darknetarmy scraper completed

Forums/DarkNetArmy/crawler_selenium.py

@@ -0,0 +1,282 @@
+__author__ = 'DarkWeb'
+
+'''
+DarkNetArmy Forum Crawler (Selenium)
+'''
+
+from selenium import webdriver
+from selenium.common.exceptions import NoSuchElementException
+from selenium.webdriver.firefox.firefox_profile import FirefoxProfile
+from selenium.webdriver.firefox.firefox_binary import FirefoxBinary
+from selenium.webdriver.firefox.service import Service
+from selenium.webdriver.common.by import By
+
+import urllib.parse as urlparse
+import os, time
+from datetime import date
+import subprocess
+from bs4 import BeautifulSoup
+from Forums.Initialization.prepare_parser import new_parse
+from Forums.DarkNetArmy.parser import darknetarmy_links_parser
+from Forums.Utilities.utilities import cleanHTML
+
+counter = 1
+baseURL = 'http://darknet77vonbqeatfsnawm5jtnoci5z22mxay6cizmoucgmz52mwyad.onion/'
+
+
+# Opens Tor Browser, crawls the website, then parses, then closes tor
+#acts like the main method for the crawler, another function at the end of this code calls this function later
+def startCrawling():
+    forumName = getForumName()
+    # driver = getAccess()
+    #
+    # if driver != 'down':
+    #     try:
+    #         crawlForum(driver)
+    #     except Exception as e:
+    #         print(driver.current_url, e)
+    #     closeDriver(driver)
+
+    new_parse(forumName, baseURL, True)
+
+
+# Returns the name of the website
+#return: name of site in string type
+def getForumName():
+    name = 'DarkNetArmy'
+    return name
+
+
+# Return the base link of the website
+#return: url of base site in string type
+def getFixedURL():
+    url = 'http://darknet77vonbqeatfsnawm5jtnoci5z22mxay6cizmoucgmz52mwyad.onion/'
+    return url
+
+
+# Closes Tor Browser
+#@param: current selenium driver
+def closeDriver(driver):
+    # global pid
+    # os.system("taskkill /pid " + str(pro.pid))
+    # os.system("taskkill /t /f /im tor.exe")
+    print('Closing Tor...')
+    driver.close()
+    time.sleep(3)
+    return
+
+
+# Creates FireFox 'driver' and configure its 'Profile'
+# to use Tor proxy and socket
+def createFFDriver():
+    from Forums.Initialization.forums_mining import config
+
+    ff_binary = FirefoxBinary(config.get('TOR', 'firefox_binary_path'))
+
+    ff_prof = FirefoxProfile(config.get('TOR', 'firefox_profile_path'))
+    ff_prof.set_preference("places.history.enabled", False)
+    ff_prof.set_preference("privacy.clearOnShutdown.offlineApps", True)
+    ff_prof.set_preference("privacy.clearOnShutdown.passwords", True)
+    ff_prof.set_preference("privacy.clearOnShutdown.siteSettings", True)
+    ff_prof.set_preference("privacy.sanitize.sanitizeOnShutdown", True)
+    ff_prof.set_preference("signon.rememberSignons", False)
+    ff_prof.set_preference("network.cookie.lifetimePolicy", 2)
+    ff_prof.set_preference("network.dns.disablePrefetch", True)#might need to turn off
+    ff_prof.set_preference("network.http.sendRefererHeader", 0)
+    ff_prof.set_preference("permissions.default.image", 3)
+    ff_prof.set_preference("browser.download.folderList", 2)
+    ff_prof.set_preference("browser.download.manager.showWhenStarting", False)
+    ff_prof.set_preference("browser.helperApps.neverAsk.saveToDisk", "text/plain")
+    ff_prof.set_preference('network.proxy.type', 1)
+    ff_prof.set_preference("network.proxy.socks_version", 5)
+    ff_prof.set_preference('network.proxy.socks', '127.0.0.1')
+    ff_prof.set_preference('network.proxy.socks_port', 9150)
+    ff_prof.set_preference('network.proxy.socks_remote_dns', True)
+    ff_prof.set_preference("javascript.enabled", True)
+    ff_prof.update_preferences()
+
+    service = Service(config.get('TOR', 'geckodriver_path'))
+
+    driver = webdriver.Firefox(firefox_binary=ff_binary, firefox_profile=ff_prof, service=service)
+
+    driver.maximize_window()
+
+    return driver
+
+
+#the driver 'gets' the url, attempting to get on the site, if it can't access return 'down'
+#return: return the selenium driver or string 'down'
+def getAccess():
+    url = getFixedURL()
+    driver = createFFDriver()
+    try:
+        driver.get(url)
+        return driver
+    except:
+        driver.close()
+        return 'down'
+
+
+# Saves the crawled html page, makes the directory path for html pages if not made
+def savePage(driver, page, url):
+    cleanPage = cleanHTML(driver, page)
+    filePath = getFullPathName(url)
+    os.makedirs(os.path.dirname(filePath), exist_ok=True)
+    open(filePath, 'wb').write(cleanPage.encode('utf-8'))
+    return
+
+
+# Gets the full path of the page to be saved along with its appropriate file name
