今天是实训的第八天,主要还是围绕爬虫来讲的。
今日笔记:
1.解析库之bs4
''''''
'''
pip3 install beautifulsoup4 # 安装bs4
pip3 install lxml # 下载lxml解析器
'''
html_doc = """
<html><head><title>The Dormouse's story</title></head>
<body>
<p class="sister"><b>$37</b></p>
<p class="story" id="p">Once upon a time there were three little sisters; and their names were
<a href="http://example.com/elsie" class="sister" >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>
"""
# 从bs4中导入BeautifulSoup
from bs4 import BeautifulSoup
# 调用BeautifulSoup实例化得到一个soup对象
# 参数一: 解析文本
# 参数二:
# 参数二: 解析器(html.parser、lxml...)
soup = BeautifulSoup(html_doc, 'lxml')
print(soup)
print('*' * 100)
print(type(soup))
print('*' * 100)
# 文档美化
html = soup.prettify()
print(html)
2.bs之遍历文档树:
html_doc = """<html><head><title>The Dormouse's story</title></head><body><p class="sister"><b>$37</b></p><p class="story" id="p">Once upon a time there were three little sisters; and their names were<b>tank</b><a href="http://example.com/elsie" class="sister" >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.<hr></hr></p><p class="story">...</p>"""
from bs4 import BeautifulSoup
soup = BeautifulSoup(html_doc, 'lxml')
'''
遍历文档树:
1、直接使用
2、获取标签的名称
3、获取标签的属性
4、获取标签的内容
5、嵌套选择
6、子节点、子孙节点
7、父节点、祖先节点
8、兄弟节点
'''
# 1、直接使用
print(soup.p) # 查找第一个p标签
print(soup.a) # 查找第一个a标签
# 2、获取标签的名称
print(soup.head.name) # 获取head标签的名称
# 3、获取标签的属性
print(soup.a.attrs) # 获取a标签中的所有属性
print(soup.a.attrs['href']) # 获取a标签中的href属性
# 4、获取标签的内容
print(soup.p.text) # $37
# 5、嵌套选择
print(soup.html.head)
# 6、子节点、子孙节点
print(soup.body.children) # body所有子节点,返回的是迭代器对象
print(list(soup.body.children)) # 强转成列表类型
print(soup.body.descendants) # 子孙节点
print(list(soup.body.descendants)) # 子孙节点
# 7、父节点、祖先节点
print(soup.p.parent) # 获取p标签的父亲节点
# 返回的是生成器对象
print(soup.p.parents) # 获取p标签所有的祖先节点
print(list(soup.p.parents))
# 8、兄弟节点
# 找下一个兄弟
print(soup.p.next_sibling)
# 找下面所有的兄弟,返回的是生成器
print(soup.p.next_siblings)
print(list(soup.p.next_siblings))
# 找上一个兄弟
print(soup.a.previous_sibling) # 找到第一个a标签的上一个兄弟节点
# 找到a标签上面的所有兄弟节点
print(soup.a.previous_siblings) # 返回的是生成器
print(list(soup.a.previous_siblings))
3.bs之搜索文档树:
''''''
html_doc = """<html><head><title>The Dormouse's story</title></head><body><p class="sister"><b>$37</b></p><p class="story" id="p">Once upon a time there were three little sisters; and their names were<b>tank</b><a href="http://example.com/elsie" class="sister" >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.<hr></hr></p><p class="story">...</p>"""
'''
搜索文档树:
find() 找一个
find_all() 找多个
标签查找与属性查找:
标签:
name 属性匹配
attrs 属性查找匹配
text 文本匹配
- 字符串过滤器
字符串全局匹配
- 正则过滤器
re模块匹配
- 列表过滤器
列表内的数据匹配
- bool过滤器
True匹配
- 方法过滤器
用于一些要的属性以及不需要的属性查找。
属性:
- class_
- id
'''
from bs4 import BeautifulSoup
soup = BeautifulSoup(html_doc, 'lxml')
# # 字符串过滤器
# # name
p_tag = soup.find(name='p')
print(p_tag) # 根据文本p查找某个标签
# 找到所有标签名为p的节点
tag_s1 = soup.find_all(name='p')
print(tag_s1)
#
#
# # attrs
# # 查找第一个class为sister的节点
p = soup.find(attrs={"class": "sister"})
print(p)
# # 查找所有class为sister的节点
tag_s2 = soup.find_all(attrs={"class": "sister"})
print(tag_s2)
#
#
# # text
text = soup.find(text="$37")
print(text)
#
#
# # 配合使用:
# # 找到一个id为link2、文本为Lacie的a标签
a_tag = soup.find(name="a", attrs={"id": "link2"}, text="Lacie")
print(a_tag)
# # 正则过滤器
import re
# name
p_tag = soup.find(name=re.compile('p'))
print(p_tag)
# 列表过滤器
import re
