Develop Reference

Python Selenium scrape data when button "Load More" doesnt change URL

I am using the following code to attempt to keep clicking a "Load More" button until all page results are shown on the website:
from selenium import webdriver
import time
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
def startWebDriver():
global driver
chrome_options = webdriver.ChromeOptions()
chrome_options.add_argument("--incognito")
chrome_options.add_argument("--window-size=1920x1080")
driver = webdriver.Chrome(options = chrome_options)
startWebDriver()
driver.get("https://together.bunq.com/all")
time.sleep(4)
while True:
try:
wait = WebDriverWait(driver, 10,10)
element = wait.until(EC.element_to_be_clickable((By.CSS_SELECTOR, "[title='Load More']")))
element.click()
print("Loading more page results")
except:
print("All page results displayed")
break;
However, since the button click does not change the URL, no new data is loaded into chromedriver and the while loop will break on the second iteration.

Selenium is overkill for this. You only need requests. Logging one's network traffic reveals that at some point JavaScript makes an XHR HTTP GET request to a REST API endpoint, the response of which is JSON and contains all the information you're likely to want to scrape.
One of the query-string parameters for that endpoint URL is page[offset], which is used to offset the query results for pagination (in this case the "load more button"). A value of 0 corresponds to no offset, or "start at the beginning". Increment this value to suit your needs - in a loop would probably be a good place to do this.
Simply imitate that XHR HTTP GET request - copy the API endpoint URL and query-string parameters and request headers, then parse the JSON response:
def get_discussions():
import requests
url = "https://together.bunq.com/api/discussions"
params = {
"include": "user,lastPostedUser,tags,firstPost",
"page[offset]": 0
}
headers = {
"user-agent": "Mozilla/5.0"
}
response = requests.get(url, params=params, headers=headers)
response.raise_for_status()
yield from response.json()["data"]
def main():
for discussion in get_discussions():
print(discussion["attributes"]["title"])
return 0
if __name__ == "__main__":
import sys
sys.exit(main())
Output:
⚡️What’s new in App Update 18.8.0
Local Currencies Accounts Fees
Local Currencies 💸
Spanish IBANs Service Coverage 🇪🇸
bunq Update 18 FAQ 📚
Phishing and Spoofing - The new ways of scamming, explained 👀
Easily rent a car with your bunq credit card 🚗
Giveaway - Hallo Deutschland! 🇩🇪
Giveaway - Hello Germany! 🇩🇪
True Name: Feedback 💬
True Name 🌈
What plans are available?
Everything about refunds 💸
Identity verification explained! 🤳
When will I receive my payment?
Together Community Guidelines 📣
What is my Tax Identification Number (TIN)?
How do I export a bank statement?
How do I change my contact info?
Large cash withdrael
If this is a new concept for you, I would suggest you look up tutorials on how to use your browser's developer tools (Google Chrome's Devtools, for example), how to log your browser's network traffic, REST APIs, HTTP, etc.

Creating POST request to scrape website with python where no network form data changes

