Python多线程爬虫与多种数据存储方式实现(Python爬虫实战2)

1. 多进程爬虫

　　对于数据量较大的爬虫，对数据的处理要求较高时，可以采用python多进程或多线程的机制完成，多进程是指分配多个CPU处理程序，同一时刻只有一个CPU在工作，多线程是指进程内部有多个类似"子进程"同时在协同工作。python中有多种多个模块可完成多进程和多线程的工作，此处此用multiprocessing模块完成多线程爬虫，测试过程中发现，由于站点具有反爬虫机制，当url地址和进程数目较多时，爬虫会报错。

2. 代码内容

#!/usr/bin/python
#_*_ coding:utf _*_

import re
import time
import requests
from multiprocessing import Pool

duanzi_list = []

def get_web_html(url):
	'''
	@params:获取url地址web站点的html数据
	'''
	headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0"}
	try:
		req = requests.get(url,headers=headers)
		if req.status_code == 200:
			response = req.text.encode('utf8')
	except Exception as e:
		print e
	return response

def scrap_qiushi_info(url):
	'''
	@params:url,获取段子数据信息
	'''
	html = get_web_html(url)
	usernames = re.findall(r'<h2>(.*?)</h2>',html,re.S|re.M)
	levels = re.findall('<div class="articleGender \w*Icon">(\d+)</div>',html,re.S|re.M)
	laugh_counts = re.findall('<span class="stats-vote">.*?<i class="number">(\d+)</i>',html,re.S|re.M)
	comment_counts = re.findall('<i class="number">(\d+)</i> 评论',html,re.S|re.M)
	contents = re.findall('<div class="content">.*?<span>(.*?)</span>',html,re.S|re.M)
	for username,level,laugh_count,comment_count,content in zip(usernames,levels,laugh_counts,comment_counts,contents):
		information = {
			"username": username.strip(),
			"level": level.strip(),
			"laugh_count": laugh_count.strip(),
			"comment_count": comment_count.strip(),
			"content": content.strip()
		}
		duanzi_list.append(information)
	time.sleep(1)
	return duanzi_list

def normal_scapper(url_lists):
	'''
	定义调用函数，使用普通的爬虫函数爬取数据
	'''
	begin_time = time.time()
	for url in url_lists:
		scrap_qiushi_info(url)
	end_time = time.time()
	print "普通爬虫一共耗费时长:%f" % (end_time - begin_time)

def muti_process_scapper(url_lists,process_num=2):
	'''
	定义多进程爬虫调用函数，使用mutiprocessing模块爬取web数据
	'''
	begin_time = time.time()
	pool = Pool(processes=process_num)
	pool.map(scrap_qiushi_info,url_lists)
	end_time = time.time()
	print "%d个进程爬虫爬取所耗费时长为:%s" % (process_num,(end_time - begin_time))

def main():
	'''
	定义main()函数，程序入口，通过列表推倒式获取url地址，调用爬虫函数
	'''
	url_lists = ['https://www.qiushibaike.com/text/page/{}'.format(i) for i in range(1,11)]
	normal_scapper(url_lists)
	muti_process_scapper(url_lists,process_num=2)

if __name__ == "__main__":
	main()

3. 爬取的数据存入到MongoDB数据库

#!/usr/bin/python
#_*_ coding:utf _*_

import re
import time
import json
import requests
import pymongo
from multiprocessing import Pool

duanzi_list = []

def get_web_html(url):
	'''
	@params:获取url地址web站点的html数据
	'''
	headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0"}
	try:
		req = requests.get(url,headers=headers)
		if req.status_code == 200:
			response = req.text.encode('utf8')
	except Exception as e:
		print e
	return response

def scrap_qiushi_info(url):
	'''
	@params:url,获取段子数据信息
	'''
	html = get_web_html(url)
	usernames = re.findall(r'<h2>(.*?)</h2>',html,re.S|re.M)
	levels = re.findall('<div class="articleGender \w*Icon">(\d+)</div>',html,re.S|re.M)
	laugh_counts = re.findall('<span class="stats-vote">.*?<i class="number">(\d+)</i>',html,re.S|re.M)
	comment_counts = re.findall('<i class="number">(\d+)</i> 评论',html,re.S|re.M)
	contents = re.findall('<div class="content">.*?<span>(.*?)</span>',html,re.S|re.M)
	for username,level,laugh_count,comment_count,content in zip(usernames,levels,laugh_counts,comment_counts,contents):
		information = {
			"username": username.strip(),
			"level": level.strip(),
			"laugh_count": laugh_count.strip(),
			"comment_count": comment_count.strip(),
			"content": content.strip()
		}
		duanzi_list.append(information)
	return duanzi_list

def write_into_mongo(datas):
	'''
	@datas: 需要插入到mongoDB的数据，封装为字典,通过遍历的方式将数据插入到mongoDB中，insert_one()表示一次插入一条数据
	'''
	client = pymongo.MongoClient('localhost',27017)
	duanzi = client['duanzi_db']
	duanzi_info = duanzi['duanzi_info']
	for data in datas:
		duanzi_info.insert_one(data)

