I've been making my headers very much like a proxy would, even sending
X-forwarded-for headers with the original client ip and such. Sending
very close to the same thing that would be sent if the user really was
using a web proxy. Seems to be keeping google happy... so far.
Jonathan
Michael
Scratch that, the code is simpler. Serves me right for not checking
things twice:
import urllib,urllib2
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor())
request =
urllib2.Request('http://books.google.com/books?bibkeys=0618379436&jscmd=
viewapi&callback=mycallback')
opener.addheaders = [