+#@param: raw url as crawler crawls through every site
+def getFullPathName(url):
+    from Forums.Initialization.forums_mining import config, CURRENT_DATE
+
+    mainDir = os.path.join(config.get('Project', 'shared_folder'), "Forums/" + getForumName() + "/HTML_Pages")
+    fileName = getNameFromURL(url)
+    if isDescriptionLink(url):
+        fullPath = os.path.join(mainDir, CURRENT_DATE + r'\\Description\\' + fileName + '.html')
+    else:
+        fullPath = os.path.join(mainDir, CURRENT_DATE + r'\\Listing\\' + fileName + '.html')
+    return fullPath
+
+
+# Creates the file name from passed URL, gives distinct name if can't be made unique after cleaned
+#@param: raw url as crawler crawls through every site
+def getNameFromURL(url):
+    global counter
+    name = ''.join(e for e in url if e.isalnum())
+    if (name == ''):
+        name = str(counter)
+        counter = counter + 1
+    return name
+
+
+# returns list of urls, here is where you can list the different urls of interest, the crawler runs through this list
+#in this example, there are a couple of categories some threads fall under such as
+#exploits, malware, and hacking tutorials
+def getInterestedLinks():
+    links = []
+
+    # rats, malware, ransomware
+    links.append('http://darknet77vonbqeatfsnawm5jtnoci5z22mxay6cizmoucgmz52mwyad.onion/forums/rats-malwares-key-loggers-ransomware-tools.24/')
+    # hacking tutorials
+    links.append('http://darknet77vonbqeatfsnawm5jtnoci5z22mxay6cizmoucgmz52mwyad.onion/forums/hacking-cracking-tutorials-courses-methods.45/')
+    # # hacking
+    # links.append('http://darknet77vonbqeatfsnawm5jtnoci5z22mxay6cizmoucgmz52mwyad.onion/forums/hacking-cracking-tools-apps.21/')
+    return links
+
+
+# gets links of interest to crawl through, iterates through list, where each link is clicked and crawled through
+#topic and description pages are crawled through here, where both types of pages are saved
+#@param: selenium driver
+def crawlForum(driver):
+    print("Crawling the BestCardingWorld forum")
+
+    linksToCrawl = getInterestedLinks()
+
+    i = 0
+    while i < len(linksToCrawl):
+        link = linksToCrawl[i]
+        print('Crawling :', link)
+        try:
+            has_next_page = True
+            count = 0
+
+            while has_next_page:
+                try:
+                    driver.get(link)
+                except:
+                    driver.refresh()
+                html = driver.page_source
+                savePage(driver, html, link)
+
+                topics = topicPages(html)
+                for topic in topics:
+                    has_next_topic_page = True
+                    counter = 1
+                    page = topic
+
+                    while has_next_topic_page:
+                        itemURL = urlparse.urljoin(baseURL, str(page))
+                        try:
+                            driver.get(itemURL)
+                        except:
+                            driver.refresh()
+                        
+                        if isListingLink(driver.current_url):
+                            break
+
+                        savePage(driver, driver.page_source, topic + f"page{counter}")  # very important
+
+                        # comment out
+                        # if counter == 2:
+                        #     break
+
+                        try:
+                            page = link = driver.find_element(by=By.XPATH, value='//a[contains(text(), "Next")]').get_attribute('href')
+                            if page == "":
+                                raise NoSuchElementException
+                            counter += 1
+
+                        except NoSuchElementException:
+                            has_next_topic_page = False
+
+                    # making sure we go back to the listing page (browser back button simulation)
+                    try:
+                        driver.get(link)
+                    except:
+                        driver.refresh()
+
+                    # comment out
+                    # break
+
+                # comment out
+                # if count == 1:
+                #     break
+
+                try:
+                    link = driver.find_element(by=By.XPATH, value='//a[contains(text(), "Next")]').get_attribute('href')
+                    if link == "":
+                        raise NoSuchElementException
+                    count += 1
+
+                except NoSuchElementException:
+                    has_next_page = False
+
+        except Exception as e:
+            print(link, e)
+        i += 1
+
+    print("Crawling the DarkNetArmy forum done.")
