We Keep Coding

web scraping all universities with websites and description WHED website

anyone can help with scraping from https://www.whed.net/home.php
the code I'm using is giving me empty df. would love to have universities with websites and maybe field of study. My scraping skills are weak so if you can guide me through this would be great thanks guys.
begin=time.time()
countries=['Emirates','United States of America (all)']
result = [] # List to store all data
univ_links=[] # Links for all universities
fields = ['Street:','City:','Province:','Post Code:','WWW:','Fields of study:','Job title:']
webD = wb.Chrome(executable_path=r'C:\Users\Admin\OneDrive\Sagasit\chromedriver.exe') # To launch chrome and run script
# Trigger the target website
webD.get("https://www.whed.net/results_institutions.php")
webD.implicitly_wait(5)
#all_countries=[]
cntry_el = webD.find_elements_by_xpath('//*[#id="Chp1"]/option')
#cntry_grp = webD.find_elements_by_xpath('//*[#id="Chp1"]/optgroup')
grps=webD.find_elements_by_xpath('//*[#id="Chp1"]/optgroup/option[1]')
for c in cntry_el:countries.append(c.text)
for g in grps: countries.append(g.text)
for cntry in countries:
select = Select(webD.find_element_by_id('Chp1'))#select country dropdown
select.select_by_visible_text(cntry)#choosing country
Btn_GO = webD.find_element_by_xpath('//*[#id="fsearch"]/p/input')
Btn_GO.click()
select_rpp = Select(webD.find_element_by_name('nbr_ref_pge'))#select results per page drop down
select_rpp.select_by_visible_text('100')#choosing 100 results per page option
university_form = webD.find_element_by_xpath('//*[#id="contenu"]').find_element_by_id('results')
university_list = university_form.find_elements_by_xpath('//*[#id="results"]/li') # list of university elements
for univ in range(len(university_list)):
href = university_list[univ].find_element_by_class_name('details').find_elements_by_tag_name('a')[0].get_property('href') # University details link
univ_links.append(href)
while True:
try:
webD.find_element_by_partial_link_text('Next').click()
university_form = webD.find_element_by_xpath('//*[#id="contenu"]').find_element_by_id('results')
university_list = university_form.find_elements_by_xpath('//*[#id="results"]/li')
for univ in range(len(university_list)):
href = university_list[univ].find_element_by_class_name('details').find_elements_by_tag_name('a')[0].get_property('href') # University details link
univ_links.append(href)
except NoSuchElementException: break
for l in univ_links:
webD.get(l)
webD.implicitly_wait(2)
title=webD.find_element_by_xpath('//*[#id="page"]/div/div/div[2]/div[1]').text
title_detailed = webD.find_element_by_xpath('//*[#id="page"]/div/div/div[2]/div[2]').text
cntry_name=webD.find_element_by_xpath('//*[#id="contenu"]/p[2]').text
t1=webD.find_elements_by_class_name('dt')
t2=webD.find_elements_by_class_name('dd')
labels=webD.find_elements_by_class_name('libelle')
content=webD.find_elements_by_class_name('contenu')
temp={}
fos=''
fos1=''
temp.update({'Title': title,'Detailed Title':title_detailed,'Country':cntry_name})
for i in range(len(t1)):
if t1[i].text == '' or t1[i].text == 'Address':
continue
else:
value=t2[i].text
temp.update({t1[i].text:value.replace('\n',',')})
for j in range(len(content)):
if labels[j].text in fields:
if labels[j].text == 'Fields of study:':
info=content[j].text
fos=fos+','+info
elif labels[j].text == 'Job title:':
info1=content[j].text
fos1=fos1+','+info1
else:
key=labels[j].text
temp.update({key[:-1]: content[j].text})
temp.update({'Fields of study': fos.lstrip(','),'Job titles':fos1.lstrip(',')})
result.append(temp)
data=pd.DataFrame(result)
data
end=time.time()
print("Time taken : "+ str(end-begin) +"s")
data.to_csv("WHED1.csv",index=False)
this code what i could use taken from github project.
would be great if i can re-create the data and save it, want this to be used as a dropdown in a web application just to make sure no mistakes written in the university studied in.

