completed atlas marketplace

MarketPlaces/Atlas/crawler_selenium.py

@@ -0,0 +1,283 @@
+__author__ = 'DarkWeb'
+
+'''
+Atlas Marketplace 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.support.ui import WebDriverWait
+from selenium.webdriver.support.ui import Select
+from selenium.webdriver.support import expected_conditions as EC
+from selenium.webdriver.common.by import By
+
+from PIL import Image
+import urllib.parse as urlparse
+import os, re, time
+from datetime import date
+import subprocess
+import configparser
+from bs4 import BeautifulSoup
+from MarketPlaces.Initialization.prepare_parser import new_parse
+from MarketPlaces.Atlas.parser import atlas_links_parser
+from MarketPlaces.Utilities.utilities import cleanHTML
+
+counter = 1
+baseURL = 'http://5jbmfuip2ohxabshoapzy23kz36movs4clpivnsu7lnrn4osazhzdtqd.onion'
+
+def startCrawling():
+    mktName = getMKTName()
+    # driver = getAccess()
+    #
+    # if driver != 'down':
+    #     try:
+    #         login(driver)
+    #         crawlForum(driver)
+    #     except Exception as e:
+    #         print(driver.current_url, e)
+    #     closeDriver(driver)
+
+    new_parse(mktName, baseURL, True)
+
+
+# Returns the name of the website
+def getMKTName():
+    name = 'Atlas'
+    return name
+
+
+# Return the base link of the website
+def getFixedURL():
+    url = 'http://5jbmfuip2ohxabshoapzy23kz36movs4clpivnsu7lnrn4osazhzdtqd.onion'
+    return url
+
+
+# Closes Tor Browser
+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 MarketPlaces.Initialization.markets_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)
+    # 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", False)
+    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'
+def getAccess():
+    url = getFixedURL()
+    driver = createFFDriver()
+    try:
+        driver.get(url)
+        return driver
+    except:
+        driver.close()
+        return 'down'
+
+def agreeToTerms(driver):
+    try:
+        agree_button = driver.find_element(by=By.XPATH, value='//input[@type="submit" and @name="agree" and @value="Yes, I agree"]')
+        agree_button.click()
+    except Exception as e:
+        print('Problem with clicking agree button', e)
+
+def login(driver):
+    # input("Press ENTER when CAPTCHA is complete and login page has loaded\n")
+    input("Press ENTER when captcha is solved")
+    WebDriverWait(driver, 100).until(EC.visibility_of_element_located(
+        (By.XPATH, '//*[@id="username"]')))
+
+    # entering username and password into input boxes
+    usernameBox = driver.find_element(by=By.XPATH, value='//*[@id="username"]')
+    # Username here
+    usernameBox.send_keys('atldark')
+    passwordBox = driver.find_element(by=By.XPATH, value='//*[@id="password"]')
+    # Password here
+    passwordBox.send_keys('a1T2l3dark')
+
+    input("Press ENTER when captcha is solved and you're logged in")
+    #
+    # try:
+    #     agree_button = driver.find_element(by=By.XPATH, value='/html/body/main/div/form/div[6]/button')
+    #     agree_button.click()
+    # except Exception as e:
+    #     print('Problem with clicking login button', e)
+
+
+
+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
+
+
+def getFullPathName(url):
+    from MarketPlaces.Initialization.markets_mining import config, CURRENT_DATE
+
+    mainDir = os.path.join(config.get('Project', 'shared_folder'), "MarketPlaces/" + getMKTName() + "/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
+
+
+def getMKTName() -> str:
+    name = 'Atlas'
+    return name
+
+
+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
+
+
+def getInterestedLinks():
+    links = []
+
+    # hacking
+    # links.append('http://5jbmfuip2ohxabshoapzy23kz36movs4clpivnsu7lnrn4osazhzdtqd.onion/c/hacking')
+    # fraud
+    links.append('http://5jbmfuip2ohxabshoapzy23kz36movs4clpivnsu7lnrn4osazhzdtqd.onion/c/fraud')
+    # software
+    links.append('http://5jbmfuip2ohxabshoapzy23kz36movs4clpivnsu7lnrn4osazhzdtqd.onion/c/software')
+    # services
