多任务的爬虫

1.多线程的方法使用

在python3中，主线程主进程结束，子线程，子进程不会结束

为了能够让主线程回收子线程，可以把子线程设置为守护线程,即该线程不重要，主线程结束，子线程结束.

t1 = threading.Thread(targe=func,args=(,))
t1.setDaemon(True)
t1.start() #此时线程才会启动

2.队列模块的使用

from queue import Queue
q = Queue(maxsize=100)
item = {}
q.put_nowait(item) #不等待直接放，队列满的时候会报错
q.put(item) #放入数据，队列满的时候回等待
q.get_nowait() #不等待直接取，队列空的时候会报错
q.get() #取出数据，队列为空的时候会等待
q.qsize() #获取队列中现存数据的个数 
q.join() #队列中维持了一个计数，计数不为0时候让主线程阻塞等待，队列计数为0的时候才会继续往后执行
q.task_done() 
# put的时候计数+1，get不会-1，get需要和task_done 一起使用才会-1

3.多线程实现思路剖析

1. 把爬虫中的每个步骤封装成函数，分别用线程去执行
2. 不同的函数通过队列相互通信，函数间解耦 

代码如下:

# coding=utf-8
import requests
from lxml import etree
from queue import Queue
import threading
import time

class QiuBai:
    def __init__(self):
        self.temp_url = "http://www.qiushibaike.com/8hr/page/{}"
        self.headers = {"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36"}
        self.url_queue = Queue()
        self.html_queue = Queue()
        self.content_list_queue = Queue()

    def get_url_list(self):
        # return [self.temp_url.format(i) for i in range(1,14)]
        for i in range(1,14):
            self.url_queue.put(self.temp_url.format(i))

    def parse_url(self):
        while True:
            url = self.url_queue.get()
            response = requests.get(url,headers=self.headers)
            print(response)

            if response.status_code != 200:
                self.url_queue.put(url)
            else:
                self.html_queue.put(response.content.decode())
            self.url_queue.task_done()  #让队列的计数-1

    def get_content_list(self):#提取数据
        while True:
            html_str = self.html_queue.get()
            html = etree.HTML(html_str)
            div_list = html.xpath("//div[@id='content-left']/div")
            content_list = []
            for div in div_list:
                item = {}
                item["user_name"] = div.xpath(".//h2/text()")[0].strip()
                item["content"] = [i.strip() for i in div.xpath(".//div[@class='content']/span/text()")]
                content_list.append(item)
            self.content_list_queue.put(content_list)
            self.html_queue.task_done()

    def save_content_list(self): #保存
        while True:
            content_list = self.content_list_queue.get()
            for content in content_list:
                # print(content)
                pass
            self.content_list_queue.task_done()

    def run(self):#实现做主要逻辑
        thread_list = []
        #1. 准备url列表
        t_url = threading.Thread(target=self.get_url_list)
        thread_list.append(t_url)
        #2. 遍历发送请求，获取响应
        for i in range(3):
            t_parse = threading.Thread(target=self.parse_url)
            thread_list.append(t_parse)
        #3. 提取数据
        t_content = threading.Thread(target=self.get_content_list)
        thread_list.append(t_content)
            #4. 保存
        t_save = threading.Thread(target=self.save_content_list)
        thread_list.append(t_save)

        for t in thread_list:
            t.setDaemon(True) #把子线程设置为守护线程
            t.start()

        for q in [self.url_queue,self.html_queue,self.content_list_queue]:
            q.join() #让主线程阻塞，等待队列计数为0


if __name__ == '__main__':
    t1 = time.time()
    qiubai = QiuBai()
    qiubai.run()
    print("total cost:",time.time()-t1)

　　

多进程程的方法使用

from multiprocessing import Process
t1 = Process(targe=func,args=(,))
t1.daemon = True  #设置为守护进程
t1.start() #此时线程才会启动

多进程中队列的使用

多进程中使用普通的队列模块会发生阻塞，对应的需要使用multiprocessing提供的JoinableQueue模块，其使用过程和在线程中使用的queue方法相同.

