爬取新闻
# 1 爬取网页---requests
# 2 解析
---xml格式,用了re匹配的
---html,bs4,lxml。。。
---json:
-python :内置的
-java : fastjson---》漏洞
-java: 谷歌 Gson
-go :内置 基于反射,效率不高
# pip3.8 install beautifulsoup4
# pip3.8 install lxml
import requests
from bs4 import BeautifulSoup
import pymysql
conn = pymysql.connect(
user='root',
password='123456',
host='127.0.0.1',
database='cars',
)
cursor = conn.cursor()
res = requests.get('https://www.autohome.com.cn/news/1/#liststart')
# print(res.text)
# soup = BeautifulSoup(res.text,'html.parser') # 1.解析的字符串 2.解析器(html.parser内置的)
soup = BeautifulSoup(res.text,'lxml') # 1.解析的字符串 2.解析器 第三方 lxml
# find:找一个 find_all:找所有
ul_list = soup.find_all(name='ul',class_='article')
# print(len(ul_list))
for ul in ul_list:
li_list = ul.find_all(name='li')
# print(len(li_list))
for li in li_list:
h3 = li.find(name='h3')
print(h3)
if h3:
title = h3.text
url = 'https:' + li.find(name='a').attrs.get('href')
desc = li.find(name='p').text
img = li.find(name='img').attrs.get('src')
print("""
新闻标题:%s
新闻图片:%s
新闻地址;%s
新闻摘要:%s
"""%(title,img,url,desc))
# 图片保存到本地
# 数据入库
cursor.execute('insert into news (title,img,url,`desc`) values (%s,%s,%s,%s)',args=[title,img,url,desc])
conn.commit()
bs4介绍遍历文档树
from bs4 import BeautifulSoup
html_doc = """
<html><head><title>The Dormouse's story</title></head>
<body>
<p class="title"><b>The Dormouse's story<span>xxx</span></b><b>hahaha</b></p>
<p class="story">Once upon a time there were three little sisters; and their names were
<a href="http://example.com/elsie" class="sister" id="link1">Elsie</a>
<a href="http://example.com/lacie" class="sister" id="link2">Lacie</a> and
<a href="http://example.com/tillie" class="sister" id="link3">Tillie</a>;
and they lived at the bottom of a well.</p>
<p class="story">...</p>
"""
soup = BeautifulSoup(html_doc,'lxml')
# print(soup.prettify())
'''遍历文档树'''
# 遍历文档树:即直接通过标签名字选择,特点是选择速度快,但如果存在多个相同的标签则只返回第一个
'''1、用法 通过点的方式遍历'''
# a = soup.html.body.a
# a = soup.a
# print(a)
'''2、获取标签的名称'''
# a = soup.a.name
# print(a)
'''3、获取标签的属性'''
# a = soup.a.attrs # {'href': 'http://example.com/elsie', 'class': ['sister'], 'id': 'link1'}
# print(a)
# a = soup.a.attrs.get('id')
# print(a)
# a = soup.a.attrs.get('href')
# print(a)
'''4 获取标签的内容----文本内容'''
# p = soup.p.text # text会把当前标签的子子孙孙的文本内容都拿出来,拼到一起
# print(p)
# s1 = soup.p.string # 当前标签有且只有自己(没有子标签),把文本内容拿出来
# print(s1)
# s2 = soup.p.strings # #generator 把子子孙孙的文本内容放到生成器中
# print(list(s2)) # ["The Dormouse's story", 'xxx', 'hahaha']
'''5 嵌套选择 点完后可以继续再点'''
# print(soup.head.title.text)
'''6 子节点、子孙节点'''
# print(soup.p.contents) # p标签下所有子节点
# print(list(soup.p.children)) # 得到一个迭代器,包含p下所有直接子节点
# print(list(soup.p.descendants)) # p标签下所有子节点及文本内容
'''7 父节点、祖先节点'''
