a
p
div
img
#p1
.c1
.c1.c2
p.c1
我是段落
from bs4 import BeautifulSoup
# 1)创建soup(soup对象代表整个网页)
soup = BeautifulSoup('网页源代码', 'lxml')
# 2)获取标签
# soup.select(css选择器)
# soup.select_one(css选择器)
# 标签对象.select(css选择器)
# 标签对象.select_one(css选择器)
# 3)获取标签内容、标签属性
# 标签对象.text
# 标签对象.attrs[属性名]
import requests
from bs4 import BeautifulSoup
from re import sub, findall
import csv
def get_one_page(page):
# 1.获取网页数据
url = f'https://cd.zu.ke.com/zufang/pg{page}/#contentList'
response = requests.get(url)
# 2.解析数据
soup = BeautifulSoup(response.text, 'lxml')
# 获取每个房屋信息对应的div
div_list = soup.select('.content__list--item')
for div in div_list:
name = div.select_one('.twoline').text.strip()
info = div.select_one('.content__list--item--des').text.strip()
info = sub(r'\s+', '', info)
area = findall(r'\d+\.\d+㎡', info)[0]
house_type = findall(r'\d+室\d+厅\d+卫', info)[0]
# address = findall(r'精选/(.+?-.+?)/|^(.+?)/', info)[0]
# address = address[0] if address[0] else address[1]
# print(address)
address = div.select('.content__list--item--des>a')
new_address = '-'.join([x.text for x in address])
price = div.select_one('.content__list--item-price').text
# print(name, new_address, price, area, house_type)
w1.writerow([name, price, area, house_type, new_address])
print('------------------------一页获取完成--------------------')
if __name__ == '__main__':
w1 = csv.writer(open('files/贝壳租房.csv', 'w', encoding='utf-8', newline=''))
w1.writerow(['房屋', '价格', '面积', '户型', '地址'])
for x in range(1, 11):
get_one_page(x)
from selenium.webdriver import Chrome
# 1.创建浏览器对象
b = Chrome()
# 2.打开网页(需要爬那个页面的数据,就打开那个页面对应的网页地址)
b.get('https://movie.douban.com/top250?start=0&filter=')
# 3.获取网页源代码(注意:不管以什么样的方式更新了界面内容,page_source的内容也会更新)
print(b.page_source) # 获取豆瓣电影top250的网页源代码
print('--------------------------------华丽的分割线-------------------------------------')
b.get('https://www.baidu.com')
print(b.page_source)
input('结束:')
from selenium.webdriver import Chrome
# ======================================方法1========================================
1.找到不同页的地址的变化规律,利用循环实现多页数据的请求
b = Chrome()
for x in range(0, 76, 25):
b.get(f'https://movie.douban.com/top250?start={x}&filter=')
print(b.page_source)
print('--------------------------------华丽的分割线-------------------------------------')
input()
# ======================================方法2========================================
2.点击翻页按钮,刷新页面内容,在刷新后获取网页源代码
from selenium.webdriver.common.by import By
b = Chrome()
b.get('https://movie.douban.com/top250?start=0&filter=')
for _ in range(5):
print(b.page_source)
# 点击下一页按钮
next = b.find_element(By.CLASS_NAME, 'next')
# 点击按钮
next.click()
浏览器对象.b.find_element(获取方式, 数据) - 返回符合条件的第一个标签,结果是标签对象
浏览器对象.b.find_elements(获取方式, 数据) - 返回符合条件的所有标签,结果是列表,列表中的元素是标签对象
1)获取方式:
By.ID - 通过ID属性值获取标签
By.CLASS_NAME - 通过class属性值获取标签
By.CSS_SELECTOR - 通过css选择器获取标签
By.LINK_TEXT - 通过a标签的标签内容获取标签
By.PARTIAL_LINK_TEXT - 通过a标签的标签内容获取标签
1)输入框输入内容:输入框对应的标签.send_keys(内容)
2)点击标签:标签对象.click()
from selenium.webdriver.common.by import By
b = Chrome()
b.get('https://www.jd.com/')
# 获取id属性值为key的标签
search = b.find_element(By.ID, 'key')
search.send_keys('电脑\n')
# 获取标签内容为"便宜包邮"的a标签
a1 = b.find_element(By.LINK_TEXT, '便宜包邮')
# a1.click()
# 获取标签内容中包含'口好'的a标签
a2 = b.find_element(By.PARTIAL_LINK_TEXT, '口好')
# a2.click()
input(':')
from selenium.webdriver import Chrome
from selenium.webdriver.common.by import By
import time
from bs4 import BeautifulSoup
def analysis_data(html):
soup = BeautifulSoup(html, 'lxml')
digest = soup.select_one('#ChDivSummary').text
print(digest)
def get_net_data():
# 1.创建浏览器
b = Chrome()
# 2.打开中国知网
b.get('https://www.cnki.net/')
# 3.获取输入框,输入"数据分析"
search = b.find_element(By.ID, 'txt_SearchText')
search.send_keys('数据分析\n')
time.sleep(1)
for _ in range(3):
# 4.获取搜索结果所有论文的标题标签
titles = b.find_elements(By.CLASS_NAME, 'fz14')
for x in titles:
# 点击一个搜索结果
x.click()
time.sleep(1)
# 切换选项卡,让浏览器对象指向详情页
b.switch_to.window(b.window_handles[-1])
# 获取详情页数据, 解析数据
# print(b.page_source)
analysis_data(b.page_source)
# 关闭当前窗口
b.close()
# 将选项卡切换回第一个页面
b.switch_to.window(b.window_handles[0])
print('--------------------一页数据获取完成--------------------------')
b.find_element(By.ID, 'PageNext').click()
time.sleep(4)
input()
if __name__ == '__main__':
get_net_data()
from selenium.webdriver import Chrome
from selenium.webdriver.common.by import By
import time
b = Chrome()
b.get('https://search.jd.com/Search?keyword=%E7%94%B5%E9%A5%AD%E9%94%85&enc=utf-8&wq=%E7%94%B5%E9%A5%AD%E9%94%85&pvid=058303d3cd58499fb8f5f3459afd4d6b')
time.sleep(2)
# =====用代码控制浏览器滚动=====
# js中页面鼓动的代码:window.scrollBy(x方向的偏移量, y方向的偏移量)
# b.execute_script('window.scrollBy(0, 8000)')
for x in range(10):
b.execute_script('window.scrollBy(0, 800)')
time.sleep(1)
time.sleep(2)
result = b.find_elements(By.CSS_SELECTOR, '#J_goodsList>ul>li')
print(len(result))
input('结束:')