源码地址:https://github.com/fatezxl/reptileDemo
一、基础组件
- URL管理器:
管理已经爬取的url和新加入的url; - HTML下载器
下载网页,此时需要注意网页编码; - 数据存储器
将数据存储成文件,或者存储到数据库; - HTML解析器
使用BeautifulSoup4进行HTML解析,提取相关的URL和当前页面的有效信息; - 爬虫调度器
协调以上四个模块组件;
二、组件源码分析
1、URL管理器
#coding:utf-8
class UrlManager(object):
'''
URL管理器
'''
def __init__(self):
self.new_urls = set() #未爬取URL集合
self.old_urls = set() #已爬取URL集合
def has_new_url(self):
"""
判断是否有未爬取的URL
:return:
"""
return self.new_url_size() != 0
def get_new_url(self):
"""
获取一个人未爬取的URL
:return:
"""
new_url = self.new_urls.pop()
self.old_urls.add(new_url)
return new_url
def add_new_url(self,url):
'''
将新的URL添加到未爬取的URL集合中
:param url: 单个URL
:return:
'''
if url is None:
return
if url not in self.new_urls and url not in self.old_urls:
self.new_urls.add(url)
def add_new_urls(self,urls):
'''
将新的URL添加到未爬取的URL集合中
:param urls: url集合
:return:
'''
if urls is None or len(urls) == 0:
return
for url in urls:
self.add_new_url(url)
def new_url_size(self):
'''
获取未爬取URL集合的大小
:return:
'''
return len(self.new_urls)
def old_url_size(self):
'''
获取已经爬取URL集合的大小
:return:
'''
return len(self.old_urls)
2、HTML下载器
#coding:utf-8
import requests
class HtmlDownloader(object):
'''
HTML下载器
'''
def download(self,url):
if url is None:
return None
user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
headers = {
'User-Agent':user_agent
}
r = requests.get(url,headers=headers)
if r.status_code == 200:
# r.encoding = 'utf-8'
return r.text.encode('utf-8')
return None
3、数据存储器
#coding:utf-8
import codecs
class DataOutput(object):
'''
数据存储器
'''
def __init__(self):
self.datas = []
def store_data(self,data):
'''
将解析出来的数据存储在内存中
:param data:
:return:
'''
if data is None:
return
self.datas.append(data)
def output_html(self):
'''
将存储的数据输出为指定的文件格式
:return:
'''
fout = codecs.open('baike.html','w',encoding='utf-8')
fout.write('')
fout.write("")
fout.write("")
fout.write("")
fout.write("this page title 1366_768 1680_1050 ")
print '数组长度:%d' % self.datas.__len__()
print '@#' * 30
for data in self.datas:
fout.write("")
if data['page_url'] != None:
fout.write("this_page " % data['page_url'])
print data['page_url']
if data['title'] != '':
fout.write("%s " % data['title'])
print data['title']
if data['imgurl_1366_768'] != '':
fout.write("1366HD " % data['imgurl_1366_768'])
print data['imgurl_1366_768']
if data['imgurl_1680_1050'] != '':
fout.write("1080BD " % data['imgurl_1680_1050'])
print data['imgurl_1680_1050']
fout.write(" ")
self.datas.remove(data)
fout.write("
")
fout.write("")
fout.write('')
fout.close()
4、HTML解析器
#coding:utf-8
import re
import urlparse
from bs4 import BeautifulSoup
class HtmlParser(object):
'''
HTML解析器
'''
def parser(self,root_url,page_url,html_cont):
'''
用于解析网页内容,抽取URL和数据
:param page_url: 下载页面的URL
:param html_cont: 下载的网页内容
:return: 返回URL和数据
'''
if page_url is None or html_cont is None:
return
soup = BeautifulSoup(html_cont,'html.parser')#加上from_encoding='utf-8'会弹出警告
new_urls = self._get_new_urls(page_url,soup)
new_data = self._get_new_data(root_url,page_url,soup)
print '***' * 30 + '当前页面:'
print page_url
print '--'*30 + 'url集合'
print new_urls
