Submitting forms over HTTPS with mechanize

[Larry Bates]

Rex wrote:
> Hello,
> 
> I am working on an academic research project where I need to log in to
> a website (www.lexis.com) over HTTPS and execute a bunch of queries to
> gather a data set. I just discovered the mechanize module, which seems
> great because it's a high-level tool. However, I can't find any decent
> documentation for mechanize apart from the docstrings, which are
> pretty thin. So I just followed some other examples I found online, to
> produce the following:
> 
> baseurl = 'http://www.lexis.com/'
> br = mechanize.Browser()
> br.set_handle_robots(False)
> br.addheaders = [('User-Agent', 'Firefox')]
> br.open(baseurl)
> br.select_form(name="formauth")
> br["USER_ID"]="my_user_id"
> br["PASSWORD"]="my_password"
> result = br.submit()
> 
> This code hangs at br.submit(), and I can't tell what I'm doing wrong.
> Typically I would inspect the HTTP data with an HTTP debugging proxy
> (Fiddler), but I guess since this is HTTPS I can't do that. Any
> glaring errors in my code?
> 
> By the way, does anyone have suggestions for Python modules that I
> should use instead of mechanize (and that are sufficiently easy)? If
> mechanize fails, I might try modifying some similar Perl code a friend
> sent me that logs into lexis.com.
> 
> Thanks so much,
> 
> Rex

I've used mechanize quite successfully but others have suggested Twill
http://twill.idyll.org/.  It seems to be at least documented.

-Larry