Python爬虫:连接超时

Question

我正在尝试实现一个简单的web爬虫，我已经编写了一个简单的代码来开始:有两个模块fetcher.py和crawler.py。以下是这些文件：

fetcher.py：

    import urllib2
    import re
    def fetcher(s):
    "fetch a web page from a url"

    try:
            req = urllib2.Request(s)
            urlResponse = urllib2.urlopen(req).read()
    except urllib2.URLError as e:
            print e.reason
            return

    p,q = s.split("//")
    d = q.split("/")
    fdes = open(d[0],"w+")
    fdes.write(str(urlResponse))
    fdes.seek(0)
    return fdes



    if __name__ == "__main__":
    defaultSeed = "http://www.python.org"
    print fetcher(defaultSeed)

crawler.py：

from bs4 import BeautifulSoup
import re
from fetchpage import fetcher    

usedLinks = open("Used","a+")
newLinks = open("New","w+")

newLinks.seek(0)

def parse(fd,var=0):
        soup = BeautifulSoup(fd)
        for li in soup.find_all("a",href=re.compile("http")):
                newLinks.seek(0,2)
                newLinks.write(str(li.get("href")).strip("/"))
                newLinks.write("\n")

        fd.close()
        newLinks.seek(var)
        link = newLinks.readline().strip("\n")

        return str(link)


def crawler(seed,n):
        if n == 0:
                usedLinks.close()
                newLinks.close()
                return
        else:
                usedLinks.write(seed)
                usedLinks.write("\n")
                fdes = fetcher(seed)
                newSeed = parse(fdes,newLinks.tell())
                crawler(newSeed,n-1)

if __name__ == "__main__":
        crawler("http://www.python.org/",7)

问题是，当我运行crawler.py时，它对前4-5链接很好地工作，然后挂起，一分钟后给出以下错误：

[Errno 110] Connection timed out
   Traceback (most recent call last):
  File "crawler.py", line 37, in <module>
    crawler("http://www.python.org/",7)
  File "crawler.py", line 34, in crawler
    crawler(newSeed,n-1)        
 File "crawler.py", line 34, in crawler
    crawler(newSeed,n-1)        
  File "crawler.py", line 34, in crawler
    crawler(newSeed,n-1)        
  File "crawler.py", line 34, in crawler
    crawler(newSeed,n-1)        
  File "crawler.py", line 34, in crawler
    crawler(newSeed,n-1)        
  File "crawler.py", line 33, in crawler
    newSeed = parse(fdes,newLinks.tell())
  File "crawler.py", line 11, in parse
    soup = BeautifulSoup(fd)
  File "/usr/lib/python2.7/dist-packages/bs4/__init__.py", line 169, in __init__
    self.builder.prepare_markup(markup, from_encoding))
  File "/usr/lib/python2.7/dist-packages/bs4/builder/_lxml.py", line 68, in     prepare_markup
    dammit = UnicodeDammit(markup, try_encodings, is_html=True)
  File "/usr/lib/python2.7/dist-packages/bs4/dammit.py", line 191, in __init__
    self._detectEncoding(markup, is_html)
  File "/usr/lib/python2.7/dist-packages/bs4/dammit.py", line 362, in _detectEncoding
    xml_encoding_match = xml_encoding_re.match(xml_data)
TypeError: expected string or buffer

有人能帮我吗?我对python非常陌生，我也不知道为什么说连接超时过了一段时间？

Answer 1

连接超时并不是特定于python的，它只是意味着您向服务器发出了请求，并且服务器没有在应用程序愿意等待的时间内做出响应。

很可能发生这种情况的原因是python.org可能有某种机制来检测它何时从脚本获得多个请求，并且可能只是在4-5个请求之后完全停止为页面服务。除了在不同的站点上尝试脚本之外，没有什么可以避免的。

Answer 2

如前所述，您可以尝试使用代理来避免在多个请求上检测到。您可能需要查看这个答案，以了解如何使用代理发送urllib请求：如何使用urllib通过代理- Python打开网站。