# name
tags = soup.find_all(name=['p', 'a', re.compile('html')])
print(tags)
# - bool过滤器
# True匹配
# 找到有id的p标签
p = soup.find(name='p', attrs={"id": True})
print(p)
# 方法过滤器
# 匹配标签名为a、属性有id没有class的标签
def have_id_class(tag):
if tag.name == 'a' and tag.has_attr('id') and tag.has_attr('class'):
return tag
tag = soup.find(name=have_id_class)
print(tag)
4.爬取豌豆荚
主页:
图标地址、下载次数、大小、详情页地址
详情页:
游戏名、图标名、好评率、评论数、小编点评、简介、网友评论、1-5张截图链接地址、下载地址
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=1&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=2&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=3&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
32
'''
import requests
from bs4 import BeautifulSoup
# 1、发送请求
def get_page(url):
response = requests.get(url)
return response
# 2、开始解析
# 解析主页
def parse_index(data):
soup = BeautifulSoup(data, 'lxml')
# 获取所有app的li标签
app_list = soup.find_all(name='li', attrs={"class": "card"})
for app in app_list:
# print('tank *' * 1000)
# print(app)
# 图标地址
img = app.find(name='img').attrs['data-original']
print(img)
# 下载次数
down_num = app.find(name='span', attrs={"class": "install-count"}).text
print(down_num)
import re
# 大小
size = soup.find(name='span', text=re.compile("\d+MB")).text
print(size)
# 详情页地址
detail_url = soup.find(name='a', attrs={"class": "detail-check-btn"}).attrs['href']
print(detail_url)
def main():
for line in range(1, 33):
url = f"https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page={line}&ctoken=FRsWKgWBqMBZLdxLaK4iem9B"
# 1、往app接口发送请求
response = get_page(url)
# print(response.text)
print('*' * 1000)
# 反序列化为字典
data = response.json()
# 获取接口中app标签数据
app_li = data['data']['content']
# print(app_li)
# 2、解析app标签数据
parse_index(app_li)
if __name__ == '__main__':
main()
主页:
图标地址、下载次数、大小、详情页地址
详情页:
游戏名、好评率、评论数、小编点评、下载地址、简介、网友评论、1-5张截图链接地址、
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=1&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=2&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=3&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
32
'''
import requests
from bs4 import BeautifulSoup
# 1、发送请求
def get_page(url):
response = requests.get(url)
return response
# 2、开始解析
# 解析详情页
def parse_detail(text):
soup = BeautifulSoup(text, 'lxml')
# print(soup)
# app名称
name = soup.find(name="span", attrs={"class": "title"}).text
# print(name)
# 好评率
love = soup.find(name='span', attrs={"class": "love"}).text
# print(love)
# 评论数
commit_num = soup.find(name='a', attrs={"class": "comment-open"}).text
# print(commit_num)
# 小编点评
commit_content = soup.find(name='div', attrs={"class": "con"}).text
# print(commit_content)
# app下载链接
download_url = soup.find(name='a', attrs={"class": "normal-dl-btn"}).attrs['href']
# print(download_url)
print(
f'''
============= tank ==============
app名称:{name}
好评率: {love}
评论数: {commit_num}
小编点评: {commit_content}
app下载链接: {download_url}
============= end ==============
'''
)
# 解析主页
def parse_index(data):
soup = BeautifulSoup(data, 'lxml')
# 获取所有app的li标签
app_list = soup.find_all(name='li', attrs={"class": "card"})
for app in app_list:
# print(app)
# print('tank' * 1000)
# print('tank *' * 1000)
# print(app)
# 图标地址
# 获取第一个img标签中的data-original属性
img = app.find(name='img').attrs['data-original']
print(img)
# 下载次数
# 获取class为install-count的span标签中的文本
down_num = app.find(name='span', attrs={"class": "install-count"}).text
