To generate new headers for scraping purposes.

header_gen.py

"""
Expose methods to return new header dictionary. Supports only GET requests at the moment.
(Using phantomJS to throw them off. Maybe overkill. Could have used simple get request.)
"""
from selenium import webdriver
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities


# edit desired capabilities
dcap = dict(DesiredCapabilities.PHANTOMJS)
dcap["phantomjs.page.settings.userAgent"] = (
    "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/53 "
    "(KHTML, like Gecko) Chrome/15.0.87"
)
dcap['pahntomjs.page.settings.loadImages'] = False

driver = webdriver.PhantomJS('/home/ubuntu/.nvm/v4.2.4/bin/phantomjs', desired_capabilities=dcap)


def get_new_cookies(url):
    """Return new cookie string"""
    try:
        driver.delete_all_cookies()
        driver.get(url)
        cookies = driver.get_cookies()
        cookie_str = ' ; '.join([c['name'] + '=' + c['value'] for c in cookies])
        return cookie_str
    except Exception as e:
        print('ERROR: failed to get new cookies. Caused by: %s ' % e.__str__())
        return None


def get_new_headers(url):
    """Return new header dict"""
    cookies = get_new_cookies(url)
    if cookies:
        headers = {
            'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.111 Safari/537.36',
            'Cookie': cookies,
            'Accept': '*/*',
            'X-Requested-With': 'XMLHttpRequest'
        }
    else:
        headers = None
    return headers