Update 1/12/22 - Async
Found a much better solution using aiohttp, it also runs the entire list of countries in ~30 seconds instead of 3 hours
import json
import time
import aiohttp
import asyncio
from bs4 import BeautifulSoup
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.remote.webelement import WebElement
from selenium.webdriver.support.select import Select
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.chrome.service import Service
def main():
print("Init")
driver = init_driver()
print("Opening Homepage")
url = "https://www.whed.net/results_institutions.php"
driver.get(url)
time.sleep(1)
print("Gathering Countries")
countries = get_countries(driver)
driver.quit()
print("Scraping")
start = time.time()
institution_list = asyncio.run(fetch_all(countries))
print("Writing out")
f = open('output.json', 'w')
f.write(json.dumps(institution_list))
f.close()
end = time.time()
print(f"Total time: {end - start}s")
def init_driver():
chrome_executable = Service(executable_path='chromedriver.exe', log_path='NUL')
chrome_options = Options()
chrome_options.add_argument("--headless")
driver = webdriver.Chrome(service=chrome_executable, options=chrome_options)
return driver
def get_countries(driver):
select = Select(driver.find_element(By.ID, "Chp1"))
countries = list(map(lambda c: c.get_attribute('value'), select.options))
countries.pop(0)
return countries
def extract_institutions(html, country):
soup = BeautifulSoup(html, 'html.parser')
page = soup.find('p', {'class': 'infos'}).text
print(str(page))
number_of_institutions = str(page).split()[0]
if number_of_institutions == 'No':
print(f"No results for {country}")
return []
results = []
inst_index = 0
raw = soup.find_all('a', {'class': 'fancybox fancybox.iframe'})
for i in raw:
results.append({
'name': str(i.text).strip(),
'url': 'https://www.whed.net/' + str(i.attrs['href']).strip(),
'country': country
})
inst_index += 1
return {
'country': country,
'count': number_of_institutions,
'records': results
}
async def get_institutions(country, session):
try:
async with session.post(
url='https://www.whed.net/results_institutions.php',
data={"Chp1": country, "nbr_ref_pge": 10000}
) as response:
html = await response.read()
print(f"Successfully got {country}")
return extract_institutions(html, country)
except Exception as e:
print(f"Unable to get {country} due to {e.__class__}.")
async def fetch_all(countries):
async with aiohttp.ClientSession() as session:
return await asyncio.gather(*[get_institutions(country, session) for country in countries])
# Main call
main()
Old answer using synchronous algorithm
Improving on #Mithun's answer since it doesn't really work as it'll be stuck on the same page.
Also added direct access to the name and url to make it easier in case you want to access those.
import time
from bs4 import BeautifulSoup
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.select import Select
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.chrome.service import Service
print("Init")
chrome_executable = Service(executable_path='chromedriver.exe', log_path='NUL')
chrome_options = Options()
chrome_options.add_argument("--headless")
driver = webdriver.Chrome(service=chrome_executable, options=chrome_options)
print("Opening Homepage")
url = "https://www.whed.net/results_institutions.php"
driver.get(url)
time.sleep(1)
print("Selecting country")
select = Select(driver.find_element(By.ID, "Chp1"))
country = "Albania"
select.select_by_visible_text(country)
time.sleep(.5)
print("Searching")
driver.find_element(By.XPATH, "//input[#value='Go']").click()
time.sleep(1)
print("Parsing")
html = driver.page_source
soup = BeautifulSoup(html, 'html.parser')
page = soup.find('p', {'class': 'infos'}).text
number_of_pages = str(page).split()[0]
counter = 10
results = []
while True:
raw = soup.find_all('a', {'class': 'fancybox fancybox.iframe'})
for i in raw:
results.append({
'name': str(i.text).strip(),
'url': 'https://www.whed.net/' + str(i.attrs['href']).strip(),
'country': country
})
print(f'{len(results)}/{number_of_pages}')
if counter >= int(number_of_pages):
break
counter += 10
driver.find_element(By.LINK_TEXT, "Next page").click()
time.sleep(0.5)
soup = BeautifulSoup(driver.page_source, 'html.parser')
driver.quit()
print(results)