print(down_num)
import re
# 大小
# 根据文本正则获取到文本中包含 数字 + MB(\d+代表数字)的span标签中的文本
size = soup.find(name='span', text=re.compile("\d+MB")).text
print(size)
# 详情页地址
# 获取class为detail-check-btn的a标签中的href属性
# detail_url = soup.find(name='a', attrs={"class": "name"}).attrs['href']
# print(detail_url)
# 详情页地址
detail_url = app.find(name='a').attrs['href']
print(detail_url)
# 3、往app详情页发送请求
response = get_page(detail_url)
# 4、解析app详情页
parse_detail(response.text)
def main():
for line in range(1, 33):
url = f"https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page={line}&ctoken=FRsWKgWBqMBZLdxLaK4iem9B"
# 1、往app接口发送请求
response = get_page(url)
# print(response.text)
print('*' * 1000)
# 反序列化为字典
data = response.json()
# 获取接口中app标签数据
app_li = data['data']['content']
# print(app_li)
# 2、解析app标签数据
parse_index(app_li)
if __name__ == '__main__':
main()
5.mongoDB的简单使用:
MongoDB 非关系型数据库
一 安装与使用
1、下载安装
https://www.mongodb.com/download-center/community
2、在C盘创建一个data/db文件夹
- 数据的存放路径
3、mongod启动服务
进入终端,输入mongod启动mongoDB服务。
4、mongo进入mongoDB客户端
打开一个新的终端,输入mongo进入客户端
二 数据库操作
数据库操作:
切换库:
SQL:
use admin; 有则切换,无则报错。
MongoDB:
use tank; 有则切换,无则创建,并切换tank库中。
查数据库:
SQL:
show databases;
MongoDB:
show dbs;
显示的数据库若无数据,则不显示。
删除库:
SQL:
drop database
MongoDB:
db.dropDatabase()
集合操作: MySQL中叫做表。
创建集合:
SQL:
create table f1, f2...
MongoDB:
# 在当前库中通过.来创建集合
db.student
插入数据:
# 插入多条数据
db.student.insert([{"name1": "tank1"}, {"name2": "tank2"}])
# 插入一条
db.student.insert({"name": "tank"})
查数据:
# 查找student集合中所有数据
db.student.find({})
# 查一条 查找name为tank的记录
db.student.find({"name":"tank"})
三 python链接MongoDB
1、下载第三方模块pymongo
pip3 install pymongo
2、链接mongoDB客户端
client = MongoClient('localhost', 27017)
6.pymongo简单使用:
from pymongo import MongoClient
# 1、链接mongoDB客户端
# 参数1: mongoDB的ip地址
# 参数2: mongoDB的端口号 默认:27017
client = MongoClient('localhost', 27017)
# print(client)
# 2、进入tank_db库,没有则创建
# print(client['tank_db'])
# 3、创建集合
# print(client['tank_db']['people'])
# 4、给tank_db库插入数据
# 1.插入一条
data1 = {
'name': 'tank',
'age': 18,
'sex': 'male'
}
client['tank_db']['people'].insert(data1)
# 2.插入多条
data1 = {
'name': 'tank',
'age': 18,
'sex': 'male'
}
data2 = {
'name': '李子恒',
'age': 84,
'sex': 'female'
}
data3 = {
'name': '张庭宇',
'age': 73,
'sex': 'male'
}
client['tank_db']['people'].insert([data1, data2, data3])
#
# # 5、查数据
# # 查看所有数据
data_s = client['tank_db']['people'].find()
print(data_s) # <pymongo.cursor.Cursor object at 0x000002EEA6720128>
# # 需要循环打印所有数据
for data in data_s:
print(data)
# # 查看一条数据
data = client['tank_db']['people'].find_one()
print(data)
# 官方推荐使用
# 插入一条insert_one
client['tank_db']['people'].insert_one()
# 插入多条insert_many
client['tank_db']['people'].insert_many()
二.作业
把豌豆荚爬取的数据插入mongoDB中
- 创建一个wandoujia库
- 把主页的数据存放一个名为index集合中
- 把详情页的数据存放一个名为detail集合中
'''
主页:
图标地址、下载次数、大小、详情页地址
详情页:
游戏名、好评率、评论数、小编点评、下载地址、简介、网友评论、1-5张截图链接地址、
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=1&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=2&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page=3&ctoken=FRsWKgWBqMBZLdxLaK4iem9B
32
'''
import requests
from bs4 import BeautifulSoup
from pymongo import MongoClient
'''
3、把豌豆荚爬取的数据插入mongoDB中
- 创建一个wandoujia库
- 把主页的数据存放一个名为index集合中
- 把详情页的数据存放一个名为detail集合中
'''
# 连接MongoDB客户端
client = MongoClient('localhost', 27017)