+
+
+# Returns 'True' if the link is a description link
+#@param: url of any url crawled
+#return: true if is a description page, false if not
+def isDescriptionLink(url):
+    if 'threads' in url:
+        return True
+    return False
+
+
+# Returns True if the link is a listingPage link
+#@param: url of any url crawled
+#return: true if is a Listing page, false if not
+def isListingLink(url):
+    if 'forums' in url:
+        return True
+    return False
+
+
+# calling the parser to define the links, the html is the url of a link from the list of interested link list
+#@param: link from interested link list
+#return: list of description links that should be crawled through
+def topicPages(html):
+    soup = BeautifulSoup(html, "html.parser")
+    #print(soup.find('div', {"class": "forumbg"}).find('ul', {"class": "topiclist topics"}).find('li', {"class": "row bg1"}).find('a', {"class": "topictitle"}, href=True))
+    return darknetarmy_links_parser(soup)
+
+
+def crawler():
+    startCrawling()
+    # print("Crawling and Parsing BestCardingWorld .... DONE!")

Forums/DarkNetArmy/parser.py

@@ -0,0 +1,262 @@
+__author__ = 'DarkWeb'
+
+import re
+
+# Here, we are importing the auxiliary functions to clean or convert data
+from Forums.Utilities.utilities import *
+
+# Here, we are importing BeautifulSoup to search through the HTML tree
+from bs4 import BeautifulSoup
+
+
+# This is the method to parse the Description Pages (one page to each topic in the Listing Pages)
+#parses description pages, so takes html pages of description pages using soup object, and parses it for info it needs
+#stores info it needs in different lists, these lists are returned after being organized
+#@param: soup object looking at html page of description page
+#return: 'row' that contains a variety of lists that each hold info on the description page
+def darknetarmy_description_parser(soup):
+    # Fields to be parsed
+
+    topic = "-1"  # 0 topic name
+    user = []  # 1 all users of each post
+    status = []  # 2 all user's authority in each post such as (adm, member, dangerous)
+    reputation = []  # 3 all users's karma in each post (usually found as a number)
+    interest = []  # 4 all user's interest in each post
+    sign = []  # 5 all user's signature in each post (usually a standard message after the content of the post)
+    post = []  # 6 all messages of each post
+    feedback = []  # 7 all feedbacks of each user (this was found in just one Forum and with a number format)
+    addDate = []  # 8 all dated of each post
+    image_user = []  # 9 all user avatars of each post
+    image_post = []  # 10 all first images of each post
+
+    # Finding the topic (should be just one coming from the Listing Page)
+
+    topic = soup.find("h1", {"class": "p-title-value"})
+    topic = topic.text
+    topic = topic.replace(",", "")
+    topic = topic.replace("\n", "")
+    topic = cleanString(topic.strip())
+
+    # Finding the repeated tag that corresponds to the listing of posts
+
+    # posts = soup.find("form", {"name": "quickModForm"}).findAll('div', {"class": "windowbg"}) + \
+    #         soup.find("form", {"name": "quickModForm"}).findAll('div', {"class": "windowbg2"})
+
+    posts = soup.findAll('article', class_=re.compile("message message--post js-post js-inlineModContainer.*"))
+
+    # For each message (post), get all the fields we are interested to:
+    for ipost in posts:
+
+        # Finding a first level of the HTML page
+
+        #post_wrapper = ipost.find('div', {"class": "post_wrapper"}).find('div', {"class": "poster"})