I am scraping a website that dynamically renders with javascript. The urls don't change when hitting the > button So I have been trying to look at the inspector in the network section and more specifically the "General" section for the "Request Url" and the "Request Method" as well as in the "Form Data" section looking for any sort of ID that could be unique to distinguish each successive page. However when recording a log of clicking the > button from page to page the "Form Data" data seems to be the same each time (See images):
Currently my code doesn't incorporate this method because I can't see it helping until I can find a unique identifier in the "Form Data" section. However, I can show my code if helpful. In essence it just pulls the first page of data over and over again in my while loop even though I'm using a driver with selenium and using driver.find_elements_by_xpath("xpath of > button").click() before trying to get the data with BeautifulSoup.
(Updated code see comments)
from selenium import webdriver
from selenium.webdriver.common.by import By
from bs4 import BeautifulSoup
import pandas as pd
from pandas import *
masters_list = []
def extract_info(html_source):
# html_source will be inner HTMl of table
global lst
soup = BeautifulSoup(html_source, 'html.parser')
lst = soup.find('tbody').find_all('tr')[0]
masters_list.append(lst)
# i am printing just id because it's id set as crypto name you have to do more scraping to get more info
chrome_driver_path = '/Users/Justin/Desktop/Python/chromedriver'
driver = webdriver.Chrome(executable_path=chrome_driver_path)
url = 'https://cryptoli.st/lists/fixed-supply'
driver.get(url)
loop = True
while loop: # loop for extrcting all 120 pages
crypto_table = driver.find_element(By.ID, 'DataTables_Table_0').get_attribute(
'innerHTML') # this is for crypto data table
extract_info(crypto_table)
paginate = driver.find_element(
By.ID, "DataTables_Table_0_paginate") # all table pagination
pages_list = paginate.find_elements(By.TAG_NAME, 'li')
# we clicking on next arrow sign at last not on 2,3,.. etc anchor link
next_page_link = pages_list[-1].find_element(By.TAG_NAME, 'a')
# checking is there next page available
if "disabled" in next_page_link.get_attribute('class'):
loop = False
pages_list[-1].click() # if there next page available then click on it
df = pd.DataFrame(masters_list)
print(df)
df.to_csv("crypto_list.csv")
driver.quit()

I am using my own code to show how i am getting the table i add explanation as comment for important line
from selenium import webdriver
from selenium.webdriver.common.by import By
from bs4 import BeautifulSoup
def extract_info(html_source):
soup = BeautifulSoup(html_source,'html.parser') #html_source will be inner HTMl of table
lst = soup.find('tbody').find_all('tr')
for i in lst:
print(i.get('id')) # i am printing just id because it's id set as crypto name you have to do more scraping to get more info
driver = webdriver.Chrome()
url = 'https://cryptoli.st/lists/fixed-supply'
driver.get(url)
loop = True
while loop: #loop for extrcting all 120 pages
crypto_table = driver.find_element(By.ID,'DataTables_Table_0').get_attribute('innerHTML') # this is for crypto data table
print(extract_info(crypto_table))
paginate = driver.find_element(By.ID, "DataTables_Table_0_paginate") # all table pagination
pages_list = paginate.find_elements(By.TAG_NAME,'li')
next_page_link = pages_list[-1].find_element(By.TAG_NAME,'a') # we clicking on next arrow sign at last not on 2,3,.. etc anchor link
if "disabled" in next_page_link.get_attribute('class'): # checking is there next page available
loop = False
pages_list[-1].click() # if there next page available then click on it
so main answer of your question is when you click on button, selenium update the page then you can use driver.page_source to get updated html. some times (*not this url) page can have ajax request which can take some time so you have to wait till the selenium load the full page.

Shell script to download a lot of HTML files and store them statically with all CSS