代码如下:

# coding=utf-8
import requests
from lxml import etree
# from queue import Queue
# import threading
from multiprocessing import Process
from multiprocessing import JoinableQueue as Queue
import time

class QiuBai:
    def __init__(self):
        self.temp_url = "http://www.qiushibaike.com/8hr/page/{}"
        self.headers = {"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36"}
        self.url_queue = Queue()
        self.html_queue = Queue()
        self.content_list_queue = Queue()
        self.proxies = {"http":"http://58.247.179.94:8060"}

    def get_url_list(self):
        # return [self.temp_url.format(i) for i in range(1,14)]
        for i in range(1,14):
            self.url_queue.put(self.temp_url.format(i))

    def parse_url(self):
        while True:
            url = self.url_queue.get()
            response = requests.get(url,headers=self.headers,proxies=self.proxies)
            print(response)

            if response.status_code != 200:
                self.url_queue.put(url)
            else:
                self.html_queue.put(response.content.decode())
            self.url_queue.task_done()  #让队列的计数-1

    def get_content_list(self):#提取数据
        while True:
            html_str = self.html_queue.get()
            html = etree.HTML(html_str)
            div_list = html.xpath("//div[@id='content-left']/div")
            content_list = []
            for div in div_list:
                item = {}
                item["user_name"] = div.xpath(".//h2/text()")[0].strip()
                item["content"] = [i.strip() for i in div.xpath(".//div[@class='content']/span/text()")]
                content_list.append(item)
            self.content_list_queue.put(content_list)
            self.html_queue.task_done()

    def save_content_list(self): #保存
        while True:
            content_list = self.content_list_queue.get()
            for content in content_list:
                # print(content)
                pass
            self.content_list_queue.task_done()

    def run(self):#实现做主要逻辑
        thread_list = []
        #1. 准备url列表
        t_url = Process(target=self.get_url_list)
        thread_list.append(t_url)
        #2. 遍历发送请求，获取响应
        for i in range(13):
            t_parse = Process(target=self.parse_url)
            thread_list.append(t_parse)
        #3. 提取数据
        t_content = Process(target=self.get_content_list)
        thread_list.append(t_content)
        #4. 保存
        t_save = Process(target=self.save_content_list)
        thread_list.append(t_save)

        for process in thread_list:
            process.daemon = True #把子线程设置为守护线程
            process.start()

        for q in [self.url_queue,self.html_queue,self.content_list_queue]:
            q.join() #让主线程阻塞，等待队列计数为0


if __name__ == '__main__':
    t1 = time.time()
    qiubai = QiuBai()
    qiubai.run()
    print("total cost:",time.time()-t1)

　　

通过线程池实现更快的爬虫

1. 线程池使用方法介绍

1.实例化线程池对象

 from multiprocessing.dummy import Pool
 pool = Pool(process=5) #默认大小是cup的个数

2. 把从发送请求，提取数据，到保存合并成一个函数，交给线程池异步执行

使用方法pool.apply_async(func)

 def exetute_requests_item_save(self):
     url = self.queue.get()
     html_str = self.parse_url(url)
     content_list = self.get_content_list(html_str)
     self.save_content_list(content_list)
     self.total_response_num +=1

 pool.apply_async(self.exetute_requests_item_save)

3.添加回调函数

通过apply_async的方法能够让函数异步执行，但是只能够执行一次

为了让其能够被反复执行，通过添加回调函数的方式能够让_callback 递归的调用自己

同时需要指定递归退出的条件.

def _callback(self,temp):
     if self.is_running:
          pool.apply_async(self.exetute_requests_item_save,callback=self._callback)

 pool.apply_async(self.exetute_requests_item_save,callback=self._callback)

4.确定程序结束的条件 程序在获取的响应和url数量相同的时候可以结束

while True: #防止主线程结束
     time.sleep(0.0001)  #避免cpu空转，浪费资源
     if self.total_response_num>=self.total_requests_num:
         self.is_running= False
         break
 self.pool.close() #关闭线程池，防止新的线程开启
# self.pool.join() #等待所有的子线程结束

