字符串常用方法
# 去掉左右空格
'hello world'.strip() # 'hello world'
# 按指定字符切割
'hello world'.split(' ') # ['hello','world']
# 替换指定字符串
'hello world'.replace(' ','#') # 'hello#world'
csv模块
作用:将爬取的数据存放到本地的csv文件中
使用流程
- 导入模块
- 打开csv文件
- 初始化写入对象
- 写入数据(参数为列表)
import csv with open('test.csv','w') as f: writer = csv.writer(f) # 初始化写入对象
# 写一行
writer.writerow(['超哥哥',20])
writer.writerow(['步惊云',22]) with open('test.csv','a') as f:
writer = csv.writer(f)
# 写多行
data_list = [('聂风',23),('秦霜',30)]
writer.writerows(data_list)
Windows中使用csv模块默认会在每行后面添加一个空行,使用newline=''可解决
with open('xxx.csv','w',newline='') as f:
猫眼电影top100抓取案例
确定URL网址
猫眼电影 - 榜单 - top100榜 目标
电影名称、主演、上映时间 操作步骤
1、查看是否为动态加载
右键 - 查看网页源代码 - 搜索爬取关键字(查看在源代码中是否存在)
2、找URL规律
第1页:https://maoyan.com/board/4?offset=0
第2页:https://maoyan.com/board/4?offset=10
第n页:offset=(n-1)*10
3、正则表达式
<div class="movie-item-info">.*?title="(.*?)".*?class="star">(.*?)</p>.*?releasetime">(.*?)</p>
4、编写程序框架,完善程序
- 打印程序执行时间
- 随机的User-Agent,(确保每次发请求使用随机)
- 数据爬下来后做处理(字符串),定义成字典
- 一条龙: 获取 -> 调用解析 -> 数据处理
- 猫眼电影数据存入本地 maoyanfilm.csv 文件
from urllib import request
import time
import re
import csv
import random class MaoyanSpider(object):
def __init__(self):
self.page = 1 # 用于记录页数
self.url = 'https://maoyan.com/board/4?offset={}'
self.agent = [
'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 \
Safari/535.1',
'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20100101 Firefox/6.0',
'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; .NET CLR 2.0.50727; \
.NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; InfoPath.3)',
'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1'] # 请求
def get_page(self, url):
headers = {'User-Agent': random.choice(self.agent)} # 每次使用随机的user-agent
req = request.Request(url=url, headers=headers) # 创建请求对象
res = request.urlopen(req) # 发起请求
html = res.read().decode('utf-8') # 获取请求内容
self.parse_page(html) # 直接调用解析函数 # 解析
def parse_page(self, html):
pattren = re.compile(
'<div class="movie-item-info">.*?title="(.*?)".*?class="star">(.*?)</p>.*?releasetime">(.*?)</p>', re.S)
r_list = pattren.findall(html)
# rlist: [('霸王别姬', '\n 主演:张国荣,张丰毅,巩俐\n ', '上映时间:1993-01-01'),(...),(...)]
self.write_page(r_list) # 写入csv文件 # # 保存,打印输出
# def write_page(self,r_list):
# one_film_dict = {}
# for rt in r_list:
# one_film_dict['name'] = rt[0].strip()
# one_film_dict['star'] = rt[1].strip()
# one_film_dict['time'] = rt[2].strip()[5:15]
#
# print(one_film_dict) # 保存到csv文件(writerows) -- 推荐使用此方法
def write_page(self, r_list):
# 空列表,最终writerows()的参数: [(),(),()]
film_list = []
with open('maoyan.csv', 'a',newline="") as f:
writer = csv.writer(f)
for rt in r_list:
# 把处理过的数据定义成元组
t = (rt[0], rt[1].strip(), rt[2].strip()[5:15])
film_list.append(t) writer.writerows(film_list) def main(self):
for offset in range(0, 31, 10):
url = self.url.format(offset)
self.get_page(url)
time.sleep(random.randint(1, 3))
print('第%d页爬取完成' % self.page)
self.page += 1 if __name__ == '__main__':
start = time.time()
spider = MaoyanSpider()
spider.main()
end = time.time()
print('执行时间: %.2f' % (end - start))
数据持久化存储(MySQL数据库)
让我们来回顾一下pymysql模块的基本使用
import pymysql db = pymysql.connect('localhost', 'root', '', 'maoyandb', charset='utf8')
cursor = db.cursor() # 创建游标对象
# execute()方法第二个参数为列表传参补位
cursor.execute('insert into film values(%s,%s,%s)', ['霸王别姬', '张国荣', ''])
db.commit() # 提交到数据库执行
cursor.close() # 关闭
db.close()
让我们来回顾一下pymysql中executemany()的用法
import pymysql # 数据库连接对象
db = pymysql.connect('localhost', 'root', '', charset='utf8')