I have posted on a science forum (roughly 290 questions) that I would like to get back by downloading them with all the associated answers.
The first issue is that I have to be logged on my personal space to have the list of all the messages. How to circumvent this first barrier to be able with a shell script or a single wget command to get back all URL and their content. Can I pass to wgeta login and a password to be logged and redirected to the appropriate URL obtaining the list of all messages?
Once this first issue will be solved, the second issue is that I have to start from 6 different menu pages that all contain the title and the link of the questions.
Moreover, concerning some of my questions, the answers and the discussions may be on multiple pages.
So I wonder if I could achieve this operation of global downloading knowing I would like to store them statically with local CSS stored also on my computer (to keep the same format into my browser when I consult them on my PC).
The URL of the first menu page of questions is (once I am logged on the website : that could be an issue also to download with wget if I am obliged to be connected).
An example of URL containing the list of messages, once I am logged, is:
https://forums.futura-sciences.com/search.php?searchid=22897684
The other pages (there all 6 or 7 pages of discussions title in total appering in the main menu page) have the format:
"https://forums.futura-sciences.com/search.php?searchid=22897684&pp=&page=2" (for page 2).
https://forums.futura-sciences.com/search.php?searchid=22897684&pp=&page=5
(for page 5)
One can see on each of these pages the title and the link of each of the discussions that I would like to download with also CSS (knowing each discussion may contain multiple pages also) :
for example the first page of discussion "https://forums.futura-sciences.com/archives/804364-demonstration-dilatation-temps.html"
has page 2: "https://forums.futura-sciences.com/archives/804364-demonstration-dilatation-temps-2.html"
and page 3: "https://forums.futura-sciences.com/archives/804364-demonstration-dilatation-temps-3.html"
Naively, I tried to do all this with only one command (with the example of URL on my personal space that I have taken at the beginning of post, i.e "https://forums.futura-sciences.com/search.php?searchid=22897684"):
wget -r --no-check-certificate --html-extension --convert-links "https://forums.futura-sciences.com/search.php?searchid=22897684"
but unfortunately, this command downloads all files, and even maybe not what I want, i.e my discussions.
I don't know what the approach to use: must I firstly store all URL in a file (with all sub-pages containing all answers and the global discussion for each of mu initial question)?
And after, I could do maybe a wget -i all_URL_questions.txt. How can I carry out this operation?
Update
My issue needs a script, I tried with Python the following things:
1)
import urllib, urllib2, cookielib
username = 'USERNAME'
password = 'PASSWORD'
cj = cookielib.CookieJar()
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cj))
login_data = urllib.urlencode({'username' : username, 'password' : password})
opener.open('https://forums.futura-sciences.com/login.php', login_data)
resp = opener.open('https://forums.futura-sciences.com/search.php?do=finduser&userid=253205&contenttype=vBForum_Post&showposts=1')
print resp.read()
But the page printed is not the page of my home into personal space.
2)
import requests
# Fill in your details here to be posted to the login form.
payload = {
'inUserName': 'USERNAME',
'inUserPass': 'PASSWORD'
}
# Use 'with' to ensure the session context is closed after use.
with requests.Session() as s:
p = s.post('https://forums.futura-sciences.com/login.php?do=login', data=payload)
# print the html returned or something more intelligent to see if it's a successful login page.
print p.text.encode('utf8')
# An authorised request.
r = s.get('https://forums.futura-sciences.com/search.php?do=finduser&userid=253205&contenttype=vBForum_Post&showposts=1')
print r.text.encode('utf8')
Here too, this doesn't work
3)
import requests
import bs4
site_url = 'https://forums.futura-sciences.com/login.php?do=login'
userid = 'USERNAME'
password = 'PASSWWORD'
file_url = 'https://forums.futura-sciences.com/search.php?do=finduser&userid=253205&contenttype=vBForum_Post&showposts=1'
o_file = 'abc.html'
# create session
s = requests.Session()
# GET request. This will generate cookie for you
s.get(site_url)
# login to site.
s.post(site_url, data={'vb_login_username': userid, 'vb_login_password': password})
# Next thing will be to visit URL for file you would like to download.
r = s.get(file_url)
# Download file
with open(o_file, 'wb') as output:
output.write(r.content)
print("requests:: File {o_file} downloaded successfully!")
# Close session once all work done
s.close()
Same thing, the content is wrong
4)
from selenium import webdriver
# To prevent download dialog
profile = webdriver.FirefoxProfile()
profile.set_preference('browser.download.folderList', 2) # custom location
profile.set_preference('browser.download.manager.showWhenStarting', False)
profile.set_preference('browser.download.dir', '/tmp')
profile.set_preference('browser.helperApps.neverAsk.saveToDisk', 'text/csv')
webdriver.get('https://forums.futura-sciences.com/')
webdriver.find_element_by_id('ID').send_keys('USERNAME')
webdriver.find_element_by_id ('ID').send_keys('PASSWORD')
webdriver.find_element_by_id('submit').click()
browser = webdriver.Firefox()
browser.get('https://forums.futura-sciences.com/search.php?do=finduser&userid=253205&contenttype=vBForum_Post&showposts=1')
Still not getting to log in with USERNAME and PASSSWORD and get content of homepage of personal space
5)
from selenium import webdriver
from selenium.webdriver.firefox.webdriver import FirefoxProfile
from selenium.webdriver.firefox.options import Options
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.common.by import By
import time
def MS_login(username, passwd): # call this with username and password
firefox_capabilities = DesiredCapabilities.FIREFOX
firefox_capabilities['moz:webdriverClick'] = False
driver = webdriver.Firefox(capabilities=firefox_capabilities)
fp = webdriver.FirefoxProfile()
fp.set_preference("browser.download.folderList", 2) # 0 means to download to the desktop, 1 means to download to the default "Downloads" directory, 2 means to use the directory
fp.set_preference("browser.download.dir","/Users/user/work_archives_futura/")
driver.get('https://forums.futura-sciences.com/') # change the url to your website
time.sleep(5) # wait for redirection and rendering
driver.delete_all_cookies() # clean up the prior login sessions
driver.find_element_by_xpath("//input[#name='vb_login_username']").send_keys(username)
elem = WebDriverWait(driver, 10).until(EC.visibility_of_element_located((By.XPATH, "//input[#name='vb_login_password']")))
elem.send_keys(Keys.TAB)
driver.find_element_by_xpath("//input[#type='submit']").click()
print("success !!!!")
driver.close() # close the browser
return driver
if __name__ == '__main__':
MS_login("USERNAME","PASSWORD")
The window is well opened, username filled but impossible to fill or submit the password and click on submit.
PS: the main issue could come from that password field has display:none property, So I can't simulate TAB operation to password field and pass it, once I have put the login.

