"joe_public34" <[EMAIL PROTECTED]> writes:
> Hello all,
>
> I'm trying to write a script to log into Yahoo! (mail, groups, etc),
I don't have an answer to your problem but I *do* have a script that
logs into yahoo. It's ugly, but it works. I'm refactoring it - if I
ever get back to it.
I've appended the crucial code, but haven't tried running this
excerpt. If you have problems or questions, feel free to ask.
<mike
myYahooID = 'Your Id Here'
from urllib2 import build_opener, HTTPCookieProcessor
from urllib import urlencode
from BeautifulSoup import BeautifulSoup
from sys import argv, exit
urlopen = build_opener(HTTPCookieProcessor()).open
def login(password):
"""Get past the multiple login forms that Yahoo uses."""
print "Logging in to Yahoo groups."
passage = BeautifulSoup(urlopen('http://groups.yahoo.com').read())
passage.done()
i = 0
while passage and i < 5:
page = passage
passage = handle_login(passage, password)
i = i + 1
if i >= 5:
handle_bogus(page, 'To many login attempts') # Never returns
while 1:
metas = page.fetch('meta', {'http-equiv': 'Refresh'})
if not metas:
return page
# Sigh. It's redirect page that didn't get handled by the library.
page = meta_redirect(metas[0]['content'])
return page
def handle_login(soup, password):
"""Parse a page for login information, and hand back the result of logging
in.
Returns None if there is no login form."""
# Get the login page, scrape the form, and log in!
forms = soup.fetch('form', {'name': 'login_form'})
if not forms:
return None
form = forms[0]
postdata = []
for intag in form.fetchNext('input'):
if intag['type'] == 'hidden':
postdata.append((intag['name'], intag['value']))
elif intag['type'] == 'password':
postdata.append((intag['name'], password))
elif intag['type'] == 'checkbox':
postdata.append((intag['name'], 'n'))
elif intag['type'] == 'text' and intag['name'] == 'login':
postdata.append(('login', myYahooID))
elif intag['type'] == 'submit':
if intag.has_key(['name']):
postdata.append((intag['name'], intag['value']))
else:
postdata.append(('submit', intag['value']))
else:
print "Login form had unrecognized input tag:", str(intag)
out = BeautifulSoup(urlopen(form['action'], urlencode(postdata)).read())
out.done()
return out
def meta_redirect(value):
"""Handle a http-equiv redirect, since the library isn't reliable."""
for spam in value.split(';'):
stuff = spam.strip().split('=')
if stuff and stuff[0] == 'url':
out = BeautifulSoup(urlopen(stuff[1]).read())
out.done()
return out
--
Mike Meyer <[EMAIL PROTECTED]> http://www.mired.org/home/mwm/
Independent WWW/Perforce/FreeBSD/Unix consultant, email for more information.
--
http://mail.python.org/mailman/listinfo/python-list