+        post_wrapper = ipost.find('div', {"class": "message-inner"})
+        # Finding the author (user) of the post
+
+        # users
+        author = post_wrapper.find('div', {'class': 'message-userName'}).find('h4').text
+        user.append(cleanString(author))  # Remember to clean the problematic characters
+
+        # Finding the status of the author
+        try:
+            membergroup = post_wrapper.find('h5', {'class': 'userTitle message-userTitle'}).text
+        except:
+            membergroup = '-1'
+
+        status.append(cleanString(membergroup))
+
+        # reputation
+        temp = post_wrapper.find('div', {'class': 'message-userExtras'}).find_all('dl')
+        rep = temp[2].find('dd').text
+        if 'K' or 'k' in rep:
+            rep = rep.replace('K', '000').replace('k', '000')
+        reputation.append(rep)
+
+        # na
+        interest.append('-1')
+        sign.append('-1')
+        feedback.append('-1')
+        image_post.append('-1')
+
+        try:
+            message = post_wrapper.find('article', {'class': 'message-body js-selectToQuote'}).text
+            message = cleanString(message.strip())
+        except:
+            message = post_wrapper.find('div', {'content': 'message-content js-messageContent'}).text
+
+            message = cleanString(message.strip())
+
+        post.append(message)
+
+
+        time = post_wrapper.find('ul', class_ = re.compile(r'message-attribution-main listInline.*')).find('time').text
+        if ',' in time:
+            time = time.replace(',', '')
+        if 'today' in time:
+            today = datetime.today()
+            time = today
+        elif 'at' or 'AM' or 'PM' in time:
+            today = datetime.today()
+            start_of_week = today - timedelta(days=(today.weekday() + 1) % 7)
+            days_mapping = {
+                'Sunday': start_of_week,
+                'Monday': start_of_week + timedelta(days=1),
+                'Tuesday': start_of_week + timedelta(days=2),
+                'Wednesday': start_of_week + timedelta(days=3),
+                'Thursday': start_of_week + timedelta(days=4),
+                'Friday': start_of_week + timedelta(days=5),
+                'Saturday': start_of_week + timedelta(days=6),
+            }
+            for day, date in days_mapping.items():
+                if day in time:
+                    time = date.strftime('%Y-%m-%d')
+                    break
+        addDate.append(time)
+
+        try:
+            image = post_wrapper.find('div', {'class': 'message-avatar '}).find('img').get('src').split('base64,')[-1]
+        except:
+            image = '-1'
+        image_user.append(image)
+
+    # Populate the final variable (this should be a list with all fields scraped)
+
+    row = (topic, user, status, reputation, interest, sign, post, feedback, addDate, image_user, image_post)
+
+    # Sending the results
+
+    return row
+
+
+# This is the method to parse the Listing Pages (one page with many posts)
+#parses listing pages, so takes html pages of listing pages using soup object, and parses it for info it needs
+#stores info it needs in different lists, these lists are returned after being organized
+#@param: soup object looking at html page of listing page
+#return: 'row' that contains a variety of lists that each hold info on the listing page
+def darknetarmy_listing_parser(soup):
+    nm = 0  # *this variable should receive the number of topics
+    forum = "DarkNetArmy"  # 0 *forum name
+    board = "-1"  # 1 *board name (the previous level of the topic in the Forum categorization tree.
