init nexus market crawler and parser

MarketPlaces/NexusMarket/crawler_selenium.py

@@ -0,0 +1,302 @@
+__author__ = 'Helium'
+
+'''
+CityMarket Marketplace Crawler (Selenium)
+not complete
+need to go through multiple pages...
+'''
+
+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 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.NexusMarket.parser import nexus_links_parser
+from MarketPlaces.Utilities.utilities import cleanHTML
+
+counter = 1
+baseURL = 'http://nexusabcdkq4pdlubs6wk6ad7pobuupzoomoxi6p7l32ci4vjtb2z7yd.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():
+    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
+#return: name of site in string type
+def getMKTName():
+    name = 'NexusMarket'
+    return name
+
+
+# Return the base link of the website
+#return: url of base site in string type
+def getFixedURL():
+    url = 'http://nexusabcdkq4pdlubs6wk6ad7pobuupzoomoxi6p7l32ci4vjtb2z7yd.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 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'
+#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'
+
+
+# Manual captcha solver, waits fora specific element so that the whole page loads, finds the input box, gets screenshot of captcha
+# then allows for manual solving of captcha in the terminal
+#@param: current selenium web driver
+def login(driver):
+    input("Press ENTER when CAPTCHA is complete and login page has loaded\n")
+
+    # entering username and password into input boxes
+    usernameBox = driver.find_element(by=By.XPATH, value='//input[@id="username"]')
+    # Username here
+    usernameBox.send_keys('findingmykeys')
+    passwordBox = driver.find_element(by=By.XPATH, value='//input[@id="inputPassword3"]')
+    # Password here
+    passwordBox.send_keys('ican’tFindMycarKey$')
+
+    input("Press ENTER when CAPTCHA is completed and you exit the newsletter\n")
+
+    # wait for listing page show up (This Xpath may need to change based on different seed url)
+    WebDriverWait(driver, 100).until(EC.visibility_of_element_located(
+        (By.XPATH, '//*[@id="collapse3"]')))
+
+
+# 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 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
+
+
+# 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
+# Guides and Tutorials, Digital Products, and Software and Malware
+#as you can see they are categories of products
+def getInterestedLinks():
+    links = []
+
+    # # Hire hacker
+    # links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=3')
+    # # other
+    # links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=14')
+    # malware
+    links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=15')
+    # ddos
+    links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=16')
+    # software
+    links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=17')
+    # botnet
+    links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=18')
+    # hacking service
+    links.append('http://wsptlnuoo3johqzcdlwuj5zcwfh2dwmswz6hahqctuxttvxpanypmwad.onion/?sub_id=31')
+
+    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 CityMarket 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, linksToCrawl[i] + f"page{count+1}")
+
+                list = productPages(html)
+                for item in list:
+                    # what is this line doing?
+                    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:
+                    link = driver.find_element(by=By.XPATH, value='//a[@rel="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 CityMarket market 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 'single_product' 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 'sub_id' 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 ie. getInterestingLinks()
+#return: list of description links that should be crawled through
+def productPages(html):
+    soup = BeautifulSoup(html, "html.parser")
+    return nexus_links_parser(soup)
+
+
+# Drop links that "signout"
+# def isSignOut(url):
+#     #absURL = urlparse.urljoin(url.base_url, url.url)
+#     if 'signout' in url.lower() or 'logout' in url.lower():
+#         return True
+#
+#     return False
+
+
+def crawler():
+    startCrawling()
+    # print("Crawling and Parsing NexusMarket .... DONE!")

