Good morning!
I have recently started learning Python and moved onto applying the little I know to create the monstrosity seen below.
In brief:
I am attempting to scrape SEC Edgar (https://www.sec.gov/edgar/searchedgar/cik.htm) for CIK codes of companies which I want to study in more detail (for now just 1 company to see if it's the right approach).
To scrape the CIK-code, I created a scrapy spider, imported selenium and created 3 functions - 1st to insert "company name" in the input name, 2nd to activate the "submit" button and finally, a function to scrape the CIK code once the submit is activated and return item.
Apart from adding the item to items.py, I haven't changed the middlewares or settings.
For some reason, I am getting ModuleNotFoundError for 'selenium', although I have installed the packages and imported selenium & webdriver along with everything else.
I have tried to mess around with indentation and rephrased the code but achieved no improvement.
import selenium
from selenium import webdriver
import scrapy
from ..items import Sec1Item
from scrapy import Selector
class SecSpSpider(scrapy.Spider):
name = 'SEC_sp'
start_urls =
['http://https://www.sec.gov/edgar/searchedgar/cik.htm/']
def parse(self,response):
company_name = 'INOGEN INC'
return scrapy.FormRequest.from_response(response, formdata ={
'company': company_name
}, callback=self.start_requests())
def start_requests(self):
driver = webdriver.Chrome()
driver.get(self.start_urls)
while True:
next_url = driver.find_element_by_css_selector(
'.search-button'
)
try:
self.parse(driver.page_source)
next_url.click()
except:
break
driver.close()
def parse_page(self, response):
items = Sec1Item()
CIK_code = response.css('a::text').extract()
items["CIK Code: "] = Sec1Item
yield items
I seem not to be able to get over the import selenium error, hence I am not sure about the extent of needed adjustments to the remainder of my spider.
Error message:
"File/Users/user1/PycharmProjects/Scraper/SEC_1/SEC_1/spiders/SEC_sp.py", line 1, in <module>
import selenium
ModuleNotFoundError: No module named 'selenium'
Thank you for any assistance and help!
Related
I made a dataset of linkedin links for data science jobs (with 1000 rows) and made a for-loop to open each link by selenium and extract the job description by beautifulsoup.
It had worked until 121 rows extracted and then it displayed an error and stopped. when I wanted to start again from row 122, it displays this error from link
the link (name of the dataset column), has problem
and does not open linkedin. I tested and see selenium can open google for example.
my loop is:
# keep the loop from 123
from selenium import webdriver
from bs4 import BeautifulSoup
import time
driver = webdriver.Chrome(r"C:\Users\shahi\chromedriver.exe")
jobdescriptions2=[]
for link in links[121:1000] :
try:
# This instance will be used to log into LinkedIn
driver.get(link)
time.sleep(5)
src = driver.page_source
except:
scr= 'problem in scraping data'
print(f'the link {link}, has problem')
try:
# Now using beautiful soup
soup = BeautifulSoup(src, 'lxml')
# Extracting the HTML of the complete introduction box
job = soup.find("div", {'class': 'decorated-job-posting__details'})
jobdescription = job.find ('div', {'class':'show-more-less-html__markup'})
except:
jobdescription= 'an error in data parcing'
jobdescriptions2.append(jobdescription)
Could you please advise me how I can scrape all 1000 rows?
I am scraping a website for most recent customer rating, with several pages.
The problem is that I am able to interact with the "sortby" option and select "most recent" using Selenium, and scrape data for first page using Scrapy. However, I am unable to extract the data for other pages, the Selenium Web driver somehow does not render the next page. My intension is to automate data scraping.
I am a newb to web scraping. A snippet of the code is attached here (Some information is removed due to confidentiality)
import scrapy
import selenium.webdriver as webdriver
from selenium.webdriver.edge.service import Service
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait,Select
import time
from selenium.webdriver.support import expected_conditions as EC
from scrapy import Selector
from selenium.webdriver.edge.options import Options
from scrapy.utils.project import get_project_settings
class ABC(scrapy.Spider):
#"........."
def start_requests(self):
#" ...... "
yield scrapy.Request(url)
def parse(self, response):
settings =get_project_settings()
driver_path = settings.get('EDGE_DRIVER_PATH')
options = Options()
options.add_argument("headless")
ser=Service(driver_path)
driver = webdriver.Edge(service=ser,options = options)
driver.get(response.url)
WebDriverWait(driver, 10).until(EC.element_to_be_clickable((By.ID,"sort-order-dropdown")))
element_dropdown=driver.find_element(By.ID,"sort-order-dropdown")
select=Select(element_dropdown)
select.select_by_value("recent")
time.sleep(5)
for review in response.css('[data-hook="review"]':
res={
"rating": review.css('[class="a-icon-alt"]::text').get(),
}
yield res
next_page =response.xpath('//a[text()="Next page"]/#href').get()
if next_page:
yield scrapy.Request(response.urljoin(next_page))
driver.quit()
Looks that you're using Scrapy and Selenium instead of scrapy_selenium (I don't see any SeleniumRequest in your code.
