登录
首页精彩阅读Python多线程、异步+多进程爬虫实现代码
Python多线程、异步+多进程爬虫实现代码
2017-09-01
收藏

Python多线程、异步+多进程爬虫实现代码

这篇文章主要介绍了Python多线程、异步+多进程爬虫实现代码,需要的朋友可以参考下

安装Tornado
省事点可以直接用grequests库,下面用的是tornado的异步client。 异步用到了tornado,根据官方文档的例子修改得到一个简单的异步爬虫类。可以参考下最新的文档学习下。

pip install tornado

异步爬虫

#!/usr/bin/env python
# -*- coding:utf-8 -*-

import time
from datetime import timedelta
from tornado import httpclient, gen, ioloop, queues
import traceback


class AsySpider(object):
  """A simple class of asynchronous spider."""
  def __init__(self, urls, concurrency=10, **kwargs):
    urls.reverse()
    self.urls = urls
    self.concurrency = concurrency
    self._q = queues.Queue()
    self._fetching = set()
    self._fetched = set()

  def fetch(self, url, **kwargs):
    fetch = getattr(httpclient.AsyncHTTPClient(), 'fetch')
    return fetch(url, **kwargs)

  def handle_html(self, url, html):
    """handle html page"""
    print(url)

  def handle_response(self, url, response):
    """inherit and rewrite this method"""
    if response.code == 200:
      self.handle_html(url, response.body)

    elif response.code == 599:  # retry
      self._fetching.remove(url)
      self._q.put(url)

  @gen.coroutine
  def get_page(self, url):
    try:
      response = yield self.fetch(url)
      print('######fetched %s' % url)
    except Exception as e:
      print('Exception: %s %s' % (e, url))
      raise gen.Return(e)
    raise gen.Return(response)

  @gen.coroutine
  def _run(self):
    @gen.coroutine
    def fetch_url():
      current_url = yield self._q.get()
      try:
        if current_url in self._fetching:
          return

        print('fetching****** %s' % current_url)
        self._fetching.add(current_url)

        response = yield self.get_page(current_url)
        self.handle_response(current_url, response)  # handle reponse

        self._fetched.add(current_url)

        for i in range(self.concurrency):
          if self.urls:
            yield self._q.put(self.urls.pop())

      finally:
        self._q.task_done()

    @gen.coroutine
    def worker():
      while True:
        yield fetch_url()

    self._q.put(self.urls.pop())  # add first url

    # Start workers, then wait for the work queue to be empty.
    for _ in range(self.concurrency):
      worker()

    yield self._q.join(timeout=timedelta(seconds=300000))
    assert self._fetching == self._fetched

  def run(self):
    io_loop = ioloop.IOLoop.current()
    io_loop.run_sync(self._run)


class MySpider(AsySpider):

  def fetch(self, url, **kwargs):
    """重写父类fetch方法可以添加cookies,headers,timeout等信息"""
    cookies_str = "PHPSESSID=j1tt66a829idnms56ppb70jri4; pspt=%7B%22id%22%3A%2233153%22%2C%22pswd%22%3A%228835d2c1351d221b4ab016fbf9e8253f%22%2C%22_code%22%3A%22f779dcd011f4e2581c716d1e1b945861%22%7D; key=%E9%87%8D%E5%BA%86%E5%95%84%E6%9C%A8%E9%B8%9F%E7%BD%91%E7%BB%9C%E7%A7%91%E6%8A%80%E6%9C%89%E9%99%90%E5%85%AC%E5%8F%B8; think_language=zh-cn; SERVERID=a66d7d08fa1c8b2e37dbdc6ffff82d9e|1444973193|1444967835; CNZZDATA1254842228=1433864393-1442810831-%7C1444972138"  # 从浏览器拷贝cookie字符串
    headers = {
      'User-Agent': 'mozilla/5.0 (compatible; baiduspider/2.0; +http://www.baidu.com/search/spider.html)',
      'cookie': cookies_str
    }
    return super(MySpider, self).fetch(  # 参数参考tornado文档
      url, headers=headers, request_timeout=1
    )

  def handle_html(self, url, html):
    print(url, html)


def main():
  urls = []
  for page in range(1, 100):
    urls.append('http://www.baidu.com?page=%s' % page)
  s = MySpider(urls)
  s.run()


if __name__ == '__main__':
  main()

可以继承这个类,塞一些url进去,然后重写handle_page处理得到的页面。

异步+多进程爬虫
还可以再变态点,加个进程池,使用了multiprocessing模块。效率飕飕的,

#!/usr/bin/env python
# -*- coding:utf-8 -*-

import time
from multiprocessing import Pool
from datetime import timedelta
from tornado import httpclient, gen, ioloop, queues


class AsySpider(object):
  """A simple class of asynchronous spider."""
  def __init__(self, urls, concurrency):
    urls.reverse()
    self.urls = urls
    self.concurrency = concurrency
    self._q = queues.Queue()
    self._fetching = set()
    self._fetched = set()

  def handle_page(self, url, html):
    filename = url.rsplit('/', 1)[1]
    with open(filename, 'w+') as f:
      f.write(html)