2. 使用线程池实现爬虫的具体实现

# coding=utf-8
import requests
from lxml import etree
import time
from queue import Queue
from multiprocessing.dummy import Pool

class QiuBai:
    def __init__(self):
        self.temp_url = "http://www.qiushibaike.com/8hr/page/{}"
        self.headers = {"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36"}
        self.queue = Queue()
        self.pool= Pool(5)
        self.is_running = True
        self.total_request_num = 0
        self.total_response_num =0
        self.proxies = {"http":"http://58.247.179.94:8060"}


    def get_url_list(self):
        for i in range(1,14):
            self.queue.put(self.temp_url.format(i))
            self.total_request_num += 1

    def parse_url(self,url):
        # response = requests.get(url,headers=self.headers,proxies=self.proxies)
        response = requests.get(url,headers=self.headers)
        print(response)
        return response.content.decode()

    def get_content_list(self,html_str):#提取数据
        html = etree.HTML(html_str)
        div_list = html.xpath("//div[@id='content-left']/div")
        content_list = []
        for div in div_list:
            item = {}
            item["user_name"] = div.xpath(".//h2/text()")[0].strip()
            item["content"] = [i.strip() for i in div.xpath(".//div[@class='content']/span/text()")]
            content_list.append(item)
        return content_list

    def save_content_list(self,content_list): #保存
        for content in content_list:
            # print(content)
            pass

    def _execete_request_content_save(self):  #进行一次url地址的请求，提取，保存
        url = self.queue.get()
        html_str = self.parse_url(url)
        #3. 提取数据
        content_list = self.get_content_list(html_str)
        #4. 保存
        self.save_content_list(content_list)
        self.total_response_num +=1

    def _callback(self,temp):
        if self.is_running:
            self.pool.apply_async(self._execete_request_content_save,callback=self._callback)


    def run(self):#实现做主要逻辑
        #1. 准备url列表
        self.get_url_list()
        for i in range(3): #设置并发数为3
            self.pool.apply_async(self._execete_request_content_save,callback=self._callback)

        while True:
            time.sleep(0.0001)
            if self.total_response_num>= self.total_request_num:
                self.is_running = False
                break


if __name__ == '__main__':
    t1 = time.time()
    qiubai = QiuBai()
    qiubai.run()
    print("total cost:",time.time()-t1)

3. 使用协程池实现爬虫的具体实现

# coding=utf-8
import gevent.monkey
gevent.monkey.patch_all()

from gevent.pool import  Pool
import requests
from lxml import etree
import time
from queue import Queue
# from multiprocessing.dummy import Pool

class QiuBai:
    def __init__(self):
        self.temp_url = "http://www.qiushibaike.com/8hr/page/{}"
        self.headers = {"User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_4) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.139 Safari/537.36"}
        self.queue = Queue()
        self.pool= Pool(5)
        self.is_running = True
        self.total_request_num = 0
        self.total_response_num =0
        self.proxies = {"http":"http://58.247.179.94:8060"}


    def get_url_list(self):
        for i in range(1,14):
            self.queue.put(self.temp_url.format(i))
            self.total_request_num += 1

    def parse_url(self,url):
        # response = requests.get(url,headers=self.headers,proxies=self.proxies)
        response = requests.get(url,headers=self.headers)
        print(response)
        return response.content.decode()

    def get_content_list(self,html_str):#提取数据
        html = etree.HTML(html_str)
        div_list = html.xpath("//div[@id='content-left']/div")
        content_list = []
        for div in div_list:
            item = {}
            item["user_name"] = div.xpath(".//h2/text()")[0].strip()
            item["content"] = [i.strip() for i in div.xpath(".//div[@class='content']/span/text()")]
            content_list.append(item)
        return content_list

    def save_content_list(self,content_list): #保存
        for content in content_list:
            # print(content)
            pass

    def _execete_request_content_save(self):  #进行一次url地址的请求，提取，保存
        url = self.queue.get()
        html_str = self.parse_url(url)
        #3. 提取数据
        content_list = self.get_content_list(html_str)
        #4. 保存
        self.save_content_list(content_list)
        self.total_response_num +=1

    def _callback(self,temp):
        if self.is_running:
            self.pool.apply_async(self._execete_request_content_save,callback=self._callback)


    def run(self):#实现做主要逻辑
        #1. 准备url列表
        self.get_url_list()
        for i in range(3): #设置并发数为3
            self.pool.apply_async(self._execete_request_content_save,callback=self._callback)

        while True:
            time.sleep(0.0001)
            if self.total_response_num>= self.total_request_num:
                self.is_running = False
                break


if __name__ == '__main__':
    t1 = time.time()
    qiubai = QiuBai()
    qiubai.run()
    print("total cost:",time.time()-t1)