Search code examples
pythonconnectionscreen-scrapingreset

Python urllib2.open Connection reset by peer error


I'm trying to scrape a page using python

The problem is, I keep getting Errno54 Connection reset by peer.

The error comes when I run this code -

urllib2.urlopen("http://www.bkstr.com/webapp/wcs/stores/servlet/CourseMaterialsResultsView?catalogId=10001&categoryId=9604&storeId=10161&langId=-1&programId=562&termId=100020629&divisionDisplayName=Stanford&departmentDisplayName=ILAC&courseDisplayName=126&sectionDisplayName=01&demoKey=d&purpose=browse")

this happens for all the urls on this pag- what is the issue?


Solution

  • $> telnet www.bkstr.com 80
    Trying 64.37.224.85...
    Connected to www.bkstr.com.
    Escape character is '^]'.
    GET /webapp/wcs/stores/servlet/CourseMaterialsResultsView?catalogId=10001&categoryId=9604&storeId=10161&langId=-1&programId=562&termId=100020629&divisionDisplayName=Stanford&departmentDisplayName=ILAC&courseDisplayName=126&sectionDisplayName=01&demoKey=d&purpose=browse HTTP/1.0
    
    Connection closed by foreign host.
    

    You're not going to have any joy fetching that URL from python, or anywhere else. If it works in your browser then there must be something else going on, like cookies or authentication or some such. Or, possibly, the server's broken or they've changed their configuration.

    Try opening it in a browser that you've never accessed that site in before to check. Then log in and try it again.

    Edit: It was cookies after all:

    import cookielib, urllib2
    
    cj = cookielib.CookieJar()
    opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
    #Need to set a cookie
    opener.open("http://www.bkstr.com/")
    #Now open the page we want
    data = opener.open("http://www.bkstr.com/webapp/wcs/stores/servlet/CourseMaterialsResultsView?catalogId=10001&categoryId=9604&storeId=10161&langId=-1&programId=562&termId=100020629&divisionDisplayName=Stanford&departmentDisplayName=ILAC&courseDisplayName=126&sectionDisplayName=01&demoKey=d&purpose=browse").read()
    

    The output looks ok, but you'll have to check that it does what you want :)