  @gen.coroutine
  def get_page(self, url):
    try:
      response = yield httpclient.AsyncHTTPClient().fetch(url)
      print('######fetched %s' % url)
    except Exception as e:
      print('Exception: %s %s' % (e, url))
      raise gen.Return('')
    raise gen.Return(response.body)

  @gen.coroutine
  def _run(self):

    @gen.coroutine
    def fetch_url():
      current_url = yield self._q.get()
      try:
        if current_url in self._fetching:
          return

        print('fetching****** %s' % current_url)
        self._fetching.add(current_url)
        html = yield self.get_page(current_url)
        self._fetched.add(current_url)

        self.handle_page(current_url, html)

        for i in range(self.concurrency):
          if self.urls:
            yield self._q.put(self.urls.pop())

      finally:
        self._q.task_done()

    @gen.coroutine
    def worker():
      while True:
        yield fetch_url()

    self._q.put(self.urls.pop())

    # Start workers, then wait for the work queue to be empty.
    for _ in range(self.concurrency):
      worker()
    yield self._q.join(timeout=timedelta(seconds=300000))
    assert self._fetching == self._fetched

  def run(self):
    io_loop = ioloop.IOLoop.current()
    io_loop.run_sync(self._run)


def run_spider(beg, end):
  urls = []
  for page in range(beg, end):
    urls.append('http://127.0.0.1/%s.htm' % page)
  s = AsySpider(urls, 10)
  s.run()


def main():
  _st = time.time()
  p = Pool()
  all_num = 73000
  num = 4  # number of cpu cores
  per_num, left = divmod(all_num, num)
  s = range(0, all_num, per_num)
  res = []
  for i in range(len(s)-1):
    res.append((s[i], s[i+1]))
  res.append((s[len(s)-1], all_num))
  print res

  for i in res:
    p.apply_async(run_spider, args=(i[0], i[1],))
  p.close()
  p.join()

  print time.time()-_st


if __name__ == '__main__':
  main()

多线程爬虫
线程池实现.

#!/usr/bin/env python
# -*- coding:utf-8 -*-
import Queue
import sys
import requests
import os
import threading
import time

class Worker(threading.Thread):  # 处理工作请求
  def __init__(self, workQueue, resultQueue, **kwds):
    threading.Thread.__init__(self, **kwds)
    self.setDaemon(True)
    self.workQueue = workQueue
    self.resultQueue = resultQueue


  def run(self):
    while 1:
      try:
        callable, args, kwds = self.workQueue.get(False)  # get task
        res = callable(*args, **kwds)
        self.resultQueue.put(res)  # put result
      except Queue.Empty:
        break

class WorkManager:  # 线程池管理,创建
  def __init__(self, num_of_workers=10):
    self.workQueue = Queue.Queue()  # 请求队列
    self.resultQueue = Queue.Queue()  # 输出结果的队列
    self.workers = []
    self._recruitThreads(num_of_workers)

  def _recruitThreads(self, num_of_workers):
    for i in range(num_of_workers):
      worker = Worker(self.workQueue, self.resultQueue)  # 创建工作线程
      self.workers.append(worker)  # 加入到线程队列


  def start(self):
    for w in self.workers:
      w.start()

  def wait_for_complete(self):
    while len(self.workers):
      worker = self.workers.pop()  # 从池中取出一个线程处理请求
      worker.join()
      if worker.isAlive() and not self.workQueue.empty():
        self.workers.append(worker)  # 重新加入线程池中
    print 'All jobs were complete.'


  def add_job(self, callable, *args, **kwds):
    self.workQueue.put((callable, args, kwds))  # 向工作队列中加入请求

  def get_result(self, *args, **kwds):
    return self.resultQueue.get(*args, **kwds)


def download_file(url):
  #print 'beg download', url
  requests.get(url).text


def main():
  try:
    num_of_threads = int(sys.argv[1])
  except:
    num_of_threads = 10
  _st = time.time()
  wm = WorkManager(num_of_threads)
  print num_of_threads
  urls = ['http://www.baidu.com'] * 1000
  for i in urls:
    wm.add_job(download_file, i)
  wm.start()
  wm.wait_for_complete()
  print time.time() - _st

if __name__ == '__main__':
  main()
这三种随便一种都有很高的效率,但是这么跑会给网站服务器不小的压力,尤其是小站点,还是有点节操为好。


数据分析咨询请扫描二维码

客服在线
立即咨询