It seems you're pretty knowledgeable already about scraping using the various methods. All that was missing were the correct field names in the post request.
I used the chrome dev tools (f12 - then go to networking tab). With this open if you login and quickly stop the browser window from redirecting, you'll be able to see the full request to login.php and look at the fields etc.
With that I was able to build this for you. It includes a nice dumping function for responses. To test my code works you can use your real password for positive case and the bad password line for negative case.
import requests
import json
s = requests.Session()
def dumpResponseData(r, fileName):
print(r.status_code)
print(json.dumps(dict(r.headers), indent=1))
cookieDict = s.cookies.get_dict()
print(json.dumps(cookieDict, indent=1))
outfile = open(fileName, mode="w")
outfile.write(r.text)
outfile.close()
username = "your-username"
password = "your-password"
# password = "bad password"
def step1():
data = dict()
data["do"] = "login"
data["vb_login_md5password"] = ""
data["vb_login_md5password_utf"] = ""
data["s"] = ""
data["securitytoken"] = "guest"
data["url"] = "/search.php?do=finduser&userid=1077817&contenttype=vBForum_Post&showposts=1"
data["vb_login_username"] = username
data["vb_login_password"] = password
p = s.post('https://forums.futura-sciences.com/login.php?do=login', data=data)
# Logged In?
if "vbseo_loggedin" in s.cookies.keys():
print("Logged In!")
else:
print("Login Failed :(")
if __name__ == "__main__":
step1()
I don't have any posts in my newly created Futura account so I can't really do any more testing for you - I don't want to spam their forum with garbage.
But I would probably start by doing a request of post search url and scrape the links using bs4.
Then you could probably just use wget -r for each link you've scraped.

#Researcher is correct on their advice when it comes to the requests library. You are not posting all of the request params that the browser would send. Overall, I think it will be difficult to get requests to pull everything when you factor in static content and client side javascript
Your selenium code from section 4 has a few mistakes in it:
# yours
webdriver.find_element_by_id('ID').send_keys('USERNAME')
webdriver.find_element_by_id ('ID').send_keys('PASSWORD')
webdriver.find_element_by_id('submit').click()
# should be
webdriver.find_element_by_id('vb_login_username').send_keys('USERNAME')
webdriver.find_element_by_id('vb_login_password').send_keys('PASSWORD')
webdriver.find_element_by_xpath("//input[#type='submit']").click()
You may need to fiddle with the xpath for the submit button.
Hint: You can debug along the way by taking a screenshots :
webdriver.find_element_by_id('vb_login_username').send_keys('USERNAME')
webdriver.find_element_by_id('vb_login_password').send_keys('PASSWORD')
webdriver.get_screenshot_as_file('before_submit.png')
webdriver.find_element_by_xpath("//input[#type='submit']").click()
webdriver.get_screenshot_as_file('after_submit.png')