+    # links.append('http://5jbmfuip2ohxabshoapzy23kz36movs4clpivnsu7lnrn4osazhzdtqd.onion/c/services')
+
+    return links
+
+
+def crawlForum(driver):
+
+    print("Crawling the Atlas market")
+
+    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)
+                list = productPages(html)
+
+                for item in list:
+                    itemURL = urlparse.urljoin(baseURL, str(item))
+                    try:
+                        driver.get(itemURL)
+                    except:
+                        driver.refresh()
+                    savePage(driver, driver.page_source, item)
+                    driver.back()
+
+                    # comment out
+                    break
+
+                # comment out
+                # if count == 1:
+                #     break
+
+                try:
+                    temp = driver.find_element(by=By.XPATH, value='/html/body/div[2]/form/section/div/div/div[2]/div[2]')
+                    temp = temp.find_elements(by=By.TAG_NAME, value='a')
+                    a_tag = temp[-1]
+
+                    if a_tag:
+                        temp = a_tag.find_element(by=By.CLASS_NAME, value='double-arrow-right')
+                        if temp:
+                            link = a_tag.get_attribute('href')
+                        else:
+                            link = ''
+
+                    if link == "":
+                        raise NoSuchElementException
+                    count += 1
+
+                except NoSuchElementException:
+                    has_next_page = False
+
+        except Exception as e:
+            print(link, e)
+        i += 1
+
+    print("Crawling the vortex market done.")
+
+
+# Returns 'True' if the link is Topic link, may need to change for every website
+def isDescriptionLink(url):
+    if 'product' in url:
+        return True
+    return False
+
+
+# Returns True if the link is a listingPage link, may need to change for every website
+def isListingLink(url):
+    if '/c/' in url:
+        return True
+    return False
+
+
+def productPages(html):
+    soup = BeautifulSoup(html, "html.parser")
+    return atlas_links_parser(soup)
+
+
+def crawler():
+    startCrawling()

MarketPlaces/Atlas/parser.py

@@ -0,0 +1,273 @@
+__author__ = 'DarkWeb'
+
+# Here, we are importing the auxiliary functions to clean or convert data
+from MarketPlaces.Utilities.utilities import *
+
+# Here, we are importing BeautifulSoup to search through the HTML tree
+from bs4 import BeautifulSoup
+import re
+
+
+# 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 atlas_description_parser(soup):
+    # Fields to be parsed
+
+    vendor = "-1"  # 0 *Vendor_Name y
+    success = "-1"  # 1 Vendor_Successful_Transactions n
+    rating_vendor = "-1"  # 2 Vendor_Rating y
+    name = "-1"  # 3 *Product_Name y
+    describe = "-1"  # 4 Product_Description y
+    CVE = "-1"  # 5 Product_CVE_Classification (Common Vulnerabilities and Exposures)
+    MS = "-1"  # 6 Product_MS_Classification (Microsoft Security)
+    category = "-1"  # 7 Product_Category y
+    views = "-1"  # 8 Product_Number_Of_Views y
+    reviews = "-1"  # 9 Product_Number_Of_Reviews n
+    rating_item = "-1"  # 10 Product_Rating n
+    addDate = "-1"  # 11 Product_AddedDate y
+    BTC = "-1"  # 12 Product_BTC_SellingPrice y
+    USD = "-1"  # 13 Product_USD_SellingPrice y
+    EURO = "-1"  # 14 Product_EURO_SellingPrice n
+    sold = "-1"  # 15 Product_QuantitySold n
+    left = "-1"  # 16 Product_QuantityLeft n
+    shipFrom = "-1"  # 17 Product_ShippedFrom n
+    shipTo = "-1"  # 18 Product_ShippedTo n
+    image = "-1"  # 19 Product_Image n
+    vendor_image = "-1"  # 20 Vendor_Image n
+
+    # Finding Vendor
+    vendor = soup.find('div', {'class', 'flex items-center space-x-3'})
+    vendor = vendor.find('div', {'class', 'flex items-center space-x-3'}).find('a').text
+    vendor = cleanString(vendor).strip()
+
+    # sold
+    try:
+        sell = soup.find('p', {'class': 'pl-3 text-sm font-semibold text-slate-700 dark:text-slate-400'}).text
+        reg = r'product sold (.+)'