def query_data_from_mongo():
	'''
	查询mongoDB中的数据
	'''
	client = pymongo.MongoClient('localhost',27017)['duanzi_db']['duanzi_info']
	for data in client.find():
		print data
	print "一共查询到%d条数据" % (client.find().count())

def main():
	'''
	定义main()函数，程序入口，通过列表推倒式获取url地址，调用爬虫函数
	'''
	url_lists = ['https://www.qiushibaike.com/text/page/{}'.format(i) for i in range(1,11)]
	for url in url_lists:
		scrap_qiushi_info(url)
		time.sleep(1)
	write_into_mongo(duanzi_list)

if __name__ == "__main__":
	main()
	#query_data_from_mongo()

4. 插入至MySQL数据库

　　将爬虫获取的数据插入到关系性数据库MySQL数据库中作为永久数据存储，首先需要在MySQL数据库中创建库和表，如下：

1. 创建库
MariaDB [(none)]> create database qiushi;
Query OK, 1 row affected (0.00 sec)

2. 使用库
MariaDB [(none)]> use qiushi;
Database changed

3. 创建表格
MariaDB [qiushi]> create table qiushi_info(id int(32) unsigned primary key auto_increment,username varchar(64) not null,level int default 0,laugh_count int default 0,comment_count int default 0,content text default '')engine=InnoDB charset='UTF8';
Query OK, 0 rows affected, 1 warning (0.06 sec)

MariaDB [qiushi]> show create table qiushi_info;
+-------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Table       | Create Table                                                                                                                                                                                                                                                                                            |
+-------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| qiushi_info | CREATE TABLE `qiushi_info` (
  `id` int(32) unsigned NOT NULL AUTO_INCREMENT,
  `username` varchar(64) NOT NULL,
  `level` int(11) DEFAULT '0',
  `laugh_count` int(11) DEFAULT '0',
  `comment_count` int(11) DEFAULT '0',
  `content` text,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 |
+-------------+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.00 sec)

写入到MySQL数据库中的代码如下：

#!/usr/bin/python
#_*_ coding:utf _*_
#blog:http://www.cnblogs.com/cloudlab/

import re
import time
import pymysql
import requests

duanzi_list = []

def get_web_html(url):
	'''
	@params:获取url地址web站点的html数据
	'''
	headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0"}
	try:
		req = requests.get(url,headers=headers)
		if req.status_code == 200:
			response = req.text.encode('utf8')
	except Exception as e:
		print e
	return response

def scrap_qiushi_info(url):
	'''
	@params:url,获取段子数据信息
	'''
	html = get_web_html(url)
	usernames = re.findall(r'<h2>(.*?)</h2>',html,re.S|re.M)
	levels = re.findall('<div class="articleGender \w*Icon">(\d+)</div>',html,re.S|re.M)
	laugh_counts = re.findall('<span class="stats-vote">.*?<i class="number">(\d+)</i>',html,re.S|re.M)
	comment_counts = re.findall('<i class="number">(\d+)</i> 评论',html,re.S|re.M)
	contents = re.findall('<div class="content">.*?<span>(.*?)</span>',html,re.S|re.M)
	for username,level,laugh_count,comment_count,content in zip(usernames,levels,laugh_counts,comment_counts,contents):
		information = {
			"username": username.strip(),
			"level": level.strip(),
			"laugh_count": laugh_count.strip(),
			"comment_count": comment_count.strip(),
			"content": content.strip()
		}
		duanzi_list.append(information)
	return duanzi_list

def write_into_mysql(datas):
	'''
	@params: datas，将爬虫获取的数据写入到MySQL数据库中
	'''
	try:
		conn = pymysql.connect(host='localhost',port=3306,user='root',password='',db='qiushi',charset='utf8')
		cursor = conn.cursor(pymysql.cursors.DictCursor)
		for data in datas:
			data_list = (data['username'],int(data['level']),int(data['laugh_count']),int(data['comment_count']),data['content'])
			sql = "INSERT INTO qiushi_info(username,level,laugh_count,comment_count,content) VALUES('%s',%s,%s,%s,'%s')" %(data_list)
			cursor.execute(sql)
			conn.commit()
	except Exception as e:
		print e
	cursor.close()
	conn.close()

def main():
	'''
	定义main()函数，程序入口，通过列表推倒式获取url地址，调用爬虫函数
	'''
	url_lists = ['https://www.qiushibaike.com/text/page/{}'.format(i) for i in range(1,11)]
	for url in url_lists:
		scrap_qiushi_info(url)
		time.sleep(1)
	write_into_mysql(duanzi_list)

if __name__ == "__main__":
	main()