print '--' * 30 + 'data数据'
print new_data
return new_urls,new_data
def _get_new_urls(self,page_url,soup):
'''
抽取新的URL集合
:param page_url: 下载页面的URL
:param soup: soup
:return: 返回新的URL集合
'''
new_urls = set()
#抽取符合要求的a标记
# 注意:此处规则很重要,关乎爬取的内容,此处举例:/fengjing/,/bizhi/7171_88712_2.html
links = soup.find_all('a',href=re.compile(r'/\w+/'))
links_2 = soup.find_all('a', href=re.compile(r'/bizhi/\d+/.html'))
for link in links:
#提取href属性
new_url = link['href']
#拼接成完整网址
new_full_url = urlparse.urljoin(page_url,new_url).encode('utf-8')
new_urls.add(new_full_url)
for link2 in links_2:
# 提取href属性
new_url2 = link2['href']
# 拼接成完整网址
new_full_url2 = urlparse.urljoin(page_url, new_url2).encode('utf-8')
new_urls.add(new_full_url2)
return new_urls
def _get_new_data(self,root_url,page_url,soup):
'''
抽取有效数据
:param root_url:根地址
:param page_url: 下载页面的URL
:param soup:
:return: 返回有效数据
'''
data = {
'title' : '',
'page_url' : '',
'imgurl_1366_768' : '',
'imgurl_1680_1050' : ''
}
#请求页面的地址
data['page_url'] = page_url
#图片标题
#TODO 筛选存在问题
test = soup.find(id='titleName')
print 'title'
print test
# print test.string
if test != None:
print '不是none'
print test.string
title = soup.find(id='titleName').string
print '编码之后:'
print title
data['title'] = title
# 图片的地址
# 1366x768
if soup.find(id='1366x768')!=None:
imgurl_136_768 = soup.find(id='1366x768')['href']
new_full_url_1366_768 = urlparse.urljoin(root_url, imgurl_136_768)
data['imgurl_1366_768'] = new_full_url_1366_768.encode('utf-8')
#1680_1050
if soup.find(id='1680x1050') != None:
imgurl_1680_1050 = soup.find( id='1680x1050')['href']
new_full_url_1680_1050 = urlparse.urljoin(root_url, imgurl_1680_1050)
data['imgurl_1680_1050'] = new_full_url_1680_1050.encode('utf-8')
# 参考案例
# title = soup.find('i',class_='business-icon').find('img')['alt']
# data['title'] = title.get_text()
# summary = soup.find('div',class_='lemma-summary')
# #获取tag中包含的所有文本内容,包括子孙tag中的内容,并将结果作为Unicode字符串返回
# data['summary'] = summary.get_text()
return data
5、爬虫调度器
#coding:utf-8
#导入之前的所有文件
from DataOutput import DataOutput
from HtmlParser import HtmlParser
from HtmlDownloader import HtmlDownloader
from URLManager import UrlManager
class SpiderMan(object):
def __init__(self):
self.manager = UrlManager()
self.downloader = HtmlDownloader()
self.parser = HtmlParser()
self.output = DataOutput()
def crawl(self,root_url):
#添加入口URL
self.manager.add_new_url(root_url)
#判断url管理器中是否有新的url,同时判断抓取了多少个url,抓取数据数量限定在0——100之间
while(self.manager.has_new_url() and self.manager.old_url_size() < 100):
try:
#从URL管理器获取新的url
new_url = self.manager.get_new_url()
#HTML下载器下载页面
html = self.downloader.download(new_url)
#HTML解析器抽取网页数据
new_urls,data = self.parser.parser(root_url,new_url,html)
#将抽取的url添加到URL管理器中
self.manager.add_new_urls(new_urls)
print '有待爬取的url数量:%d' % self.manager.new_url_size()
#数据存储器存储文件
self.output.store_data(data)
print "已经抓取%s个链接"%self.manager.old_url_size()
except Exception as e:
print "crawl failed"
print e.message
#数据存储器将文件输出成指定格式
self.output.output_html()
if __name__ == "__main__":
spider_man = SpiderMan()
spider_man.crawl("http://desk.zol.com.cn/")