# print(soup.title.parent) # 获取title标签的父节点
# print(list(soup.title.parents)) # 找到title标签所有的祖先节点,父亲的父亲,父亲的父亲的父亲...
'''8 兄弟节点'''
# print(soup.a.next_sibling) # 下一个兄弟
# print(soup.a.previous_sibling) # 上一个兄弟
# print(list(soup.a.next_siblings)) # 下面的兄弟们=>生成器对象
# print(list(soup.a.previous_siblings)) # 上面的兄弟们=>生成器对象
bs4搜索文档树
# find:找到最符合的第一个 find_all:找到所有
# 五种过滤器: 字符串、正则表达式、列表、True、方法
from bs4 import BeautifulSoup
html_doc = """
<html><head><title>The Dormouse's story</title></head>
<body>
<p class="title"><b class ='baby'>The Dormouse's story<span>lqz</span></b><b>adfasdf<b></p>
<p class="story">Once upon a time there were three little sisters; and their names were
<a href="http://example.com/elsie" class="sister" id="link1" xx="xx">Elsie</a>
<a href="http://example.com/lacie" class="sister" id="link2">Lacie</a> and
<a href="http://example.com/tillie" class="sister" id="link3" name="zzz">Tillie</a>;
and they lived at the bottom of a well.</p>
<p class="story">...</p>
"""
soup = BeautifulSoup(html_doc,'lxml')
# 五种过滤器: 字符串、正则表达式、列表、True、方法
'''字符串 通过字符串查找'''
# a = soup.find(name='a')
# a = soup.find_all(name='a',class_='sister')
# a = soup.find_all(name='a',id='link1')
# a = soup.find(string='Elsie').parent
# a=soup.find(href='http://example.com/elsie') # # 括号中可以写 name,id,class_,href,text,所有属性
# a = soup.find(xx='xx')
# a = soup.find(attrs={'class':'sister'}) # 可以通过attrs传属性
# a = soup.find(attrs={'name':'zzz'}) # 可以通过attrs传属性
# print(a)
'''正则表达式'''
# import re
# a = soup.find_all(class_=re.compile('^b'))
# print(a)
# a = soup.find_all(href=re.compile('^http'))
# print(a)
'''列表'''
# a = soup.find_all(name=['b','body','span'])
# print(a)
# a = soup.find_all(class_=['sister','title'])
# print(a)
'''True'''
# a = soup.find_all(href=True)
# print(a)
# a = soup.find_all(name='img',src=True)
# print(a)
'''方法'''
# 查询所有有class但是没有id的标签
"""
find 本质就是find_all find的参数就是find_all的参数,但是find_all比find多
recursive=True:是否递归查找,默认是True,如果写成false,只找第一层,limit=None
"""
# a = soup.find_all(name='html',recursive=False)
# print(a)
# 联合遍历文档树使用
# a = soup.html.p.find(name='b',recursive=False)
# print(a)
# limit=None 限制找几条
# a = soup.find_all(name='a',limit=1)
# print(a)
css选择器
from bs4 import BeautifulSoup
html_doc = """
<html><head><title>The Dormouse's story</title></head>
<body>
<p class="title"><b class ='baby'>The Dormouse's story<span>lqz</span></b><b>adfasdf<b></p>
<p class="story">Once upon a time there were three little sisters; and their names were
<a href="http://example.com/elsie" class="sister" id="link1" xx="xx">Elsie</a>
<a href="http://example.com/lacie" class="sister" id="link2">Lacie</a> and
<a href="http://example.com/tillie" class="sister" id="link3" name="zzz">Tillie</a>;
and they lived at the bottom of a well.</p>
<p class="story">...</p>
"""
soup = BeautifulSoup(html_doc,'lxml')
"""
css选择器
div
.类名
#id号
div a # div下的子子孙孙中得a
div>a #div直接子节点
"""
# res = soup.select('.sister')
# res = soup.select('#link1')
# res = soup.p.find(name='b').select('span')
# print(res)
# 以后,基本所有的解析器都会支持两种解析:css,xpath,都可以去页面中复制
import requests
res = requests.get('http://it028.com/css-selectors.html')
res.encoding = res.apparent_encoding
# res.encoding = 'utf-8'
# print(res.text)
soup = BeautifulSoup(res.text,'lxml')
res = soup.select('#content > table > tbody > tr:nth-child(14) > td:nth-child(3)')
print(res)
selenium基本使用
# requests 发送请求,不能加载ajax
# selenium:直接操作浏览器,不是直接发送http请求,而是用代码控制模拟人操作浏览器的行为,js会自动加载
# appnium :直接操作手机
# 使用步骤(操作什么浏览器:1 谷歌(为例) 2 ie 3 Firefox)
1 下载谷歌浏览器驱动(跟浏览器版本一致)
https://registry.npmmirror.com/binary.html?path=chromedriver/
浏览器版本:114.0.5735.199