cursor = db.cursor() # 游标对象
ins_list = [] # 存放所有数据的大列表
for i in range(2):
name = input('请输入第%d个学生姓名:' % (i + 1))
age = input('请输入第%d个学生年龄:' % (i + 1))
ins_list.append([name, age]) ins = 'insert into t3 values(%s,%s)' # 定义插入语句
cursor.executemany(ins, ins_list) # 一次数据库的IO操作可插入多条语句,提升性能 db.commit() # 提交到数据库执行
cursor.close() # 关闭游标
db.close() # 关闭数据库 ins = 'insert into maoyanfilm values(%s,%s,%s)'
cursor.execute(['霸王', '国荣', ''])
cursor.executemany([
['月光宝盒', '周星驰', ''],
['大圣娶亲', '周星驰', '']])
练习:把猫眼电影案例中电影信息存入MySQL数据库中(尽量使用executemany方法)
from urllib import request
import time
import re
import pymysql
import random class MaoyanSpider(object):
def __init__(self):
self.page = 1 # 用于记录页数
self.url = 'https://maoyan.com/board/4?offset={}'
self.ua_list = [
'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) \
Chrome/14.0.835.163 Safari/535.1',
'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20100101 Firefox/6.0',
'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; \
.NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; InfoPath.3)']
# 创建数据库连接对象和游标对象
self.db = pymysql.connect('localhost', 'root', '', 'maoyandb', charset='utf8')
self.cursor = self.db.cursor() # 获取
def get_page(self, url):
# 每次使用随机的user-agent
headers = {'User-Agent': random.choice(self.ua_list)}
req = request.Request(url=url, headers=headers)
res = request.urlopen(req)
html = res.read().decode('utf-8')
self.parse_page(html) # 直接调用解析函数 # 解析
def parse_page(self, html):
pattren = re.compile(
'<div class="movie-item-info">.*?title="(.*?)".*?class="star">(.*?)</p>.*?releasetime">(.*?)</p>', re.S)
# rlist: [('霸王别姬','张国荣','1993'),(),()]
r_list = pattren.findall(html)
print(r_list)
self.write_page(r_list) # 存入mysql数据库(executemany([ [],[],[] ]))
def write_page(self, r_list):
film_list = []
ins = 'insert into filmtab values(%s,%s,%s)' # 定义插入语句
# 处理数据,放到大列表film_list中
for rt in r_list:
one_film = [rt[0], rt[1].strip(), rt[2].strip()[5:15]]
# 添加到大列表中
film_list.append(one_film)
# 一次数据库IO把1页数据存入
self.cursor.executemany(ins, film_list)
# 提交到数据库执行
self.db.commit() def main(self):
for offset in range(0, 31, 10):
url = self.url.format(offset)
self.get_page(url)
time.sleep(random.randint(1, 3))
print('第%d页爬取完成' % self.page)
self.page += 1 # 断开数据库(所有页爬完之后)
self.cursor.close()
self.db.close() if __name__ == '__main__':
start = time.time()
spider = MaoyanSpider()
spider.main()
end = time.time()
print('执行时间: %.2f' % (end - start))
让我们来做个SQL命令查询
1、查询20年以前的电影的名字和上映时间
select name,time from filmtab where time<(now()-interval 20 year);
2、查询1990-2000年的电影名字和上映时间
select name,time from filmtab where time>='1990-01-01' and time<='2000-12-31';
让我们来复习一下mongdb数据库
import pymongo # 1.连接对象
conn = pymongo.MongoClient(host='127.0.0.1', port=27017)
db = conn['maoyandb'] # 2.库对象
myset = db['filmtab'] # 3.集合对象
myset.insert_one({'name': '赵敏'}) # 4.插入数据库
练习:把猫眼电影案例中电影信息存入MongDB数据库中
from urllib import request
import re
import time
import random
import pymongo class MaoyanSpider(object):
def __init__(self):
self.url = 'https://maoyan.com/board/4?offset={}'
# 计数
self.num = 0
# 创建3个对象
self.conn = pymongo.MongoClient('localhost', 27017)
self.db = self.conn['maoyandb']
self.myset = self.db['filmset']
self.ua_list = [
'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1',
'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0) Gecko/20100101 Firefox/6.0',
'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET \
CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; InfoPath.3)', ] def get_html(self, url):
