def get_info(id): data = {'username': '******', 'password': '******'} url = 'http://informatics.mccme.ru/login/index.php' #browser = webdriver.Chrome('/Users/mihail/Desktop/chromedriver') #browser.get(url) #request_cookies_browser = browser.get_cookies() #s = requests.Session() #c = [s.cookies.set(c['name'], c['value']) for c in request_cookies_browser] #resp = s.post(url, data) #dict_resp_cookies = resp.cookies.get_dict() #response_cookies_browser = [{'name': name, 'value': value} for name, value in dict_resp_cookies.items()] #c = [browser.add_cookie(c) for c in response_cookies_browser] #browser.get(url) #html = browser.page_source #browser.get('http://informatics.mccme.ru/submits/view.php?user_id=' + id) #WebDriverWait(browser, 10).until( # ajax_complete, "Timeout waiting for page to load") #html = browser.page_source browser = Browser() browser.load('http://informatics.mccme.ru/submits/view.php?user_id=' + id) browser.wait_load() html = browser.html print(html)
def process_request(self, request, spider): browser = Browser() browser.create_webview() browser.set_html_parser(PyQuery) browser.load(request.url, 20) try: browser.wait_load(10) except: print '###########ERROR###########' pass body = browser.html body = body.encode('utf-8') renderedBody = str(body) return HtmlResponse(request.url, body=renderedBody)
def process_request(self, request, spider): browser = Browser() browser.create_webview() browser.set_html_parser(PyQuery) browser.load(request.url, 20) try: browser.wait_load(10) except: print '###########ERROR###########' pass body = browser.html body = body.encode('utf-8') renderedBody = str(body) return HtmlResponse(request.url, body=renderedBody)