MarketPlaces/NexusMarket/parser.py

@@ -0,0 +1,212 @@
+__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
+
+
+#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 nexus_description_parser(soup):
+
+    # Fields to be parsed
+
+    vendor = "-1"                       # 0 *Vendor_Name
+    success = "-1"                      # 1 Vendor_Successful_Transactions
+    rating_vendor = "-1"                # 2 Vendor_Rating
+    name = "-1"                         # 3 *Product_Name
+    describe = "-1"                     # 4 Product_Description
+    CVE = "-1"                          # 5 Product_CVE_Classification (Common Vulnerabilities and Exposures)
+    MS = "-1"                           # 6 Product_MS_Classification (Microsoft Security)
+    category = "-1"                     # 7 Product_Category
+    views = "-1"                        # 8 Product_Number_Of_Views
+    reviews = "-1"                      # 9 Product_Number_Of_Reviews
+    rating_item = "-1"                  # 10 Product_Rating
+    addDate = "-1"                      # 11 Product_AddedDate
+    BTC = "-1"                          # 12 Product_BTC_SellingPrice
+    USD = "-1"                          # 13 Product_USD_SellingPrice
+    EURO = "-1"                         # 14 Product_EURO_SellingPrice
+    sold = "-1"                         # 15 Product_QuantitySold
+    left = "-1"                         # 16 Product_QuantityLeft
+    shipFrom = "-1"                     # 17 Product_ShippedFrom
+    shipTo = "-1"                       # 18 Product_ShippedTo
+    image = "-1"                        # 19 Product_Image
+    vendor_image = "-1"                 # 20 Vendor_Image
+
+    divmd7 = soup.find('div', {'class': "col-md-7"})
+
+    # Finding Vendor
+    vendor = divmd7.find('a').text.strip()
+
+    # Finding Prices
+    USD = soup.find('span', {'class': "total"}).text.strip()
+
+    tempBTC = soup.find('div', {'class': "text-center"}).text.strip()
+    BTC = tempBTC.replace("BTC", "").strip()
+
+    # Finding Product Image
+    image = soup.find('img', {'class': 'img-fluid'})
+    image = image.get('src')
+    image = image.split('base64,')[-1]
+
+    # Finding the Product description
+    describe = soup.find('div', {'class': "text-white"}).text
+    describe = cleanString(describe.strip())
+
+    # Searching for CVE and MS categories
+    cve = soup.findAll(text=re.compile('CVE-\d{4}-\d{4}'))
+    if cve:
+        CVE = " "
+        for idx in cve:
+            CVE += (idx)
+            CVE += "  "
+            CVE = CVE.replace(',', ' ')
+            CVE = CVE.replace('\n', '')
+    ms = soup.findAll(text=re.compile('MS\d{2}-\d{3}'))
+    if ms:
+        MS = " "
+        for im in ms:
+            MS += (im)
+            MS += " "
+            MS = MS.replace(',', ' ')
+            MS = MS.replace('\n', '')
+
+    # 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 nexus_listing_parser(soup):
+
+    # Fields to be parsed
+    nm = 0                                    # *Total_Products (Should be Integer)
+    mktName = "CityMarket"                     # 0 *Marketplace_Name
+    vendor = []                               # 1 *Vendor y
+    rating_vendor = []                        # 2 Vendor_Rating
+    success = []                              # 3 Vendor_Successful_Transactions
+    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
+    views = []                                # 9 Product_Number_Of_Views
+    reviews = []                              # 10 Product_Number_Of_Reviews
+    rating_item = []                          # 11 Product_Rating
+    addDate = []                              # 12 Product_AddDate
+    BTC = []                                  # 13 Product_BTC_SellingPrice
+    USD = []                                  # 14 Product_USD_SellingPrice y
+    EURO = []                                 # 15 Product_EURO_SellingPrice
+    sold = []                                 # 16 Product_QuantitySold
+    qLeft = []                                # 17 Product_QuantityLeft
+    shipFrom = []                             # 18 Product_ShippedFrom
+    shipTo = []                               # 19 Product_ShippedTo
+    image = []                                # 20 Product_Image
+    image_vendor = []                         # 21 Vendor_Image
+    href = []                                 # 22 Product_Links
+
+    listing = soup.findAll('div', {"class": "p-4"})
+
+    # Populating the Number of Products
+    nm = len(listing)
+
+    for a in listing:
+        bae = a.findAll('a', href=True)
+
+        # Adding the url to the list of urls
+        link = bae[0].get('href')
+        href.append(link)
+
+        # Category
+        tempCategory = soup.find('select', {"name": "category"})
+        tempCategory = tempCategory.find('option', selected=True).text.strip()
+        category.append(tempCategory)
+
+        # Product Name
+        product = a.find('h4', {"class": "text-center"}).text
+        product = product.replace('\n', ' ')
+        product = product.replace(",", "")
+        product = product.replace("...", "")
+        product = product.strip()
+        name.append(product)
+
+        # USD and BTC Price
+        price = a.find('div', {"class": "price"}).text
+        if "~" in price:
+            tempUSD = price.split("~")[0]
+            tempUSD = tempUSD.replace("$", "")
+            tempUSD = tempUSD.strip()
+            USD.append(tempUSD)
+
+            tempBTC = price.split("~")[1]
+            tempBTC = tempBTC.replace("BTC", "")
+            tempBTC = tempBTC.strip()
+            BTC.append(tempBTC)
+        else:
+            USD.append("-1")
+            BTC.append("-1")
+
+        # Img
+        product_image = a.find('img')
+        product_image = product_image.get('src')
+        product_image = product_image.split('base64,')[-1]
+        image.append(product_image)
+
+        # Searching for CVE and MS categories
+        cve = a.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', '')
+            cveValue=cee
+        CVE.append(cveValue)
+        
+        ms = a.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)
+
+    # 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 nexus_links_parser(soup):
+
+    # Returning all links that should be visited by the Crawler
+
+    href = []
+    listing = soup.findAll('div', {"class": "p-4"})
+
+    for a in listing:
+        bae = a.find('a', href=True)
+        link = bae['href']
+        href.append(link)
+
+    return href