As a beginner programmer, I have found a lot of useful information on this site, but could not find an answer to my specific question. I want to scrape data from a webpage, but some of the data I am interested in scraping can only be obtained after clicking a "more" button. The below code executes without producing an error, but it does not appear to click the "more" button and display the additional data on the page. I am only interested in viewing the information on the "Transcripts" tab, which seems to complicate things a bit for me because there are "more" buttons on the other tabs. The relevant portion of my code is as follows:
from mechanize import Browser
from selenium import webdriver
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver import ActionChains
import urllib2
import mechanize
import logging
import time
import httplib
import os
import selenium
url="http://seekingalpha.com/symbol/IBM/transcripts"
ua='Mozilla/5.0 (X11; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0 (compatible;)'
br=Browser()
br.addheaders=[('User-Agent', ua), ('Accept', '*/*')]
br.set_debug_http(True)
br.set_debug_responses(True)
logging.getLogger('mechanize').setLevel(logging.DEBUG)
br.set_handle_robots(False)
chromedriver="~/chromedriver"
os.environ["webdriver.chrome.driver"]=chromedriver
driver=webdriver.Chrome(chromedriver)
time.sleep(1)
httplib.HTTPConnection._http_vsn=10
httplib.HTTPConnection._http_vsn_str='HTTP/1.0'
page=br.open(url)
driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")
time.sleep(5)
actions=ActionChains(driver)
elem=driver.find_element_by_css_selector("div #transcripts_show_more div#more.older_archives")
actions.move_to_element(elem).click()
A couple of things:
Given you're using selenium, you don't need either mechanize or urllib2 as selenium is doing the actual page loading. As for the other imports (httplib, logging, os and time), they're either unused or redundant.
For my own convenience, I changed the code to use Firefox; you can change it back to Chrome (or other any browser).
In regards to the ActionChains, you don't them here as you're only doing a single click (nothing to chain really).
Given the browser is receiving data (via AJAX) instead of loading a new page, we don't know when the new data has appeared; so we need to detect the change.
We know that 'clicking' the button loads more <li> tags, so we can check if the number of <li> tags has changed. That's what this line does:
WebDriverWait(selenium_browser, 10).until(lambda driver: len(driver.find_elements_by_xpath("//div[#id='headlines_transcripts']//li")) != old_count)
It will wait up to 10 seconds, periodically comparing the current number of <li> tags from before and during the button click.
import selenium
from selenium import webdriver
from selenium.common.exceptions import StaleElementReferenceException
from selenium.common.exceptions import WebDriverException
from selenium.common.exceptions import TimeoutException as SeleniumTimeoutException
from selenium.webdriver.support.ui import WebDriverWait
url = "http://seekingalpha.com/symbol/IBM/transcripts"
selenium_browser = webdriver.Firefox()
selenium_browser.set_page_load_timeout(30)
selenium_browser.get(url)
selenium_browser.execute_script("window.scrollTo(0, document.body.scrollHeight);")
elem = selenium_browser.find_element_by_css_selector("div #transcripts_show_more div#more.older_archives")
old_count = len(selenium_browser.find_elements_by_xpath("//div[#id='headlines_transcripts']//li"))
elem.click()
try:
WebDriverWait(selenium_browser, 10).until(lambda driver: len(driver.find_elements_by_xpath("//div[#id='headlines_transcripts']//li")) != old_count)
except StaleElementReferenceException:
pass
except SeleniumTimeoutException:
pass
print(selenium_browser.page_source.encode("ascii", "ignore"))
I'm on python2.7; if you're on python3.X, you probably won't need .encode("ascii", "ignore").
Related
I can't figure out how to get around this in Python. The email pop-up is preventing Selenium from clicking on one of the footer links because the pop-up blocks the view of it. I ideally would like to click the "X" and not enter an email.
I've tried using what was in the Selenium documentation about prompts but none of it worked or perhaps I implemented it incorrectly. I tried some of what I already found in stack overflow, which you can see in the commented out code, but kept getting all kinds of errors.
import requests
from urllib.parse import urljoin
from bs4 import BeautifulSoup
from urllib.request import urlopen as uReq
from selenium.webdriver.support.ui import WebDriverWait
from selenium import webdriver
from selenium.webdriver.support import ui
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
url = "https://www.standardmedia.co.ke/"
driver = webdriver.Chrome()
driver.get(url)
html = driver.page_source.encode('utf-8')
page_num = 0
##options = Firefox_options();
##options.addPreference("dom.disable_beforeunload", true)
##driver = webdriver.Firefox(options);
#click on the headings in the footer
for i in range (0,1):
footer = driver.find_elements_by_css_selector('#top-footer li')[i]
## if(driver.switch_to.alert != null):
## alert = driver.switch_to.alert
## alert.dismiss()
try:
WebDriverWait(driver, 10).until(EC.alert_is_present())
alert = driver.switch_to_alert()
alert.dismiss()
print("Alert dismissed.")
except TimeoutException:
print("No alert.")
footer.click()
print("alert dismissed")
page_num += 1
subheadings = driver.find_elements_by_css_selector('.title a')
len(subheadings)
The most recent error for a Firefox web driver was "No connection could be made because the target machine actively refused it."
