Scrape multiple accounts aka multiple logins

Question

I can successfully scrape data for a single account. I want to scrape multiple accounts on a single website, which means multiple logins. How do I manage login/logout?

Answer 1

you can scrape multiples accounts in parallel using multiple cookiejars per account session, see "cookiejar" request meta key at http://doc.scrapy.org/en/latest/topics/downloader-middleware.html?highlight=cookiejar#std:reqmeta-cookiejar

To clarify: suppose we have an array of accounts in settings.py:

MY_ACCOUNTS = [
    {'login': 'my_login_1', 'pwd': 'my_pwd_1'},
    {'login': 'my_login_2', 'pwd': 'my_pwd_2'},
]

And this is the link to the login page: http://example.com/login

Create start_requestsfunction in your spider, in this function we can loop on the MY_ACCOUNTS array and login to each account:

def start_requests(self):
    requests = []

    for i, account in enumerate(self.crawler.settings['MY_ACCOUNTS']):
        request = FormRequest('http://example.com/login', 
            formdata={'form_login_name': account['login'], 'form_pwd_name': account['pwd']}, 
            callback=self.parse,
            dont_filter=True)

        request.meta['cookiejar'] = i
        requests.append(request)

    return requests

form_login_name and form_pwd_name are respectively fields names on the login form.

dont_filter=True For ignoring filter on duplicate requests, because here we make a POST request to login on the same page http://example.com/login

request.meta['cookiejar'] = i to separate cookies of each session(login), dont forget to add cookiejar identifier in your sub request, suppose you want to redirect scrapy to a page after login:

def parse(self, response): 
    """ make some manipulation here ... """

    yield Request(my_url, meta={'cookiejar': response.meta['cookiejar']}, callback = my_callback)