+        sell = re.search(reg, sell)
+        if sell:
+            sell = sell.group(1)
+        sell = sell.replace('product sold ', '').replace('time', '').replace('s', '')
+        sell = cleanString(sell.strip())
+    except:
+        sell = '-1'
+    sold = sell
+
+    # successful transaction
+    suc = soup.find('div', {'class': 'mt-4 grid w-full grid-cols-6 gap-4 rounded border p-3 shadow-sm dark:border-slate-700'}).find_all('div')
+    suc = suc[2].find_all('p')
+    suc = suc[1].text
+    success = cleanString(suc.strip())
+
+
+    # Finding Vendor Rating (bug in their system shows standard rating)
+    rating_vendor = '-1'
+    rating_vendor = cleanNumbers(rating_vendor).strip()
+
+    # Finding Product Name
+    name = soup.find('h2').text
+    name = cleanString(name).strip()
+
+    # Finding Product description
+    try:
+        describe = soup.find('div', {'class': 'prose mb-12 mt-5 break-words dark:prose-invert'}).text
+        describe = cleanString(describe).strip()
+    except:
+        describe = '-1'
+
+    # Finding category
+    try:
+        div_category = soup.find('ol', {'class': "leading-node flex items-center gap-1 text-sm font-medium text-gray-600 dark:text-gray-300"}).find_all('li', class_=lambda x: x is None)
+        category = div_category[1].find('a').text
+    except:
+        category = '-1'
+    category = cleanString(category).strip()
+
+    # Product rating - check
+    rating = '-1'
+    rating_item = cleanString(rating).strip()
+
+    # Finding BTC and USD/GOLD
+    BTC = '-1'
+
+    usd = soup.find('h1', {'class': 'text-2xl font-bold leading-none tracking-tight text-slate-800 dark:text-slate-300'}).text
+    if '$' in usd:
+        usd = usd.replace('$', '')
+        usd = cleanString(usd).strip()
+    USD = usd
+
+    try:
+        image = soup.find('div', {'class', 'w-full flex-1 flex-shrink-0'}).find('img').get('src').split('base64,')[-1]
+    except:
+        image = '-1'
+
+
+    # Populating the final variable (this should be a list with all fields scraped)
+    row = (vendor, rating_vendor, success, name, describe, CVE, MS, category, views, reviews, rating_item, addDate,
+           BTC, USD, EURO, sold, left, shipFrom, shipTo, image, vendor_image)
+
+    # Sending the results
+    return row
+
+
+# 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 atlas_listing_parser(soup):
+    # Fields to be parsed
+    nm = 0  # *Total_Products (Should be Integer)
+    mktName = "Atlas"  # 0 *Marketplace_Name y
+    vendor = []  # 1 *Vendor y
+    rating_vendor = []  # 2 Vendor_Rating y
+    success = []  # 3 Vendor_Successful_Transactions n
+    name = []  # 4 *Product_Name y
+    CVE = []  # 5 Product_CVE_Classification (Common Vulnerabilities and Exposures) dont worry about this
+    MS = []  # 6 Product_MS_Classification (Microsoft Security) dont worry about this
+    category = []  # 7 Product_Category y
+    describe = []  # 8 Product_Description n
+    views = []  # 9 Product_Number_Of_Views y
+    reviews = []  # 10 Product_Number_Of_Reviews n
+    rating_item = []  # 11 Product_Rating n
+    addDate = []  # 12 Product_AddDate y
+    BTC = []  # 13 Product_BTC_SellingPrice y
+    USD = []  # 14 Product_USD_SellingPrice y
+    EURO = []  # 15 Product_EURO_SellingPrice n
+    sold = []  # 16 Product_QuantitySold n
+    qLeft = []  # 17 Product_QuantityLeft n
+    shipFrom = []  # 18 Product_ShippedFrom n
+    shipTo = []  # 19 Product_ShippedTo n
+    image = []  # 20 Product_Image n
+    image_vendor = []  # 21 Vendor_Image n
+    href = []  # 22 Product_Links y
+
+    listings = soup.find('div', {'class': 'grid h-fit grid-cols-1 gap-x-8 gap-y-10 lg:grid-cols-4 md:grid-cols-3'}).findAll('div', class_=lambda x: x is None)
+
+    temp = soup.find('h1', {'class': 'text-4xl font-bold tracking-tight text-gray-900 dark:text-slate-200'}).text
+    reg = r'results in (.+)'
+    cat = re.search(reg, temp)
+    if cat:
+        cat = cat.group(1)
+    cat = cat.replace('results in ', '')
+
+    # Populating the Number of Products
+    nm = len(listings)
+