WebDriver Alert class is designed to work with JavaScript Alerts to wit:
Window.alert()
Window.confirm()
Window.prompt()
At the page you're trying to test this modal popup is a normal <div>
therefore you will not be able to use Alert class, you will have to normally locate close button using find_element() function and click it.
Open the page:
driver.get("https://www.standardmedia.co.ke/")
Wait for the popup to appear
popup = WebDriverWait(driver, 10).until(expected_conditions.presence_of_element_located((By.CLASS_NAME, "mc-closeModal")))
Click the close button
popup.click()
Aside from the selenium connection issue, to address the main question, it looks like you are not calling the switch_to_alert function. Once you get selenium connecting, try the following:
alert = driver.switch_to_alert()
alert.dismiss()
Also note, your wait times are in seconds which seem pretty high/long in your code.
I'm trying to get the xpath of an element in site https://www.tradingview.com/symbols/BTCUSD/technicals/
Specifically the result under the summary speedometer. Whether it's buy or sell.
Speedometer
Using Google Chrome xpath I get the result
//*[#id="technicals-root"]/div/div/div[2]/div[2]/span[2]
and to try and get that data in python I plugged it into
from lxml import html
import requests
page = requests.get('https://www.tradingview.com/symbols/BTCUSD/technicals/')
tree = html.fromstring(page.content)
status = tree.xpath('//*[#id="technicals-root"]/div/div/div[2]/div[2]/span[2]/text()')
When I print status I get an empty array. But it doesn't seem like anything is wrong with the xpath. I've read that google does some shenanigans with incorrectly written HTML tables which will output the wrong xpath but that doesn't seem to be the issue.
When I run your code, the "technicals-root" div is empty. I assume javascript is filling it in. When you can't get a page statically, you can always turn to Selenium to run a browser and let it figure everything out. You may have to tweak the driver path to get it working in your environment but this works for me:
import time
import contextlib
import selenium
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
from selenium.common.exceptions import TimeoutException
option = webdriver.ChromeOptions()
option.add_argument(" — incognito")
with contextlib.closing(webdriver.Chrome(
executable_path='/usr/lib/chromium-browser/chromedriver',
chrome_options=option)) as browser:
browser.get('https://www.tradingview.com/symbols/BTCUSD/technicals/')
# wait until js has filled in the element - and a bit longer for js churn
WebDriverWait(browser, 20).until(EC.visibility_of_element_located(
(By.XPATH,
'//*[#id="technicals-root"]/div/div/div[2]/div[2]/span')))
time.sleep(1)
status = browser.find_elements_by_xpath(
'//*[#id="technicals-root"]/div/div/div[2]/div[2]/span[2]')
print(status[0].text)
I am trying to scrape a website and want to get the url's and images from Google AdSense. But it seems I am not getting any details of Google Adsense.
Here I want
If we search "refrigerator" in google then we will get some ads there which I need to fetch. Or some blogs, website showing Google Ads like See image
But when I inspect I can find related divs and url but when I hit url then i am getting only static html data.
Here is code which I need to fetch
Here is script which I have written in Selenium, Python.
from contextlib import closing
from selenium.webdriver import Firefox # pip install selenium
import time
from selenium.common.exceptions import TimeoutException
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
url = "http://www.compiletimeerror.com/"
# use firefox to get page with javascript generated content
with closing(Firefox()) as browser:
browser.get(url) # load page
delay = 10 # seconds
try:
WebDriverWait(browser, delay).until(EC.presence_of_element_located(browser.find_element_by_xpath("(//div[#class='pla-unit'])[0]")))
print "Page is ready!"
Element=browser.find_element(By.ID,value="google_image_div")
print Element
print Element.text
except TimeoutException:
print "Loading took too much time!"
But I'm still unable to get data. Please give me any reference or hint.
You need to first select the frame which contains the elements you want to work with.
select_frame("id=google_ads_frame1");
NOTE: I am not sure about the python syntax. But it should be something similar to this.