You can use Selenium to scrape data. The following code will help you scrape the university names for "United States of America (all)". Similarly, you can scrape for other countries as well using Loop or entering the name manually. If you need the field of study for every university, you can scrape its href using bs4 and its field of study.
from bs4 import BeautifulSoup
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.select import Select
driver = webdriver.Chrome(r"chromedriver.exe")
url = "https://www.whed.net/results_institutions.php"
driver.get(url)
time.sleep(1)
select = Select(driver.find_element(By.ID, "Chp1"))
select.select_by_visible_text("United States of America (all)")
time.sleep(1)
driver.find_element(By.XPATH, "//input[#value='Go']").click()
time.sleep(1)
html = driver.page_source
soup = BeautifulSoup(html, 'html.parser')
page = soup.find('p', {'class': 'infos'}).text
number_of_pages = str(page).split()[0]
counter = 10
while counter < int(number_of_pages):
raw = soup.find_all('div', {'class': 'details'})
for i in raw:
i = (str(i.text).lstrip())
i = i.replace("\n","")
i = i.replace("\r", "")
i = i.replace("\t", "")
print(i)
next_page = driver.find_element(By.LINK_TEXT, "Next page").click()
counter += 10
driver.quit()

Python scraping pages after hitting the "load more results" button

How do you scrape a web page with infinite scrolling?
My first try was using Selenium, but it detects as robot:
from selenium import webdriver
import time
import pandas as pd
url = 'https://www.bloomberg.com/search?query=indonesia%20mining'
options = webdriver.ChromeOptions()
options.add_argument("start-maximized")
options.add_argument("disable-infobars")
options.add_argument("--disable-extensions")
driver = webdriver.Chrome(options=options)
driver.get(url)
html = driver.page_source.encode('utf-8')
page_num = 0
while driver.find_elements_by_css_selector('.contentWell__a8d28605a5'):
driver.find_element_by_css_selector('.contentWell__a8d28605a5').click()
page_num += 1
print("getting page number "+str(page_num))
time.sleep(1)
html = driver.page_source.encode('utf-8')
soup = BeautifulSoup(html, 'lxml')
titles = soup.find_all('div', {"class":"text__d88756958e withThumbnail__c4ffc902a6"})
df = pd.DataFrame(columns=['judul', 'link'])
news = {}
for t in titles:
news['judul'] = t.find('a', {'class':'headline__96ba1917df'}).text.strip()
news['link'] = t.find('a', {'class':'headline__96ba1917df'}).get('href')
df = df.append(news, ignore_index=True)
any idea how to limit the maximum page number?

Python - Selenium next page

I am trying to make a scraping application to scrape Hants.gov.uk and right now I am working on it just clicking the pages instead of scraping. When it gets to the last row on page 1 it just stopped, so what I did was make it click button "Next Page" but first it has to go back to the original URL. It clicks page 2, but after page 2 is scraped it doesn't go to page 3, it just restarts page 2.
Can somebody help me fix this issue?
Code:
import time
import config # Don't worry about this. This is an external file to make a DB
import urllib.request
from bs4 import BeautifulSoup
from selenium import webdriver
url = "https://planning.hants.gov.uk/SearchResults.aspx?RecentDecisions=True"
driver = webdriver.Chrome(executable_path=r"C:\Users\Goten\Desktop\chromedriver.exe")
driver.get(url)
driver.find_element_by_id("mainContentPlaceHolder_btnAccept").click()
def start():
elements = driver.find_elements_by_css_selector(".searchResult a")
links = [link.get_attribute("href") for link in elements]
result = []
for link in links:
if link not in result:
result.append(link)
else:
driver.get(link)
goUrl = urllib.request.urlopen(link)
soup = BeautifulSoup(goUrl.read(), "html.parser")
#table = soup.find_element_by_id("table", {"class": "applicationDetails"})
for i in range(20):
pass # Don't worry about all this commented code, it isn't relevant right now
#table = soup.find_element_by_id("table", {"class": "applicationDetails"})
#print(table.text)
# div = soup.select("div.applicationDetails")
# getDiv = div[i].split(":")[1].get_text()
# log = open("log.txt", "a")
# log.write(getDiv + "\n")
#log.write("\n")
start()
driver.get(url)
for i in range(5):
driver.find_element_by_id("ctl00_mainContentPlaceHolder_lvResults_bottomPager_ctl02_NextButton").click()
url = driver.current_url
start()
driver.get(url)
driver.close()