Loop through url with Selenium Webdriver

The below request finds the contest id's for the day. I am trying to pass that str into the driver.get url so it will go to each individual contest url and download each contests CSV. I would imagine you have to write a loop but I'm not sure what that would look like with a webdriver.
import time
from selenium import webdriver
import requests
import datetime
req = requests.get('https://www.draftkings.com/lobby/getlivecontests?sport=NBA')
data = req.json()
for ids in data:
contest = ids['id']
driver = webdriver.Chrome() # Optional argument, if not specified will search path.
driver.get('https://www.draftkings.com/account/sitelogin/false?returnurl=%2Flobby');
time.sleep(2) # Let DK Load!
search_box = driver.find_element_by_name('username')
search_box.send_keys('username')
search_box2 = driver.find_element_by_name('password')
search_box2.send_keys('password')
submit_button = driver.find_element_by_xpath('//*[#id="react-mobile-home"]/section/section[2]/div[3]/button/span')
submit_button.click()
time.sleep(2) # Let Page Load, If not it will go to Account!
driver.get('https://www.draftkings.com/contest/exportfullstandingscsv/' + str(contest) + '')

Try in following order:
import time
from selenium import webdriver
import requests
import datetime
req = requests.get('https://www.draftkings.com/lobby/getlivecontests?sport=NBA')
data = req.json()
driver = webdriver.Chrome() # Optional argument, if not specified will search path.
driver.get('https://www.draftkings.com/account/sitelogin/false?returnurl=%2Flobby')
time.sleep(2) # Let DK Load!
search_box = driver.find_element_by_name('username')
search_box.send_keys('Pr0c3ss')
search_box2 = driver.find_element_by_name('password')
search_box2.send_keys('generic1!')
submit_button = driver.find_element_by_xpath('//*[#id="react-mobile-home"]/section/section[2]/div[3]/button/span')
submit_button.click()
time.sleep(2) # Let Page Load, If not it will go to Account!
for ids in data:
contest = ids['id']
driver.get('https://www.draftkings.com/contest/exportfullstandingscsv/' + str(contest) + '')

You do not need to send load selenium for x nos of times to download x nos of files. Requests and selenium can share cookies. This means you can login to site with selenium, retrieve the login details and share them with requests or any other application. Take a moment to check out httpie, https://httpie.org/doc#sessions it seems you manually control sessions like requests does.
For requests look at: http://docs.python-requests.org/en/master/user/advanced/?highlight=sessions
For selenium look at: http://selenium-python.readthedocs.io/navigating.html#cookies
Looking at the Webdriver block,you can add proxies and load the browser headless or live: Just comment the headless line and it should load the browser live, this makes debugging easy, easy to understand movements and changes to site api/html.
import time
from selenium import webdriver
from selenium.common.exceptions import WebDriverException
import requests
import datetime
import shutil
LOGIN = 'https://www.draftkings.com/account/sitelogin/false?returnurl=%2Flobby'
BASE_URL = 'https://www.draftkings.com/contest/exportfullstandingscsv/'
USER = ''
PASS = ''
try:
data = requests.get('https://www.draftkings.com/lobby/getlivecontests?sport=NBA').json()
except BaseException as e:
print(e)
exit()
ids = [str(item['id']) for item in data]
# Webdriver block
driver = webdriver.Chrome()
options.add_argument('headless')
options.add_argument('window-size=800x600')
# options.add_argument('--proxy-server= IP:PORT')
# options.add_argument('--user-agent=' + USER_AGENT)
try:
driver.get(URL)
driver.implicitly_wait(2)
except WebDriverException:
exit()
def login(USER, PASS)
'''
Login to draftkings.
Retrieve authentication/authorization.
http://selenium-python.readthedocs.io/waits.html#implicit-waits
http://selenium-python.readthedocs.io/api.html#module-selenium.common.exceptions
'''
search_box = driver.find_element_by_name('username')
search_box.send_keys(USER)
search_box2 = driver.find_element_by_name('password')
search_box2.send_keys(PASS)
submit_button = driver.find_element_by_xpath('//*[#id="react-mobile-home"]/section/section[2]/div[3]/button/span')
submit_button.click()
driver.implicitly_wait(2)
cookies = driver.get_cookies()
return cookies
site_cookies = login(USER, PASS)
def get_csv_files(id):
'''
get each id and download the file.
'''
session = rq.session()
for cookie in site_cookies:
session.cookies.update(cookies)
try:
_data = session.get(BASE_URL + id)
with open(id + '.csv', 'wb') as f:
shutil.copyfileobj(data.raw, f)
except BaseException:
return
map(get_csv_files, ids)