驱动版本对应
放到项目路径下
2 代码实现
# pip3.8 install selenium
from selenium import webdriver
import time
from selenium.webdriver.common.by import By
bro = webdriver.Chrome(executable_path='./chromedriver.exe')
bro.get('https://www.baidu.com')
time.sleep(1)
# 有id优先用id找
input_name = bro.find_element(by=By.ID,value='kw')
# 往标签中写内容
input_name.send_keys('性感美女诱惑')
button = bro.find_element(by=By.ID,value='su')
button.click()
time.sleep(3)
bro.close()
模拟登录百度
from selenium import webdriver
import time
from selenium.webdriver.common.by import By
bro = webdriver.Chrome(executable_path='./chromedriver.exe')
bro.get('https://www.baidu.com')
bro.implicitly_wait(10) #隐士等待---》找标签,如果找不到就先等,等10s,如果10s内,标签有了,直接往下执行,如果登录10s还没有,就报错
bro.maximize_window()
# 如果是a标签,可以根据a标签文字找
submit_login = bro.find_element(By.LINK_TEXT,'登录')
submit_login.click()
# 点击短信登录
sms_login = bro.find_element(By.ID,'TANGRAM__PSP_11__changeSmsCodeItem')
sms_login.click()
time.sleep(1)
username_login = bro.find_element(By.ID,'TANGRAM__PSP_11__changePwdCodeItem')
username_login.click()
time.sleep(1)
username = bro.find_element(By.ID,'TANGRAM__PSP_11__userName')
username.send_keys('[email protected]')
password = bro.find_element(By.ID,'TANGRAM__PSP_11__password')
password.send_keys('hahaha123')
login = bro.find_element(By.ID,'TANGRAM__PSP_11__submit')
time.sleep(1)
login.click()
time.sleep(3)
bro.close()
selenium其他用法
# 无头浏览器
# 无界面浏览器(一堆配置)
import time
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
chrome_options = Options()
chrome_options.add_argument('window-size=1920x3000') #指定浏览器分辨率
chrome_options.add_argument('--disable-gpu') #谷歌文档提到需要加上这个属性来规避bug
chrome_options.add_argument('--hide-scrollbars') #隐藏滚动条, 应对一些特殊页面
chrome_options.add_argument('blink-settings=imagesEnabled=false') #不加载图片, 提升速度
chrome_options.add_argument('--headless') #浏览器不提供可视化页面. linux下如果系统不支持可视化不加这条会启动失败
bro = webdriver.Chrome(executable_path='./chromedriver.exe',chrome_options=chrome_options)
bro.get('https://www.douyin.com/')
print(bro.page_source) # 当前页面内容
time.sleep(3)
bro.close()
查找标签
from selenium import webdriver
from selenium.webdriver.common.by import By
bro = webdriver.Chrome(executable_path='./chromedriver.exe') # 打开了浏览器
bro.get('https://www.baidu.com/')
bro.implicitly_wait(10)
# bs4 find和find_all 也支持css
# selenium find_element和 find_elements也支持css和xpath
# bro.find_element(by=By.ID) # 根据id号找一个 id号唯一
# bro.find_element(by=By.NAME) # 根据name属性找一个
# bro.find_element(by=By.TAG_NAME,value='div') # 根据标签名找
# bro.find_element(by=By.LINK_TEXT) # 根据a标签文字
# bro.find_element(by=By.PARTIAL_LINK_TEXT) # 根据a标签文字模糊找
# bro.find_element(by=By.CLASS_NAME) # 根据类名
# bro.find_element(by=By.CSS_SELECTOR) # 根据css选择器找
# bro.find_element(by=By.XPATH) # 根据xpath
bro.find_elements() # 找所有
bro.close()
获取标签属性,文本,大小
'''
print(tag.get_attribute('src')) # 用的最多
tag.text # 文本内容
#获取标签ID,位置,名称,大小(了解)
print(tag.id)
print(tag.location)
print(tag.tag_name)
print(tag.size)
'''
from selenium import webdriver
from selenium.webdriver.common.by import By
bro = webdriver.Chrome(executable_path='./chromedriver.exe')
bro.get('https://www.chaojiying.com/apiuser/login/')
bro.implicitly_wait(10)
tag = bro.find_element(By.CSS_SELECTOR,'body > div.wrapper_danye > div > div.content_login > div.login_form > form > div > img')
print(tag.get_attribute('src')) # 拿属性
print(tag.text) # 拿文本
print(tag.id) # 不是属性id,是selenium提供的id,没有用
print(tag.location) # x,y的坐标 {'x': 648, 'y': 291}
print(tag.tag_name) # 标签名字 img
print(tag.size) # 大小 {'height': 50, 'width': 180}
标签:name,bs4,selenium,bro,soup,使用,print,import,find
From: https://www.cnblogs.com/XxMa/p/17542497.html