5. 将爬虫数据写入到CSV文件

　　CSV文件是以逗号,形式分割的文本读写方式，能够通过纯文本或者Excel方式读取，是一种常见的数据存储方式，此处将爬取的数据存入到CSV文件内。

将数据存入到CSV文件代码内容如下：

#!/usr/bin/python
#_*_ coding:utf _*_
#blog:http://www.cnblogs.com/cloudlab/

import re
import csv
import time
import requests

duanzi_list = []

def get_web_html(url):
	'''
	@params:获取url地址web站点的html数据
	'''
	headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0"}
	try:
		req = requests.get(url,headers=headers)
		if req.status_code == 200:
			response = req.text.encode('utf8')
	except Exception as e:
		print e
	return response

def scrap_qiushi_info(url):
	'''
	@params:url,获取段子数据信息
	'''
	html = get_web_html(url)
	usernames = re.findall(r'<h2>(.*?)</h2>',html,re.S|re.M)
	levels = re.findall('<div class="articleGender \w*Icon">(\d+)</div>',html,re.S|re.M)
	laugh_counts = re.findall('<span class="stats-vote">.*?<i class="number">(\d+)</i>',html,re.S|re.M)
	comment_counts = re.findall('<i class="number">(\d+)</i> 评论',html,re.S|re.M)
	contents = re.findall('<div class="content">.*?<span>(.*?)</span>',html,re.S|re.M)
	for username,level,laugh_count,comment_count,content in zip(usernames,levels,laugh_counts,comment_counts,contents):
		information = {
			"username": username.strip(),
			"level": level.strip(),
			"laugh_count": laugh_count.strip(),
			"comment_count": comment_count.strip(),
			"content": content.strip()
		}
		duanzi_list.append(information)
	return duanzi_list

def write_into_csv(datas,filename):
	'''
	@datas: 需要写入csv文件的数据内容，是一个列表
	@params:filename,需要写入到目标文件的csv文件名
	'''
	with file(filename,'w+') as f:
		writer = csv.writer(f)
		writer.writerow(('username','level','laugh_count','comment_count','content'))
		for data in datas:
			writer.writerow((data['username'],data['level'],data['laugh_count'],data['comment_count'],data['content']))

def main():
	'''
	定义main()函数，程序入口，通过列表推倒式获取url地址，调用爬虫函数
	'''
	url_lists = ['https://www.qiushibaike.com/text/page/{}'.format(i) for i in range(1,11)]
	for url in url_lists:
		scrap_qiushi_info(url)
		time.sleep(1)
	write_into_csv(duanzi_list,'/root/duanzi_info.csv')

if __name__ == "__main__":
	main()

6. 将爬取数据写入到文本文件中

#!/usr/bin/python
#_*_ coding:utf _*_
#blog:http://www.cnblogs.com/cloudlab/

import re
import csv
import time
import requests

duanzi_list = []

def get_web_html(url):
	'''
	@params:获取url地址web站点的html数据
	'''
	headers = {"User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0"}
	try:
		req = requests.get(url,headers=headers)
		if req.status_code == 200:
			response = req.text.encode('utf8')
	except Exception as e:
		print e
	return response

def scrap_qiushi_info(url):
	'''
	@params:url,获取段子数据信息
	'''
	html = get_web_html(url)
	usernames = re.findall(r'<h2>(.*?)</h2>',html,re.S|re.M)
	levels = re.findall('<div class="articleGender \w*Icon">(\d+)</div>',html,re.S|re.M)
	laugh_counts = re.findall('<span class="stats-vote">.*?<i class="number">(\d+)</i>',html,re.S|re.M)
	comment_counts = re.findall('<i class="number">(\d+)</i> 评论',html,re.S|re.M)
	contents = re.findall('<div class="content">.*?<span>(.*?)</span>',html,re.S|re.M)
	for username,level,laugh_count,comment_count,content in zip(usernames,levels,laugh_counts,comment_counts,contents):
		information = {
			"username": username.strip(),
			"level": level.strip(),
			"laugh_count": laugh_count.strip(),
			"comment_count": comment_count.strip(),
			"content": content.strip()
		}
		duanzi_list.append(information)
	return duanzi_list

def write_into_files(datas,filename):
	'''
	定义数据存入写文件的函数
	@params:datas需要写入的数据
	@filename:将数据写入到指定的文件名
	'''
	print "开始写入文件.."
	with file(filename,'w+') as f:
		f.write("用户名" + "\t" + "用户等级" + "\t" + "笑话数" + "\t" + "评论数" + "\t" + "段子内容" + "\n")
		for data in datas:
			f.write(data['username'] + "\t" + \
				data['level'] + "\t" + \
				data['laugh_count'] + "\t" + \
				data['comment_count'] + "\t" + \
				data['content'] + "\n" + "\n"
			)

def main():
	'''
	定义main()函数，程序入口，通过列表推倒式获取url地址，调用爬虫函数
	'''
	url_lists = ['https://www.qiushibaike.com/text/page/{}'.format(i) for i in range(1,11)]
	for url in url_lists:
		scrap_qiushi_info(url)
		time.sleep(1)
	write_into_files(duanzi_list,'/root/duanzi.txt')

if __name__ == "__main__":
	main()