Your current spider works like this:
Get page using Scrapy
Get the same page using Selenium webdriver
Perform some actions using Selenium
Parse Scrapy response (for rating and next_page)
As you see you never use / parse Selenium result.
I have made code, see following lines:
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
import time
from selenium.webdriver.common.by import By
PATH = "C:\Program Files (x86)\chromedriver.exe"
driver = webdriver.Chrome(PATH)
driver.get("https://www.flashscore.com/match/jBvNMej6/#match-summary")
print(driver.title)
driver.maximize_window() # For maximizing window
driver.implicitly_wait(10) # gives an implicit wait for 20 seconds
driver.find_element_by_id('onetrust-reject-all-handler').click()
time.sleep(2)
driver.find_element(By.CLASS_NAME,'previewShowMore.showMore').click()
main = driver.find_element(By.CLASS_NAME,'previewLine'[b[text()="Hot stat:"]]/text)
print(main.text)
time.sleep(2)
driver.close()
However, I get the following error.
main = driver.find_element(By.CLASS_NAME,'previewLine'[b[text()="Hot stat:"]]/text)
^
SyntaxError: invalid syntax
What can I do to avoid this?
thx! : )
Well, in this line
main = driver.find_element(By.CLASS_NAME,'previewLine'[b[text()="Hot stat:"]]/text)
You have made a great mix :)
Your locator is absolutely invalid.
Also, if you want to print the paragraph text without the "Hot streak" you will need to remove that string from the entire div (paragraph) text.
This should do what you are trying to achieve:
main = driver.find_element(By.XPATH,"//div[#class='previewLine' and ./b[text()='Hot streak']]").text
main = main.replace('Hot streak','')
print(main)
I'm not finding any text 'Hot stat:'. You'll have to attach the html code where you found that.
I assume that you want to retrieve the text of a specific previewLine?
main = driver.find_element(By.XPATH ,'//div[#class="previewLine"]/b[contains(text(),"Hot streak")]/..')
print(main.text)
This might be the stupidest question i asked yet but this is driving me nuts...
Basically i want to get all links from profiles but for some reason selenium gives different amounts of links most of the time ( sometimes all sometimes only a tenth)
I experimented with time.sleep and i know its affecting the output somehow but i dont understand where the problem is.
(but thats just my hypothesis maybe thats wrong)
I have no other explanation why i get incosistent output. Since i get all profile links from time to time the program is able to find all relevant profiles.
heres what the output should be (for different gui input)
input:anlagenbau output:3070
Fahrzeugbau output:4065
laserschneiden output:1311
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
from selenium.common.exceptions import TimeoutException
from urllib.request import urlopen
from datetime import date
from datetime import datetime
import easygui
import re
from selenium.common.exceptions import NoSuchElementException
import time
#input window suchbegriff
suchbegriff = easygui.enterbox("Suchbegriff eingeben | Hinweis: suchbegriff sollte kein '/' enthalten")
#get date and time
now = datetime.now()
current_time = now.strftime("%H-%M-%S")
today = date.today()
date = today.strftime("%Y-%m-%d")
def get_profile_url(label_element):
# get the url from a result element
onlick = label_element.get_attribute("onclick")
# some regex magic
return re.search(r"(?<=open\(\')(.*?)(?=\')", onlick).group()
def load_more_results():