+    for listing in listings:
+        # Finding vendor name
+        vendor_name = listing.find('p', {'class': 'text-sm font-medium leading-none dark:text-slate-100'}).text
+        vendor_name = vendor_name.strip()
+        vendor.append(vendor_name)
+
+        # Finding the vendor rating
+        vendor_level = listing.find('p', {'class': "flex space-x-1 text-xs text-gray-700 dark:text-slate-500"}).text
+        vendor_level = vendor_level.strip().split(' ')
+        vendor_level = cleanNumbers(vendor_level[-1])
+        rating_vendor.append(vendor_level)
+
+        # Finding the product name
+        product_name = listing.find('p', {'class':'my-1 line-clamp-2 text-sm text-slate-700 group-hover:underline dark:text-slate-300'}).text
+        product_name = cleanString(product_name).strip()
+        name.append(product_name)
+
+        # Searching for CVE and MS categories
+        cve = listing.findAll(text=re.compile('CVE-\d{4}-\d{4}'))
+        if not cve:
+            cveValue = "-1"
+        else:
+            cee = " "
+            for idx in cve:
+                cee += (idx)
+                cee += "  "
+                cee = cee.replace(',', ' ')
+                cee = cee.replace('\n', '')
+                cee = cee.replace('\t', '')
+            cveValue = cee
+        CVE.append(cveValue)
+
+        ms = listing.findAll(text=re.compile('MS\d{2}-\d{3}'))
+        if not ms:
+            MSValue = "-1"
+        else:
+            me = " "
+            for im in ms:
+                me += (im)
+                me += " "
+                me = me.replace(',', ' ')
+                me = me.replace('\n', '')
+            MSValue = me
+        MS.append(MSValue)
+
+        # Finding the category
+        category_text = cleanString(cat).strip()
+        category.append(category_text)
+
+        # Finding the hrefs
+        description_link = listing.find('a', {'class': 'group relative block'})['href']
+        href.append(description_link)
+
+        # Finding the views - check
+        views_text = '-1'
+        views.append(views_text)
+
+        # Finding the date added
+        date = '-1'
+        # date = datetime.strptime(date, "%d-%m-%Y")
+        addDate.append(date)
+
+        # EURO
+        EURO.append('-1')
+
+        # Finding the BTC and USD/GOLD
+        btc_price = '-1'
+        BTC.append(btc_price)
+
+        try:
+            money = listing.find('p', {'class': 'mt-2 text-lg font-bold tracking-wider text-slate-900 dark:text-slate-200'}).text
+            if '$' in money:
+                usd_price = money.strip().replace('$', '')
+                usd_price = cleanString(usd_price).strip()
+        except:
+            usd_price = '-1'
+        USD.append(usd_price)
+
+        sold.append('-1')
+        qLeft.append('-1')
+        shipTo.append('-1')
+
+        shipf = '-1'
+        shipFrom.append(shipf)
+
+        try:
+            temp = listing.find('div', {'class', 'overflow-hidden rounded-lg'}).find('img').get('src').split('base64,')[-1]
+        except:
+            temp = '-1'
+        image.append(temp)
+        image_vendor.append('-1')
+
+    # Populate the final variable (this should be a list with all fields scraped)
+    return organizeProducts(mktName, nm, vendor, rating_vendor, success, name, CVE, MS, category, describe, views,
+                            reviews, rating_item, addDate, BTC, USD, EURO, sold, qLeft, shipFrom, shipTo, href, image,
+                            image_vendor)
+
+
+# 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 atlas_links_parser(soup):
+    # Returning all links that should be visited by the Crawler
+
+    href = []
+    listings = soup.find('div', {'class': 'grid h-fit grid-cols-1 gap-x-8 gap-y-10 lg:grid-cols-4 md:grid-cols-3'}).findAll('div')
+
+    for listing in listings:
+        # Adding the url to the list of urls
+        try:
+            description_link = listing.find('a', {'class': 'group relative block'}).get('href')
+            href.append(description_link)
+        except:
+            pass
+
+    return href