will this help
for ids in data:
contest = ids['id']
driver.get('https://www.draftkings.com/contest/exportfullstandingscsv/' + str(contest) + '')

May be its time to decompose it a bit.
Create few isolated functions, which are:
0. (optional) Provide authorisation to target url.
1. Collecting all needed id (first part of your code).
2. Exporting CSV for specific id (second part of your code).
3. Loop through list of id and call func #2 for each.
Share chromedriver as input argument for each of them to save driver state and auth-cookies.
Its works fine, make code clear and readable.

I think you can set the URL of a contest to an a element in the landing page, and then click on it. Then repeat the step with other ID.
See my code below.
req = requests.get('https://www.draftkings.com/lobby/getlivecontests?sport=NBA')
data = req.json()
contests = []
for ids in data:
contests.append(ids['id'])
driver = webdriver.Chrome() # Optional argument, if not specified will search path.
driver.get('https://www.draftkings.com/account/sitelogin/false?returnurl=%2Flobby');
time.sleep(2) # Let DK Load!
search_box = driver.find_element_by_name('username')
search_box.send_keys('username')
search_box2 = driver.find_element_by_name('password')
search_box2.send_keys('password')
submit_button = driver.find_element_by_xpath('//*[#id="react-mobile-home"]/section/section[2]/div[3]/button/span')
submit_button.click()
time.sleep(2) # Let Page Load, If not it will go to Account!
for id in contests:
element = driver.find_element_by_css_selector('a')
script1 = "arguments[0].setAttribute('download',arguments[1]);"
driver.execute_script(script1, element, str(id) + '.pdf')
script2 = "arguments[0].setAttribute('href',arguments[1]);"
driver.execute_script(script2, element, 'https://www.draftkings.com/contest/exportfullstandingscsv/' + str(id))
time.sleep(1)
element.click()
time.sleep(3)

How to get HTML code after logging in?

I am quite new to Selenium, it would be great if you guys can point me to the right direction.
I'm trying to access the HTML code of a website AFTER the login sequence.
I've used Selenium to direct the browser to initiate the login sequence, the part of the HTML I need will show up after I login. But when I tried to call the HTML code after the login sequence with page_source, it just gave me the HTML code for the site before logging in.
def test_script(ticker):
base_url = "http://amigobulls.com/stocks/%s/income-statement/quarterly" %ticker
driver = webdriver.Firefox()
verificationErrors = []
accept_next_alert = True
driver.get(base_url)
driver.maximize_window()
driver.implicitly_wait(30)
driver.find_element_by_xpath("//header[#id='header_cont']/nav/div[4]/div/span[3]").click()
driver.find_element_by_id("login_email").clear()
driver.find_element_by_id("login_email").send_keys(email)
driver.find_element_by_id("login_pswd").clear()
driver.find_element_by_id("login_pswd").send_keys(pwd)
driver.find_element_by_id("loginbtn").click()
amigo_script = driver.page_source