try this:
import time
# import config # Don't worry about this. This is an external file to make a DB
import urllib.request
from bs4 import BeautifulSoup
from selenium import webdriver
url = "https://planning.hants.gov.uk/SearchResults.aspx?RecentDecisions=True"
driver = webdriver.Chrome()
driver.get(url)
driver.find_element_by_id("mainContentPlaceHolder_btnAccept").click()
result = []
def start():
elements = driver.find_elements_by_css_selector(".searchResult a")
links = [link.get_attribute("href") for link in elements]
result.extend(links)
def start2():
for link in result:
# if link not in result:
# result.append(link)
# else:
driver.get(link)
goUrl = urllib.request.urlopen(link)
soup = BeautifulSoup(goUrl.read(), "html.parser")
#table = soup.find_element_by_id("table", {"class": "applicationDetails"})
for i in range(20):
pass # Don't worry about all this commented code, it isn't relevant right now
#table = soup.find_element_by_id("table", {"class": "applicationDetails"})
#print(table.text)
# div = soup.select("div.applicationDetails")
# getDiv = div[i].split(":")[1].get_text()
# log = open("log.txt", "a")
# log.write(getDiv + "\n")
#log.write("\n")
while True:
start()
element = driver.find_element_by_class_name('rdpPageNext')
try:
check = element.get_attribute('onclick')
if check != "return false;":
element.click()
else:
break
except:
break
print(result)
start2()
driver.get(url)

As per the url https://planning.hants.gov.uk/SearchResults.aspx?RecentDecisions=True to click through all the pages you can use the following solution:
Code Block:
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
options = Options()
options.add_argument("start-maximized")
options.add_argument("disable-infobars")
options.add_argument("--disable-extensions")
driver = webdriver.Chrome(chrome_options=options, executable_path=r'C:\Utility\BrowserDrivers\chromedriver.exe')
driver.get('https://planning.hants.gov.uk/SearchResults.aspx?RecentDecisions=True')
WebDriverWait(driver, 20).until(EC.element_to_be_clickable((By.ID, "mainContentPlaceHolder_btnAccept"))).click()
numLinks = len(WebDriverWait(driver, 20).until(EC.visibility_of_all_elements_located((By.CSS_SELECTOR, "div#ctl00_mainContentPlaceHolder_lvResults_topPager div.rdpWrap.rdpNumPart>a"))))
print(numLinks)
for i in range(numLinks):
print("Perform your scrapping here on page {}".format(str(i+1)))
WebDriverWait(driver, 20).until(EC.element_to_be_clickable((By.XPATH, "//div[#id='ctl00_mainContentPlaceHolder_lvResults_topPager']//div[#class='rdpWrap rdpNumPart']//a[#class='rdpCurrentPage']/span//following::span[1]"))).click()
driver.quit()
Console Output:
8
Perform your scrapping here on page 1
Perform your scrapping here on page 2
Perform your scrapping here on page 3
Perform your scrapping here on page 4
Perform your scrapping here on page 5
Perform your scrapping here on page 6
Perform your scrapping here on page 7
Perform your scrapping here on page 8

hi #Feitan Portor you have written the code absolutely perfect the only reason that you are redirected back to the first page is because you have given url = driver.current_url in the last for loop where it is the url that remains static and only the java script that instigates the next click event so just remove url = driver.current_url and driver.get(url)
and you are good to go i have tested my self
also to get the current page that your scraper is in just add this part in the for loop so you will get to know where your scraper is :
ss = driver.find_element_by_class_name('rdpCurrentPage').text
print(ss)
Hope this solves your confusion

Selenium no another page

I am scraping one page but the problem i came up today was that the page didn`t have another page and it gave me the previous page without any error from which i could determine that page was last one..
for ex: https://example/page-7
when i want to go to: https://example/page-8 which doesn`t exist it gives me
the last page: https://example/page-7
How could i determine that https://example/page-7 was the last page using python3???
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from bs4 import BeautifulSoup
import urllib.request
page = "https://www.supermama.lt/forumas/topic/214375-vilma/page-1"
driver = webdriver.Firefox()
driver.get(page)
dd = driver.page_source
for j in range(100):
soup = BeautifulSoup(dd, "html.parser")
my_text = list(soup.findAll("div", class_ = "post-content"))
for i in my_text:
#collect some data
pass
page = "https://www.supermama.lt/forumas/topic/214375-vilma/page{0}".format(j+2)
driver.get(page)
dd = driver.page_source
At first i was thinking about checking dublicates of collected data but this is too slow cause i have 30 000 links from which i have to collect data. Maybe there is easier solution??