Use Selenium's switch_to.frame method to direct your browser to the iframe in your html, before selecting your element variable (untested):
from contextlib import closing
from selenium.webdriver import Firefox # pip install selenium
import time
from selenium.common.exceptions import TimeoutException
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
url = "http://www.compiletimeerror.com/"
# use firefox to get page with javascript generated content
with closing(Firefox()) as browser:
browser.get(url) # load page
delay = 10 # seconds
try:
WebDriverWait(browser, delay).until(EC.presence_of_element_located(browser.find_element_by_xpath("(//div[#class='pla-unit'])[0]")))
print "Page is ready!"
browser.switch_to.frame(browser.find_element_by_id('google_ads_frame1'))
element=browser.find_element(By.ID,value="google_image_div")
print element
print element.text
except TimeoutException:
print "Loading took too much time!"
http://elementalselenium.com/tips/3-work-with-frames
A note on Python style best practices: use lowercase when declaring local variables (element vs. Element).
I have written many scrapers but I am not really sure how to handle infinite scrollers. These days most website etc, Facebook, Pinterest has infinite scrollers.
You can use selenium to scrap the infinite scrolling website like twitter or facebook.
Step 1 : Install Selenium using pip
pip install selenium
Step 2 : use the code below to automate infinite scroll and extract the source code
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support.ui import Select
from selenium.webdriver.support.ui import WebDriverWait
from selenium.common.exceptions import TimeoutException
from selenium.webdriver.support import expected_conditions as EC
from selenium.common.exceptions import NoSuchElementException
from selenium.common.exceptions import NoAlertPresentException
import sys
import unittest, time, re
class Sel(unittest.TestCase):
def setUp(self):
self.driver = webdriver.Firefox()
self.driver.implicitly_wait(30)
self.base_url = "https://twitter.com"
self.verificationErrors = []
self.accept_next_alert = True
def test_sel(self):
driver = self.driver
delay = 3
driver.get(self.base_url + "/search?q=stckoverflow&src=typd")
driver.find_element_by_link_text("All").click()
for i in range(1,100):
self.driver.execute_script("window.scrollTo(0, document.body.scrollHeight);")
time.sleep(4)
html_source = driver.page_source
data = html_source.encode('utf-8')
if __name__ == "__main__":
unittest.main()
Step 3 : Print the data if required.
Most sites that have infinite scrolling do (as Lattyware notes) have a proper API as well, and you will likely be better served by using this rather than scraping.
But if you must scrape...
Such sites are using JavaScript to request additional content from the site when you reach the bottom of the page. All you need to do is figure out the URL of that additional content and you can retrieve it. Figuring out the required URL can be done by inspecting the script, by using the Firefox Web console, or by using a debug proxy.
For example, open the Firefox Web Console, turn off all the filter buttons except Net, and load the site you wish to scrape. You'll see all the files as they are loaded. Scroll the page while watching the Web Console and you'll see the URLs being used for the additional requests. Then you can request that URL yourself and see what format the data is in (probably JSON) and get it into your Python script.
Finding the url of the ajax source will be the best option but it can be cumbersome for certain sites. Alternatively you could use a headless browser like QWebKit from PyQt and send keyboard events while reading the data from the DOM tree. QWebKit has a nice and simple api.
I have these includes:
from selenium import webdriver
from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver.common.keys import Keys
Browser set up via
browser = webdriver.Firefox()
browser.get(loginURL)
However sometimes I do
browser.switch_to_frame("nameofframe")
And it won't work (sometimes it does, sometimes it doesn't).
I am not sure if this is because Selenium isn't actually waiting for pages to load before it executes the rest of the code or what. Is there a way to force a webpage load?
Because sometimes I'll do something like
browser.find_element_by_name("txtPassword").send_keys(password + Keys.RETURN)
#sends login information, goes to next page and clicks on Relevant Link Text
browser.find_element_by_partial_link_text("Relevant Link Text").click()
And it'll work great most of the time, but sometimes I'll get an error where it can't find "Relevant Link Text" because it can't "see" it or some other such thing.
Also, is there a better way to check if an element exists or not? That is, what is the best way to handle:
browser.find_element_by_id("something")
When that element may or may not exist?
You could use WebDriverWait:
from contextlib import closing
from selenium.webdriver import Chrome as Browser
from selenium.webdriver.support.ui import WebDriverWait
from selenium.common.exceptions import NoSuchFrameException
def frame_available_cb(frame_reference):
"""Return a callback that checks whether the frame is available."""
def callback(browser):
try:
browser.switch_to_frame(frame_reference)
except NoSuchFrameException:
return False
else:
return True
return callback
with closing(Browser()) as browser:
browser.get(url)
# wait for frame
WebDriverWait(browser, timeout=10).until(frame_available_cb("frame name"))