Making HTTP POST request

Question

I'm trying to make a POST request to retrieve information about a book. Here is the code that returns HTTP code: 302, Moved

import httplib, urllib
params = urllib.urlencode({
    'isbn' : '9780131185838',
    'catalogId' : '10001',
    'schoolStoreId' : '15828',
    'search' : 'Search'
    })
headers = {"Content-type": "application/x-www-form-urlencoded",
           "Accept": "text/plain"}
conn = httplib.HTTPConnection("bkstr.com:80")
conn.request("POST", "/webapp/wcs/stores/servlet/BuybackSearch",
             params, headers)
response = conn.getresponse()
print response.status, response.reason
data = response.read()
conn.close()

When I try from a browser, from this page: http://www.bkstr.com/webapp/wcs/stores/servlet/BuybackMaterialsView?langId=-1&catalogId=10001&storeId=10051&schoolStoreId=15828 , it works. What am I missing in my code?

EDIT: Here's what I get when I call print response.msg

302 Moved Date: Tue, 07 Sep 2010 16:54:29 GMT
Vary: Host,Accept-Encoding,User-Agent
Location: http://www.bkstr.com/webapp/wcs/stores/servlet/BuybackSearch
X-UA-Compatible: IE=EmulateIE7
Content-Length: 0
Content-Type: text/plain; charset=utf-8

Seems that the location points to the same url I'm trying to access in the first place?

EDIT2:

I've tried using urllib2 as suggested here. Here is the code:

import urllib, urllib2

url = 'http://www.bkstr.com/webapp/wcs/stores/servlet/BuybackSearch'
values = {'isbn' : '9780131185838',
          'catalogId' : '10001',
          'schoolStoreId' : '15828',
          'search' : 'Search' }


data = urllib.urlencode(values)
req = urllib2.Request(url, data)
response = urllib2.urlopen(req)
print response.geturl()
print response.info()
the_page = response.read()
print the_page

And here is the output:

http://www.bkstr.com/webapp/wcs/stores/servlet/BuybackSearch
Date: Tue, 07 Sep 2010 16:58:35 GMT
Pragma: No-cache
Cache-Control: no-cache
Expires: Thu, 01 Jan 1970 00:00:00 GMT
Set-Cookie: JSESSIONID=0001REjqgX2axkzlR6SvIJlgJkt:1311s25dm; Path=/
Vary: Accept-Encoding,User-Agent
X-UA-Compatible: IE=EmulateIE7
Content-Length: 0
Connection: close
Content-Type: text/html; charset=utf-8
Content-Language: en-US
Set-Cookie: TSde3575=225ec58bcb0fdddfad7332c2816f1f152224db2f71e1b0474c866f3b; Path=/

Caeoma

The 302 response also indicated where it was moved to - find that URL and use it. – Excursus

Answer 1

Their server seems to want you to acquire the proper cookie. This works:

import urllib, urllib2, cookielib

cookie_jar = cookielib.CookieJar()
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookie_jar))
urllib2.install_opener(opener)

# acquire cookie
url_1 = 'http://www.bkstr.com/webapp/wcs/stores/servlet/BuybackMaterialsView?langId=-1&catalogId=10001&storeId=10051&schoolStoreId=15828'
req = urllib2.Request(url_1)
rsp = urllib2.urlopen(req)

# do POST
url_2 = 'http://www.bkstr.com/webapp/wcs/stores/servlet/BuybackSearch'
values = dict(isbn='9780131185838', schoolStoreId='15828', catalogId='10001')
data = urllib.urlencode(values)
req = urllib2.Request(url_2, data)
rsp = urllib2.urlopen(req)
content = rsp.read()

# print result
import re
pat = re.compile('Title:.*')
print pat.search(content).group()

# OUTPUT: Title:&nbsp;&nbsp;Statics & Strength of Materials for Arch (w/CD)<br />

Myrmecophagous

@infrared: Glad to help. I probably should have added that one method to solve these types of things is to run an HTTP proxy which shows you the trace of requests/responses. Then, use the browser and your code, and compare the two traces. Usually, you're looking for differences between cookies or headers. Sometimes it takes a bit of trial and error. I like to use Fiddler, but any such tool will do. – Myrmecophagous

Answer 2

You might want to use the urllib2 module which should handle redirects better. Here's an example of POSTING with urllib2.

Bow

Answer 3

1. Perhaps that's what the browser gets, and you'll just have to follow the 302 redirect.

2. If all else fails, you can monitor the dialogue between Firefox and the Web Server using FireBug or tcpdump or wireshark, and see which HTTP headers are different. Possibly it's just the User Agent: header.

Delvalle