# load more results if needed // use only on the search page!
button_wrapper = wd.find_element_by_class_name("loadNextBtn")
button_wrapper.find_element_by_tag_name("span").click()
#### Script starts here ####
# Set some Selenium Options
options = webdriver.ChromeOptions()
options.add_argument("--no-sandbox")
options.add_argument("--disable-dev-shm-usage")
# Webdriver
wd = webdriver.Chrome(options=options)
# Load URL
wd.get("https://www.techpilot.de/zulieferer-suchen?"+str(suchbegriff))
# lets first wait for the timeframe
iframe = WebDriverWait(wd, 5).until(
EC.frame_to_be_available_and_switch_to_it("efficientSearchIframe")
)
# the result parent
result_pane = WebDriverWait(wd, 5).until(
EC.presence_of_element_located((By.ID, "resultPane"))
)
#get all profilelinks as list
time.sleep(5)
href_list = []
wait = WebDriverWait(wd, 15)
while True:
try:
#time.sleep(1)
wd.execute_script("loadFollowing();")
#time.sleep(1)
try:
wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR, ".fancyCompLabel")))
except TimeoutException:
break
#time.sleep(1) # beeinflusst in irgeneiner weise die findung der ergebnisse
result_elements = wd.find_elements_by_class_name("fancyCompLabel")
#time.sleep(1)
for element in result_elements:
url = get_profile_url(element)
href_list.append(url)
#time.sleep(2)
while True:
try:
element = wd.find_element_by_class_name('fancyNewProfile')
wd.execute_script("""var element = arguments[0];element.parentNode.removeChild(element);""", element)
except NoSuchElementException:
break
except NoSuchElementException:
break
wd.close #funktioniert noch nicht
print("####links secured: "+str(len(href_list)))
Since you say that the sleep is affecting the number of results, it sounds like they're loading asynchronously and populating as they're loaded, instead of all at once.
The first question is whether you can ask the web site developers to change this, to only show them when they're all loaded at once.
Assuming you don't work for the same company as them, consider:
Is there something else on the page that shows up when they're all loaded? It could be a button or a status message, for instance. Can you wait for that item to appear, and then get the list?
How frequently do new items appear? You could poll for the number of results relatively infrequently, such as only every 2 or 3 seconds, and then consider the results all present when you get the same number of results twice in a row.
The issue is the method presence_of_all_elements_located doesn't wait for all elements matching a passed locator. It waits for presence of at least 1 element matching the passed locator and then returns a list of elements found on the page at that moment matching that locator.
In Java we have
wait.until(ExpectedConditions.numberOfElementsToBeMoreThan(element, expectedElementsAmount));
and
wait.until(ExpectedConditions.numberOfElementsToBe(element, expectedElementsAmount));
With these methods you can wait for predefined amount of elements to appear etc.
Selenium with Python doesn't support these methods.
The only thing you can see with Selenium in Python is to build some custom method to do these actions.
So if you are expecting some amount of elements /links etc. to appear / be presented on the page you can use such method.
This will make your test stable and will avoid usage of hardcoded sleeps.
UPD
I have found this solution.
This looks to be the solution for the mentioned above methods.
This seems to be a Python equivalent for wait.until(ExpectedConditions.numberOfElementsToBeMoreThan(element, expectedElementsAmount));
myLength = 9
WebDriverWait(browser, 20).until(lambda browser: len(browser.find_elements_by_xpath("//img[#data-blabla]")) > int(myLength))
And this
myLength = 10
WebDriverWait(browser, 20).until(lambda browser: len(browser.find_elements_by_xpath("//img[#data-blabla]")) == int(myLength))
Is equivalent for Java wait.until(ExpectedConditions.numberOfElementsToBe(element, expectedElementsAmount));
With the Selenium (3.141), BeautifulSoup (4.7.9), and Python (3.79), I'm trying to scrape what streaming, rental, and buying options are available for a given movie/show. I've spent hours trying to solve this, so any help would be appreciated. Apologies for the poor formatting, in terms of mixing in comments and prior attempts.
Example Link: https://www.justwatch.com/us/tv-show/24
Desired Outcome is a Beautiful soup element that I can then parse (e.g., which streaming services have it, how many seasons are available, etc.),
which has 3 elements (as of now) - Hulu, IMDB TV, and DirecTV.
I tried numerous variations, but only get one of the 3 streaming services for the example link, and even then it's not a consistent result. Often, I get an empty object.
Some of the things that I've tried included waiting for an expected condition (presence or visibility), explicitly using sleep() from the time library. I'm using a Mac (but running Linux via a USB), so there is no "PAGE DOWN" on the physical keyboard. For the keys module, I've tried control+arrow down, page down, and and space (space bar), but on this particular web page they don't work. However, if I'm browsing it in a normal fashion, control+arrow down and space bar help scrolling the desired section into view. As far as I know, there is no fn + arrow down option that works in Keys, but that's another way that I can move in a normal fashion.
I've run both headless and regular options to try to debug, as well as trying both Firefox and Chrome drivers.