Found the answer to my own question.
To find the page url just use driver.current_url
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from bs4 import BeautifulSoup
import urllib.request
page = "https://www.supermama.lt/forumas/topic/214375-vilma/"
driver = webdriver.Firefox()
driver.get(page)
dd = driver.page_source
current_pages = []
for j in range(100):
page_url = driver.current_url
if(page_url not in current_pages):
current_pages.append(page_url)
soup = BeautifulSoup(dd, "html.parser")
my_text = list(soup.findAll("div", class_ = "post-content"))
for i in my_text:
#collect some data
pass
page = "https://www.supermama.lt/forumas/topic/214375-vilma/page-{0}".format(j+2)
driver.get(page)
dd = driver.page_source
else:
print(current_pages)
driver.quit()
break

crawl playstore with python and phantomjs

This code will crawl again from the beginning every time an error occurs. I want to change this to crawl only new text, not just from the beginning.
and I would like to ask for further advice.
from selenium import webdriver
from time import sleep
from bs4 import BeautifulSoup, Comment
import pandas as pd
#Setting up Chrome webdriver Options
#chrome_options = webdriver.ChromeOptions()
#setting up local path of chrome binary file
#chrome_options.binary_location = "/Users/Norefly/chromedriver2/chromedriver.exec"
#creating Chrome webdriver instance with the set chrome_options
driver = webdriver.PhantomJS("C:/Python/phantomjs-2.1.1-windows/bin/phantomjs.exe")
link = "https://play.google.com/store/apps/details?id=com.supercell.clashofclans&hl=en"
driver.get(link)
#driver.execute_script("window.scrollTo(0, document.body.scrollHeight)")
Ptitle = driver.find_element_by_class_name('id-app-title').text.replace(' ','')
print(Ptitle)
#driver.find_element_by_xpath('//*[#id="body-content"]/div/div/div[1]/div[2]/div[2]/div[1]/div[4]/button[2]/div[2]').click()
sleep(1)
driver.find_element_by_xpath('//*[#id="body-content"]/div/div/div[1]/div[2]/div[2]/div[1]/div[4]/button[2]/div[2]/div/div').click()
#select_newest.select_by_visible_text('Newest')
#driver.find_element_by_xpath('//*[#id="body- content"]/div/div/div[1]/div[2]/div[2]/div[1]/div[4]/button[2]/div[2]/div/div').click()
sleep(2)
#driver.find_element_by_css_selector('.review-filter.id-review-sort-filter.dropdown-menu-container').click()
driver.find_element_by_css_selector('.displayed-child').click()
#driver.find_element_by_xpath("//button[#data-dropdown-value='1']").click()
driver.execute_script("document.querySelectorAll('button.dropdown-child')[0].click()")
reviews_df = []
for i in range(1,10):
try:
for elem in driver.find_elements_by_class_name('single-review'):
print(str(i))
content = elem.get_attribute('outerHTML')
soup = BeautifulSoup(content, "html.parser")
#print(soup.prettify())
date = soup.find('span',class_='review-date').get_text()
rating = soup.find('div',class_='tiny-star')['aria-label'][6:7]
title = soup.find('span',class_='review-title').get_text()
txt = soup.find('div',class_='review-body').get_text().replace('Full Review','')[len(title)+1:]
print(soup.get_text())
temp = pd.DataFrame({'Date':date,'Rating':rating,'Review Title':title,'Review Text':txt},index=[0])
print('-'*10)
reviews_df.append(temp)
#print(elem)
except:
print('what i can do?')
driver.find_element_by_xpath('//*[#id="body-content"]/div/div/div[1]/div[2]/div[2]/div[1]/div[4]/button[2]/div[2]/div/div').click()
#driver.execute_script("document.querySelectorAll('button.dropdown-child')[0].click()")
#driver.find_element_by_xpath('//*[#id="body-content"]/div/div/div[1]/div[2]/div[2]/div[1]/div[4]/button[2]/div[2]/div/div').click()
reviews_df = pd.concat(reviews_df,ignore_index=True)
reviews_df.to_csv(Ptitle+'review_google.csv', encoding='utf-8')
driver.close()
And I wonder if this is a problem with phantom js