python - A problem happen when I'm trying to fetch documents from a website

I tried to download documents from this page Securities Class Action Filings

I tried to download the 25 documents on the page. I thought it was simple, and here's my code:

from BeautifulSoup import BeautifulSoup
import re
import urllib2
import os

if __name__ == "__main__":
  pre_url = "http://securities.stanford.edu"
  url = "http://securities.stanford.edu/fmi/xsl/SCACPUDB/recordlist.xsl?-db=SCACPUDB&-lay=Search&FIC_DateFiled_Quater=Q1&FIC_DateFiled_Year=2011&-sortfield.1=FIC_DateFiled&-sortfield.2=LitigationName&-sortorder.1=ascend&开发者_如何学Goamp;-max=25&-find" 
  response = urllib2.urlopen(url)
  soup = BeautifulSoup(response.read()).findAll('tr')
  url_list = []
  for s in soup[8:]:
    url_list.append(pre_url + s.a['href'])
  for x in url_list:
    name = x.split("/")[4]  
    context = urllib2.urlopen(x).read()
    soup = BeautifulSoup(context)
    file = open(name + ".txt", "w")
    file.write(soup.prettify())
  print "DONE"

After executing the script, I downloaded 25 files successfully. But then I found 10 of them are full of garbage characters! How come? Can anyone help me?

Thanks a lot, and I'm sorry for my poor English.

Update: This is one of the pages which would be downloaded incorrectly by the script http://securities.stanford.edu/1046/BWEN00_01/

The sample page is encoded in UTF-16 without properly providing that factoid in the header.

>>> page = urllib2.urlopen( "http://securities.stanford.edu/1046/BWEN00_01" )
>>> page.info().headers
['Date: Mon, 22 Aug 2011 13:13:56 GMT\r\n', 'Server: Apache/1.3.33 (Darwin) mod_jk/1.2.2 DAV/1.0.3\r\n', 'Cache-Control: max-age=60\r\n', 'Expires: Mon, 22 Aug 2011 13:14:56 GMT\r\n', 'Last-Modified: Thu, 21 Jul 2011 22:06:51 GMT\r\n', 'ETag: "18b9a6e-9af6-4e28a2fb"\r\n', 'Accept-Ranges: bytes\r\n', 'Content-Length: 39670\r\n', 'Connection: close\r\n', 'Content-Type: text/html\r\n']

Try page.decode('utf-16') to see the page in proper Unicode characters instead of bytes.

open(name + ".txt", "w")

It's possible that your problem is that you're opening the files in text mode, but they're being downloaded in binary mode. Replace the above expression with

open(name + ".txt", "wb")

and see if it improves things.