+    # For instance: Security/Malware/Tools to hack Facebook. The board here should be Malware)
+    author = []  # 2 *all authors of each topic
+    topic = []  # 3 *all topics
+    views = []  # 4 number of views of each topic
+    posts = []  # 5 number of posts of each topic
+    href = []  # 6 this variable should receive all cleaned urls (we will use this to do the marge between
+    # Listing and Description pages)
+    addDate = []  # 7 when the topic was created (difficult to find)
+    image_author = []  # 8 all author avatars used in each topic
+
+    # Finding the board (should be just one)
+
+    board = soup.find('h1', {"class": "p-title-value"}).text
+    board = board.replace(u"\xbb", "")
+    board = cleanString(board.strip())
+
+    # Finding the repeated tag that corresponds to the listing of topics
+
+    itopics = soup.find('div', {"class": "structItemContainer-group js-threadList"}).find_all('div', class_=re.compile(
+        r'^structItem structItem--thread js-inlineModContainer js-threadListItem.*'))
+
+    nm = len(itopics)
+
+    index = 0
+    for itopic in itopics:
+        # authors
+        a = itopic.find('ul', {"class": "structItem-parts"}).find('li').text
+        a = cleanString(a.strip())
+        author.append(a)
+
+        # topic
+        top = itopic.find('div', {"class": 'structItem-title'}).text
+        top = cleanString(top.strip())
+        topic.append(top)
+
+        # href
+        ref = itopic.find('div', {"class": 'structItem-title'}).find('a').get('href')
+        href.append(ref)
+
+        # image
+        try:
+            image = itopic.find('div', {"class": 'structItem-iconContainer'}).find('img').get('src').split('base64,')[
+                -1]
+        except:
+            image = '-1'
+
+        image_author.append(image)
+
+        # add date
+        try:
+            time = soup.find('li', {"class": 'structItem-startDate'}).find('time').text
+            if ',' in time:
+                time = time.replace(',', '')
+            time = time.strip()
+            if 'today' in time:
+                today = datetime.today()
+                time = today
+            elif 'at' or 'AM' or 'PM' in time:
+                today = datetime.today()
+                start_of_week = today - timedelta(days=(today.weekday() + 1) % 7)
+                days_mapping = {
+                    'Sunday': start_of_week,
+                    'Monday': start_of_week + timedelta(days=1),
+                    'Tuesday': start_of_week + timedelta(days=2),
+                    'Wednesday': start_of_week + timedelta(days=3),
+                    'Thursday': start_of_week + timedelta(days=4),
+                    'Friday': start_of_week + timedelta(days=5),
+                    'Saturday': start_of_week + timedelta(days=6),
+                }
+                for day, date in days_mapping.items():
+                    if day in time:
+                        time = date.strftime('%Y-%m-%d')
+                        break
+            addDate.append(time)
+        except:
+            addDate.append('-1')
+
+        try:
+            temp = itopic.find('div', class_=re.compile(r'^structItem-cell structItem-cell--meta.*')).find_all('dl')
+            try:
+                reply = temp[0].find('dd').text
+                reply = cleanString(reply.strip())
+                if 'K' or 'k' in reply:
+                    reply = reply.replace('K', '000').replace('k', '000')
+            except:
+                reply = '-1'
+            posts.append(reply)
+
+            # views
+            try:
+                view = temp[1].find('dd').text
+                view = cleanString(view.strip())
+                if 'K' or 'k' in view:
+                    view = view.replace('K', '000').replace('k', '000')
+            except:
+                view = '-1'
+            views.append(view)
+        except:
+            reply = '-1'
+            posts.append(reply)
+            view = '-1'
+            views.append(view)
+
+    return organizeTopics(forum, nm, board, author, topic, views, posts, href, addDate, image_author)
+
+
+#called by the crawler to get description links on a listing page
+#@param: beautifulsoup object that is using the correct html page (listing page)
+#return: list of description links from a listing page
+def darknetarmy_links_parser(soup):
+    # Returning all links that should be visited by the Crawler
+
+    href = []
+
+    listing = soup.find('div', {"class": "block-container block-container--nodes"}).findAll('div', {
+        "class": "structItem-title"})
+
+    for a in listing:
+        bae = a.find('a', href=True)
+        link = bae['href']
+        href.append(link)
+
+    return href