import os from datetime import datetime, timedelta from queue import Queue from urllib.parse import urlparse, urljoin import requests from bs4 import BeautifulSoup import indexing from myapp import search_index def time_passed(link): reader = ix.reader() try: link_time = reader.stored_fields(reader.first_id("url", link))["date"] time_difference = datetime.now() - link_time return time_difference > crawl_cooldown except: return True #Index: ix = indexing.get_index() ALLOWED_DOMAINS = ('minecraft.fandom.com',) crawl_cooldown = timedelta(hours=1) depth_limit = 2 queue = Queue() queue.put(('https://minecraft.fandom.com/wiki/Crafting_Table', 0)) #initialize queue with start URL visited_list = set() #list with visited websites while not queue.empty(): current_url, depth = queue.get() #take the first element of the queue visited_list.add(current_url) #working? r = requests.get(current_url, timeout=3) # get the current url soup = BeautifulSoup(r.content, 'html.parser') urltitle = str(soup.title.string) request_time = datetime.now() writer = ix.writer() writer.update_document(title=urltitle, url=r.url, content=str(soup.text), date=request_time) #add website information to crawler writer.commit()# put crawler content into the index print(F"Queue length: {queue.qsize()}, searching on depth: {depth}, site: {r.url}") if depth <= depth_limit: new_links = soup.find_all('a', href=True, limit=20) # find all links on this website for link in new_links: # for each of the links on the current website href = urljoin(current_url, link['href']) # join the domain with its relative links url = urlparse(href) if os.path.splitext(url.path)[1] in ('.html', '', '.htm') and url.hostname in ALLOWED_DOMAINS and href not in visited_list and time_passed(url): # if the url has the right format, is inside the allowed domains and has not been visited recently queue.put((href, depth + 1)) print(visited_list)