dark_web_forums/MarketPlaces/BlackPyramid/parser.py

__author__ = 'cern'

# 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 blackpyramid_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

    # Finding Product Name
    name = soup.find('div', {'class': 'panel39002'}).find('span').next_sibling
    name = name.replace('\n', ' ')
    name = name.replace(",", "")
    name = name.strip()

    # Finding Product Rating
    rating_span = soup.find('span', {'class': 'to3098503t'}).find_next_sibling('span')
    rating_num = rating_span.find('b').text
    if rating_num != 'N/A':
        rating_item = rating_num[0:3]

    # product description
    describe = soup.findAll('div', {'class': 'fer048953'})[1].text
    describe = describe.replace('\n', ' ')
    describe = describe.replace(",", "")
    describe = describe.strip()

    # Finding Vendor
    vendor = soup.find('div', {'class': 'bold03905 vstat364'}).text
    vendor = vendor.split(" ")
    vendor = vendor[2][:-1]
    vendor = vendor.replace('\n', ' ')
    vendor = vendor.replace(",", "")
    vendor = vendor.strip()

    # Finding Product Rating
    rating_div = soup.find('div', {'class': 'bold03905 vstat364'}).find_next_sibling('div').find_next_sibling('div')
    rating_vendor = cleanNumbers(rating_div.text)
    if rating_vendor == "":
        rating_vendor = "-1"

    # Finding Successful Transactions
    success_container = soup.find('ul', {'class': 'ul3o00953'}).findAll('li')[1]
    success = success_container.find('div').text
    success = success.replace('"', '')
    success = success.replace("\n", " ")
    success = success.replace(",", "")
    success = success.strip()

    # Finding Prices
    USD_text = soup.find('li', {'class': 'vul2994 vghul995'}).find('div').text
    USD = USD_text.split(',')[1]
    USD = USD.replace('\n', ' ')
    USD = USD.replace(",", "")
    USD = USD.strip()

    container = soup.find('ul', {'class': 'bic03095'})

    # Finding Number Sold
    sold_container = container.find('li')
    sold_div = sold_container.findAll('div')[2]
    sold = sold_div.find('b').next_sibling
    sold = sold.replace('"', '')
    sold = sold.replace("\n", " ")
    sold = sold.replace(",", "")
    sold = sold.strip()

    # Finding the Product Quantity Available
    left_container = container.find('li')
    left_div = left_container.findAll('div')[3]
    left = left_div.find('b').next_sibling
    left = left.replace('"', '')
    left = left.replace("\n", " ")
    left = left.replace(",", "")
    left = left.strip()

    # Finding number of reviews
    positive = soup.find('span', {'class': 'ar04999324'}).text
    neutral = soup.find('span', {'class': 'ti9400005 can39953'}).text
    negative = soup.find('span', {'class': 'ti9400005 ti90088 can39953'}).text
    reviews = int(positive) + int(neutral) + int(negative)

    # Finding product image
    image = soup.find('img', {'class': 'img0390503'})
    image = image.get('src')
    image = image.split('base64,')[-1]

    vendor_image = soup.find('img', {'class': 'img0390503'})
    vendor_image = vendor_image.get('src')
    vendor_image = vendor_image.split('base64,')[-1]

    # 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 blackpyramid_listing_parser(soup):

    # Fields to be parsed
    nm = 0                                      # *Total_Products (Should be Integer)
    mktName = "BlackPyramid"                   # 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('article', {"class": "product"})

    # Some listing pages have an additional article section which is blank
    if not listing[-1].findAll('a', href=True):
        listing = listing[:-1]


    # Populating the Number of Products
    nm = len(listing)

    for card in listing:
        bae = card.findAll('a', href=True)

        # Adding the url to the list of urls
        link = bae[2].get('href')
        href.append(link)

        # Finding the Product
        product = bae[3].text
        product = product.replace('\n', ' ')
        product = product.replace(",", "")
        product = product.replace("...", "")
        product = product.strip()
        name.append(product)

        # # Finding description
        # # 'recurisve = False' only searches direct children
        # desc = card.findChildren('div', recursive=False)[0]
        # desc = desc.findAll('div', recursive=False)[3].text
        # desc = desc.replace('\n', ' ')
        # desc = desc.replace(",", "")
        # desc = desc.strip()
        # describe.append(desc)

        # Finding Vendor Name
        vendor_name = bae[4].find('span').text
        vendor_name = vendor_name.split(' ')[1]
        vendor_name = vendor_name.replace('\n', ' ')
        vendor_name = vendor_name.replace(",", "")
        vendor_name = vendor_name.strip()
        vendor.append(vendor_name)

        # Finding the Category
        cat = card.findAll('div', recursive=False)[0].findAll('div', recursive=False)[1].find('span').text
        cat = cat.replace("\n", "")
        cat = cat.replace(",", "")
        cat = cat.strip()
        category.append(cat)

        bae = card.findAll('div', recursive=False)[1].findAll('div', recursive=False)[1]

        # Finding amount left
        left = bae.findAll('div', recursive=False)[1].text
        left = left.replace("x", "")
        left = left.replace('\n', ' ')
        left = left.replace(",", "")
        left = left.strip()
        qLeft.append(left)

        # Finding amount sold
        qsold = bae.findAll('div', recursive=False)[2].text
        qsold = qsold.replace('\n', ' ')
        qsold = qsold.replace("x", "")
        qsold = qsold.replace(",", "")
        qsold = qsold.strip()
        sold.append(qsold)

        # Finding product image
        product_image = card.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 = card.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 = card.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 BlackPyramid_links_parser(soup):

    # Returning all links that should be visited by the Crawler
    href = []

    listing = soup.findAll('article', {"class": "product"})

    for item in listing:

        link = item.find('a', {"class": "ah39063"})['href']
        href.append(link)

    return href