dark_web_forums/MarketPlaces/SilkRoad4/parser.py

__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


# This is the method to parse the Description Pages (one page to each Product in the Listing Pages)
def silkroad4_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
    LTC = "-1"                          # 19 Product_LTC_SellingPrice
    XMR = "-1"                          # 20 Product_XMR_SellingPrice
    image = "-1"                        # 19 Product_Image
    vendor_image = "-1"                 # 20 Vendor_Image

    bae = soup.find('div', {'id': 'cats'})

    # Finding Vendor
    vendor = bae.find('font', {'color':'blue'}).text
    vendor = vendor.strip()

    #Finding Product Name
    name = bae.find('b', {'style': 'color:#333;'}).text
    name = name.strip()

    #Finding image
    image = bae.find('img')
    image = image.get('src')
    image = image.split('base64,')[-1]

    #Finding Price
    temp = bae.find('span').next_sibling
    price_list = temp.split("/")

    USD = price_list[0].replace("$", "").strip()
    BTC = price_list[1].replace("BTC","").strip()

    # Not stored into databse/PGAdmin yet!
    LTC = price_list[2].replace("LTC", "").strip()
    XMR = price_list[3].replace("XMR", "").strip()
    #print(USD, "", BTC,"", LTC,"", XMR)

    # Finding Ships From
    a = bae.find_all('a',href=True, limit = 2)
    shipFrom = a[0].text.strip()

    #Finding Category
    category = a[1].text.strip()

    # Finding the Product description
    describe = soup.find('div', {'style': 'color:#555;font-weight:normal;font-size:12px'}).text
    describe = cleanString(describe.strip())

    # Finding Rating
    rate = soup.find('div', {'style': 'padding:0px; margin-bottom:10px; font-size:12px;'})
    rate = rate.find('p')
    if rate is not None:
        rate = rate.text.strip()

        #Some descriptions have rating as 'No Rating yet', can convert to -1 for consistency in database
        #if(rate is 'No rating yet'):
           # rating_item = -1

        #Only extract rating part
        rating_item = rate[:rate.index('Note')]



    # Finding Number of Reviews
    table = soup.find('div', {'class': 'table-responsive'})
    if table is not None:
        num_rev = table.findAll('tr')
        reviews = len(num_rev) - 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


# This is the method to parse the Listing Pages
def silkroad4_listing_parser(soup):

    # Fields to be parsed
    nm = 0                                    # *Total_Products (Should be Integer)
    mktName = "SilkRoad4"                     # 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)
    MS = []                                   # 6 Product_MS_Classification (Microsoft Security)
    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
    href = []                                 # 20 Product_Links
    LTC = []                                  # 21 Product_LTC_SellingPrice
    XMR = []                                  # 22 Product_XMR_SellingPrice
    image = []                                # 19 Product_Image
    image_vendor = []                         # 20 Vendor_Image


    listing = soup.findAll('div', {'style': "padding:10px; width:100%; margin-bottom:5px; margin-top:0px"})

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

    # Finding category of listing page
    cat = listing[0].find_all('a', href=True)
    cat = cat[len(cat)-2].text
    cat = cat.replace(",", "")
    cat = cat.strip()

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

        # Adding the category
        category.append(cat)


        # Adding the url to the list of urls
        link = bae[0].get('href')
        link = cleanLink(link)
        href.append(link)

        # Finding Price
        temp = a.find('b', {'style': 'color:#333'})
        temp = temp.text
        price = temp.split("/")
        USD.append(price[0].replace('$', '').strip())
        BTC.append(price[1].replace('BTC', '').strip())

        # LTC and XMR will not be stored in Pgadmin as of now
        LTC.append(price[2].replace('LTC', '').strip())
        XMR.append(price[3].replace('XMR', '').strip())
        #print(USD, " ", BTC, ' ', LTC, '', XMR)

        # Finding the Vendor
        ven = a.findAll('b')
        v = ven[2].find('font').text
        v = v.strip()
        vendor.append(v)

        # Finding the Product
        product = bae[0].text
        product = product.replace(",", "")
        product = product.replace(",", "")
        product = product.replace("...", "")
        product = product.strip()
        name.append(product)


        # Finding ShipFrom
        shipf = bae[len(bae)-1].text.strip()
        shipFrom.append(shipf)

        # Finding image
        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)


def silkroad4_links_parser(soup):

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

    # finds all div with id, vp
    divs = soup.findAll('div', {"id": "vp"})

    listing = []

    # for all div with id:vp, find first a with href
    for div in divs:
        a_s = div.find('a', href=True)

        # if div has no href True, then move on. otherwise, store the a with href into listing
        if a_s is not None:
            listing.append(a_s)

    # loop through listing with a with href and extract/return the href
    for a in listing:
        link = a['href']

        # if '?listing' is in link, then it is a product, so append to the href list. Otherwise, move on.
        if "?listing" in link:
            href.append(link)

    return href