如何处理BeautifulSoup网页无法加载的情况

Question

目前，如果在获取网页时出现错误，soup 就不会填充网页内容，而是会得到 beautifulsoup 的默认返回值。

我想找一种方法来检查这个情况，这样如果获取网页时出错，我就可以跳过一段代码，比如

if soup:
  do stuff

但我不想完全终止程序。抱歉我问的问题有点初级。

def getwebpage(address):
  try:
      user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
      headers = { 'User-Agent' : user_agent }
      req = urllib2.Request(address, None, headers)
      web_handle = urllib2.urlopen(req)
  except urllib2.HTTPError, e:
      error_desc = BaseHTTPServer.BaseHTTPRequestHandler.responses[e.code][0]
      appendlog('HTTP Error: ' + str(e.code) + ': ' + address)
      return
  except urllib2.URLError, e:
      appendlog('URL Error: ' + e.reason[1] + ': ' + address)
      return
  except:
      appendlog('Unknown Error: ' + address)
      return
  return web_handle


def test():
  soup = BeautifulSoup(getwebpage('http://doesnotexistblah.com/'))
  print soup

  if soup:
    do stuff

test()

Answer 1

soup = getwebpage('http://doesnotexistblah.com/')
if soup is not None:
    soup = BeautifulSoup(soup)

这就是你想要的吗？

Answer 2

把代码整理一下，让一个函数专门负责从网址获取数据，另一个函数则专门处理这些数据。

import urllib2, httplib
from BeautifulSoup import BeautifulSoup

def append_log(message):
    print message

def get_web_page(address):
    try:
        user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
        headers = { 'User-Agent' : user_agent }
        request = urllib2.Request(address, None, headers)
        response = urllib2.urlopen(request, timeout=20)
        try:
            return response.read()
        finally:
            response.close()
    except urllib2.HTTPError as e:
        error_desc = httplib.responses.get(e.code, '')
        append_log('HTTP Error: ' + str(e.code) + ': ' +
                  error_desc + ': ' + address)
    except urllib2.URLError as e:
        append_log('URL Error: ' + e.reason[1] + ': ' + address)
    except Exception as e:
        append_log('Unknown Error: ' + str(e) + address)

def process_web_page(data):
    if data is not None:
        print BeautifulSoup(data)
    else:
        pass # do something else

data = get_web_page('http://doesnotexistblah.com/')
process_web_page(data)

data = get_web_page('http://docs.python.org/copyright.html')
process_web_page(data)