Here's my code:
import time
from selenium import webdriver
from selenium.webdriver.firefox.options import Options
from selenium.common.exceptions import NoSuchElementException
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
from bs4 import BeautifulSoup
firefox_options = Options()
firefox_options.add_argument('--enable-javascript') # double-checking to make sure that javascript is enabled
firefox_options.add_argument('--headless')
firefox_driver_path = 'geckodriver'
driver = webdriver.Firefox(executable_path=firefox_driver_path, options=firefox_options)
url_link = 'https://www.justwatch.com/us/tv-show/24'
driver.get(url_link) # initial page
cookies = driver.get_cookies()
Examples of things I've tried around this part of the code
various time.sleep(3) and driver.implicitly_wait(3) commands
webdriver.ActionChains(driver).key_down(Keys.CONTROL).key_down(Keys.ARROW_DOWN).perform()
webdriver.ActionChains(driver).key_down(Keys.SPACE).perform()
This code yields a timeout error when used
stream_results = WebDriverWait(driver, 15)
stream_results.until(EC.presence_of_element_located(
(By.CLASS_NAME, "price-comparison__grid__row price-comparison__grid__row--stream")))
page_source = driver.page_source
soup = BeautifulSoup(page_source, 'html.parser') # 'lxml' didn't work either
Here's code for getting the html related to the streaming services. I've also tried to grab the html code at various levels, ids, and classes of the tree, but the code just isn't there
stream_row = soup.find('div', attrs={'class':'price-comparison__grid__row price-comparison__grid__row--stream'})
stream_row_holder = soup.find('div', attrs={'class':'price-comparison__grid__row__holder'})
stream_items = stream_row_holder\
.find_all('div', attrs={'class':'price-comparison__grid__row__element__icon'})
driver.quit()
I'm not sure if you are saying your code works in some cases or not at all, but I use chrome and the four find_all() lines at the end all produce results. If this isn't what you mean, let me know. The one thing you may be missing is a time.sleep() that is long enough. That could be the only difference...
Note you need chromedriver to run this code, but perhaps you have chrome and can download chromedriver.exe.
import time
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
chrome_options = Options()
chrome_options.add_argument("--headless")
url = 'https://www.justwatch.com/us/tv-show/24'
driver = webdriver.Chrome('chromedriver.exe', options=chrome_options)
driver.get(url)
time.sleep(5)
html = driver.page_source
soup = BeautifulSoup(html, "html.parser")
soup.find_all(class_="price-comparison__grid__row__price")
soup.find_all(class_="price-comparison__grid__row__holder")
soup.find_all(class_="price-comparison__grid__row__element__icon")
soup.find_all(class_="price-comparison__grid__row--stream")
This is the output from the last line:
[<div class="price-comparison__grid__row price-comparison__grid__row--stream"><div class="price-comparison__grid__row__title price-comparison__promoted__title"> Stream </div><div class="price-comparison__grid__row__holder"><!-- --><div class="price-comparison__grid__row__element"><div class="presentation-type price-comparison__grid__row__element__icon"><img alt="Hulu" class="jw-provider-icon price-comparison__grid__row__icon" src="https://images.justwatch.com/icon/116305230/s100" title="Hulu"/><div class="price-comparison__grid__row__price"> 9 Seasons <span class="price-comparison__badge price-comparison__badge--hd price-comparison__badge--hd"> HD </span></div></div></div><!-- --></div></div>,
<div class="price-comparison__grid__row price-comparison__grid__row--stream"><div class="price-comparison__grid__row__title"> Stream </div><div class="price-comparison__grid__row__holder"><!-- --><div class="price-comparison__grid__row__element"><div class="presentation-type price-comparison__grid__row__element__icon"><img alt="Hulu" class="jw-provider-icon price-comparison__grid__row__icon" src="https://images.justwatch.com/icon/116305230/s100" title="Hulu"/><div class="price-comparison__grid__row__price"> 9 Seasons <span class="price-comparison__badge price-comparison__badge--hd price-comparison__badge--hd"> HD </span></div></div></div><div class="price-comparison__grid__row__element"><div class="presentation-type price-comparison__grid__row__element__icon"><img alt="IMDb TV" class="jw-provider-icon price-comparison__grid__row__icon" src="https://images.justwatch.com/icon/134049674/s100" title="IMDb TV"/><div class="price-comparison__grid__row__price"> 8 Seasons <!-- --></div></div></div><div class="price-comparison__grid__row__element"><div class="presentation-type price-comparison__grid__row__element__icon"><img alt="DIRECTV" class="jw-provider-icon price-comparison__grid__row__icon" src="https://images.justwatch.com/icon/158260222/s100" title="DIRECTV"/><div class="price-comparison__grid__row__price"> 1 Season <span class="price-comparison__badge price-comparison__badge--hd price-comparison__badge--hd"> HD </span></div></div></div><!-- --></div></div>]