headers = {
'User-Agent': random.choice(self.ua_list)
}
req = request.Request(url=url, headers=headers)
res = request.urlopen(req)
html = res.read().decode('utf-8')
# 直接调用解析函数
self.parse_html(html) def parse_html(self, html):
re_bds = r'<div class="movie-item-info">.*?title="(.*?)".*?class="star">(.*?)</p>.*?releasetime">(.*?)</p>'
pattern = re.compile(re_bds, re.S)
# film_list: [('霸王别姬','张国荣','1993'),()]
film_list = pattern.findall(html)
# 直接调用写入函数
self.write_html(film_list) # mongodb数据库
def write_html(self, film_list):
for film in film_list:
film_dict = {
'name': film[0].strip(),
'star': film[1].strip(),
'time': film[2].strip()[5:15]
}
# 插入mongodb数据库
self.myset.insert_one(film_dict) def main(self):
for offset in range(0, 31, 10):
url = self.url.format(offset)
self.get_html(url)
time.sleep(random.randint(1, 2)) if __name__ == '__main__':
start = time.time()
spider = MaoyanSpider()
spider.main()
end = time.time()
print('执行时间:%.2f' % (end - start))
电影天堂案例(二级页面抓取)
1、查看是否为静态页面,是否为动态加载
右键 - 查看网页源代码
2、确定URL地址
百度搜索 :电影天堂 - 2019年新片 - 更多
3、目标
*********一级页面***********
1、电影名称
2、电影链接
*********二级页面***********
1、下载链接
4、步骤
找URL规律
第1页 :https://www.dytt8.net/html/gndy/dyzz/list_23_1.html
第2页 :https://www.dytt8.net/html/gndy/dyzz/list_23_2.html
第n页 :https://www.dytt8.net/html/gndy/dyzz/list_23_n.html
写正则表达式
1、一级页面正则表达式(电影名称、电影详情链接)
<table width="100%".*?<td height="26">.*?<a href="(.*?)".*?>(.*?)</a>
2、二级页面正则表达式
<td style="WORD-WRAP.*?>.*?>(.*?)</a>
代码实现
# decode('gbk','ignore') 注意ignore参数
# 注意结构和代码可读性(一个函数不要太冗余)
from urllib import request
import re
import time
import random
from useragents import *
import pymysql class FilmSky(object):
def __init__(self):
self.url = 'https://www.dytt8.net/html/gndy/dyzz/list_23_{}.html'
# 定义两个对象
self.db = pymysql.connect('127.0.0.1', 'root', '', 'maoyandb', charset='utf8')
self.cursor = self.db.cursor() # 获取html函数(因为两个页面都需要发请求)
def get_page(self, url):
req = request.Request(url=url, headers={'User-Agent': random.choice(ua_list)})
res = request.urlopen(req)
# ignore参数,实在处理不了的编码错误忽略
# 查看网页源码,发现网页编码为 gb2312,不是 utf-8
html = res.read().decode('gbk', 'ignore')
return html # 解析提取数据(把名称和下载链接一次性拿到)
# html为一级页面响应内容
def parse_page(self, html):
# 1. 先解析一级页面(电影名称 和 详情链接)
pattern = re.compile('<table width="100%".*?<td height="26">.*?<a href="(.*?)".*?>(.*?)</a>', re.S)
# film_list: [('详情链接','名称'),()]
film_list = pattern.findall(html)
# [('/html/gndy/dyzz/20190806/58956.html', '019年惊悚动作《报仇雪恨/血债血偿》BD中英双字幕'),(),()]
ins = 'insert into filmsky values(%s,%s)'
for film in film_list:
film_name = film[1]
film_link = 'https://www.dytt8.net' + film[0]
# 2. 拿到详情链接后,再去获取详情链接html,提取下载链接
download_link = self.parse_two_html(film_link) self.cursor.execute(ins, [film_name, film_link])
self.db.commit() # 打印测试
d = {'电影名称': film_name, '下载链接': download_link}
print(d) # {'电影名称': '019年惊悚动作《报仇雪恨/血债血偿》BD中英双字幕', '下载链接': 'ftp://ygdy8:[email protected]:8590/阳光电影www.ygdy8.com.报仇雪恨.BD.720p.中英双字幕.mkv'} # 解析二级页面,获取下载链接
def parse_two_html(self, film_link):
two_html = self.get_page(film_link)
pattern = re.compile('<td style="WORD-WRAP.*?>.*?>(.*?)</a>', re.S)
download_link = pattern.findall(two_html)[0]
return download_link # 主函数
def main(self):
for page in range(1, 11):
url = self.url.format(page)
html = self.get_page(url)
self.parse_page(html)
time.sleep(random.randint(1, 3))
print('第%d页完成' % page) if __name__ == '__main__':
start = time.time()
spider = FilmSky()
spider.main()
end = time.time()
print('执行时间:%.2f' % (end - start))