# 创建或选择wandoujia库,index集合
index_col = client['wandoujia']['index']
# 创建或选择wandoujia库,detail集合
detail_col = client['wandoujia']['detail']
# 1、发送请求
def get_page(url):
response = requests.get(url)
return response
# 2、开始解析
# 解析详情页
def parse_detail(text):
soup = BeautifulSoup(text, 'lxml')
# print(soup)
# app名称
try:
name = soup.find(name="span", attrs={"class": "title"}).text
except Exception:
# 若有异常,设置为None
name = None
# print(name)
# 好评率
try:
love = soup.find(name='span', attrs={"class": "love"}).text
except Exception:
love = None
# print(love)
# 评论数
try:
commit_num = soup.find(name='a', attrs={"class": "comment-open"}).text
except Exception:
commit_num = None
# print(commit_num)
# 小编点评
try:
commit_content = soup.find(name='div', attrs={"class": "con"}).text
except Exception:
commit_content = None
# print(commit_content)
# app下载链接
try:
download_url = soup.find(name='a', attrs={"class": "normal-dl-btn"}).attrs['href']
except Exception:
# 若有异常,设置为None
download_url = None
# print(download_url)
# print(
# f'''
# ============= tank ==============
# app名称:{name}
# 好评率: {love}
# 评论数: {commit_num}
# 小编点评: {commit_content}
# app下载链接: {download_url}
# ============= end ==============
# '''
# )
# 判断所有数据都存在,正常赋值
if name and love and commit_num and commit_content and download_url :
detail_data = {
'name': name,
'love': love,
'commit_num': commit_num,
'commit_content': commit_content,
'download_url': download_url
}
# 若love没有值,则设置为 没人点赞,很惨
if not love:
detail_data = {
'name': name,
'love': "没人点赞,很惨",
'commit_num': commit_num,
'commit_content': commit_content,
'download_url': download_url
}
# 若download_url没有值,则设置为 没有安装包
if not download_url:
detail_data = {
'name': name,
'love': love,
'commit_num': commit_num,
'commit_content': commit_content,
'download_url': '没有安装包'
}
# 插入详情页数据
detail_col.insert(detail_data)
print(f'{name}app数据插入成功!')
# 解析主页
def parse_index(data):
soup = BeautifulSoup(data, 'lxml')
# 获取所有app的li标签
app_list = soup.find_all(name='li', attrs={"class": "card"})
for app in app_list:
# print(app)
# print('tank' * 1000)
# print('tank *' * 1000)
# print(app)
# 图标地址
# 获取第一个img标签中的data-original属性
img = app.find(name='img').attrs['data-original']
# print(img)
# 下载次数
# 获取class为install-count的span标签中的文本
down_num = app.find(name='span', attrs={"class": "install-count"}).text
# print(down_num)
import re
# 大小
# 根据文本正则获取到文本中包含 数字 + MB(\d+代表数字)的span标签中的文本
size = soup.find(name='span', text=re.compile("\d+MB")).text
# print(size)
# 详情页地址
# 获取class为detail-check-btn的a标签中的href属性
# detail_url = soup.find(name='a', attrs={"class": "name"}).attrs['href']
# print(detail_url)
# 详情页地址
detail_url = app.find(name='a').attrs['href']
# print(detail_url)
# 拼接数据
index_data = {
'img': img,
'down_num': down_num,
'size': size,
'detail_url': detail_url
}
# 插入数据
index_col.insert(index_data)
print('主页数据插入成功!')
# 3、往app详情页发送请求
response = get_page(detail_url)
# 4、解析app详情页
parse_detail(response.text)
def main():
for line in range(1, 33):
url = f"https://www.wandoujia.com/wdjweb/api/category/more?catId=6001&subCatId=0&page={line}&ctoken=FRsWKgWBqMBZLdxLaK4iem9B"
# 1、往app接口发送请求
response = get_page(url)
# print(response.text)
print('*' * 1000)
# 反序列化为字典
data = response.json()
# 获取接口中app标签数据
app_li = data['data']['content']
# print(app_li)
# 2、解析app标签数据
parse_index(app_li)
# 执行完所有函数关闭mongoDB客户端
client.close()
if __name__ == '__main__':
main()