict['小于1w'] = self.dict['小于1w'] + 1
def paint_table(self):
bar = (
Bar()
.add_xaxis([str(key) for key in self.dict.keys()])
.add_yaxis("斗鱼热度分析", list(self.dict.values()))
.set_global_opts(title_opts=opts.TitleOpts(title="斗鱼热度分析"))
)
bar.render()
def start(self):
self.send_request(self.url)
while True:
time.sleep(1)
self.driver.execute_script("window.scrollTo(0,document.body.scrollHeight)") # selenium 支持js
self.parse_content()
time.sleep(1)
next = self.driver.find_element_by_class_name('dy-Pagination-next')
if next.get_attribute('aria-disabled') == 'false':
next.click()
# self.page += 1
# if self.page == 2:
# self.paint_table() # 制图
# self.create_word_cloud(self.titles) # 绘制词云
# break
else:
self.paint_table() # 制图
self.create_word_cloud(self.titles) # 绘制词云
break
if __name__ == '__main__':
dys = DouYuSpider()
dys.start()