dark_web_forums/MarketPlaces/Initialization/prepare_parser.py

__author__ = 'DarkWeb'

import glob
import os
import codecs
import shutil
import traceback
from psycopg2.extras import RealDictCursor

from MarketPlaces.DB_Connection.db_connection import *
from MarketPlaces.DarkFox.parser import *
from MarketPlaces.AnonymousMarketplace.parser import *
from MarketPlaces.TheDarkMarket.parser import *
from MarketPlaces.ViceCity.parser import *
from MarketPlaces.M00nkeyMarket.parser import *
from MarketPlaces.MikesGrandStore.parser import *
from MarketPlaces.PabloEscobarMarket.parser import *
from MarketPlaces.CityMarket.parser import *
from MarketPlaces.DarkBazar.parser import *
from MarketPlaces.Sonanza.parser import *
from MarketPlaces.Kingdom.parser import *
from MarketPlaces.BlackPyramid.parser import *
from MarketPlaces.Quest.parser import *
from MarketPlaces.Ares.parser import *
from MarketPlaces.CypherMarketplace.parser import *
from MarketPlaces.WeTheNorth.parser import *
from MarketPlaces.Torzon.parser import *
from MarketPlaces.GoFish.parser import *
from MarketPlaces.ZeroDay.parser import *

from MarketPlaces.Classifier.classify_product import predict

nError = 0


def mergePages(rmm, rec):

    # key = u"Pr:" + rec[1].upper() + u" Vendor:" + rec[18].upper()
    # key = rec[23]

    print("----------------- Matched: " + rec[4] + "--------------------")

    if rec[1] == "-1":          # name_vendor
        rec[1] = rmm[0]
    if rec[2] == "-1":          # rating_vendor
        rec[2] = rmm[1]
    if rec[3] == "-1":          # success_vendor
        rec[3] = rmm[2]
    if rec[4] == "-1":          # name_item
        rec[4] = rmm[3]
    if rec[5] == "-1":          # description_item
        rec[5] = rmm[4]
    if rec[6] == "-1":          # cve_item
        rec[6] = rmm[5]
    if rec[7] == "-1":          # ms_item
        rec[7] = rmm[6]
    if rec[8] == "-1":          # category_item
        rec[8] = rmm[7]
    if rec[9] == "-1":          # views_item
        rec[9] = rmm[8]
    if rec[10] == "-1":         # reviews_item
        rec[10] = rmm[9]
    if rec[11] == "-1":         # rating_item
        rec[11] = rmm[10]
    if rec[12] == "-1":         # adddate_item
        rec[12] = rmm[11]
    if rec[13] == "-1":         # btc_item
        rec[13] = rmm[12]
    if rec[14] == "-1":         # usd_item
        rec[14] = rmm[13]
    if rec[15] == "-1":         # euro_item
        rec[15] = rmm[14]
    if rec[16] == "-1":         # quantitysold_item
        rec[16] = rmm[15]
    if rec[17] == "-1":         # quantityleft_item
        rec[17] = rmm[16]
    if rec[18] == "-1":         # shippedfrom_item
        rec[18] = rmm[17]
    if rec[19] == "-1":         # shippedto_item
        rec[19] = rmm[18]
    if rmm[19] != "-1":         # image
        rec[20] = rmm[19]
    if rmm[20] != "-1":         # image_vendor
        rec[21] = rmm[20]

    return rec


def persist_data(url, row, cur):

    marketPlace = create_marketPlace(cur, row, url)

    vendor = create_vendor(cur, row, marketPlace)

    create_items(cur, row, marketPlace, vendor)


def incrementError():
    global nError
    nError += 1


def read_file(filePath, createLog, logFile):
    try:
        html = codecs.open(filePath.strip('\n'), encoding='utf8')
        soup = BeautifulSoup(html, "html.parser")
        html.close()
        time.sleep(0.01)  # making sure the file is closed before returning soup object
        return soup
    except:

        try:
            html = open(filePath.strip('\n'))
            soup = BeautifulSoup(html, "html.parser")
            html.close()
            time.sleep(0.01)  # making sure the file is closed before returning soup object
            return soup
        except:

            incrementError()
            print("There was a problem to read the file " + filePath)
            if createLog:
                logFile.write(
                    str(nError) + ". There was a problem to read the file " + filePath + "\n" + traceback.format_exc() + "\n")
            return None


def parse_listing(marketPlace, listingFile, soup, createLog, logFile):
    try:

        if marketPlace == "DarkFox":
            rw = darkfox_listing_parser(soup)
        elif marketPlace == "AnonymousMarketplace":
            rw = anonymousMarketplace_listing_parser(soup)
        elif marketPlace == "ViceCity":
            rw = vicecity_listing_parser(soup)
        elif marketPlace == "M00nkeyMarket":
            rw = m00nkey_listing_parser(soup)
        elif marketPlace == "MikesGrandStore":
            rw = MikesGrandStore_listing_parser(soup)
        elif marketPlace == "PabloEscobarMarket":
            rw = pabloescobarmarket_listing_parser(soup)
        elif marketPlace == "CityMarket":
            rw = city_listing_parser(soup)
        elif marketPlace == "Ares":
            rw = ares_listing_parser(soup)
        elif marketPlace == "DarkBazar":
            rw = darkbazar_listing_parser(soup)
        elif marketPlace == "Sonanza":
            rw = sonanza_listing_parser(soup)
        elif marketPlace == "Kingdom":
            rw = kingdom_listing_parser(soup)
        elif marketPlace == "BlackPyramid":
            rw = blackpyramid_listing_parser(soup)
        elif marketPlace == "Quest":
            rw = quest_listing_parser(soup)
        elif marketPlace == "CypherMarketplace":
            rw = cyphermarketplace_listing_parser(soup)
        elif marketPlace == "TheDarkMarket":
            rw = darkmarket_listing_parser(soup)
        elif marketPlace == "WeTheNorth":
            rw = wethenorth_listing_parser(soup)
        elif marketPlace == "GoFish":
            rw = gofish_listing_parser(soup)
        elif marketPlace == "ZeroDay":
            rw = zeroday_listing_parser(soup)
        elif marketPlace == "Torzon":
            rw = torzon_listing_parser(soup)
        else:
            print("MISSING CALL TO LISTING PARSER IN PREPARE_PARSER.PY!")
            raise Exception
        return rw

    except:

        incrementError()
        print("There was a problem to parse the file " + listingFile + " in the listing section!")
        traceback.print_exc()
        if createLog:
            logFile.write(
                str(nError) + ". There was a problem to parse the file " + listingFile + " in the Listing section.\n"
                + traceback.format_exc() + "\n")
        return None


def parse_description(marketPlace, descriptionFile, soup, createLog, logFile):
    try:

        if marketPlace == "DarkFox":
            rmm = darkfox_description_parser(soup)
        elif marketPlace == "AnonymousMarketplace":
            rmm = anonymousMarketplace_description_parser(soup)
        elif marketPlace == "ViceCity":
            rmm = vicecity_description_parser(soup)
        elif marketPlace == "M00nkeyMarket":
            rmm = m00nkey_description_parser(soup)
        elif marketPlace == "MikesGrandStore":
            rmm = MikesGrandStore_description_parser(soup)
        elif marketPlace == "PabloEscobarMarket":
            rmm = pabloescobarmarket_description_parser(soup)
        elif marketPlace == "CityMarket":
            rmm = city_description_parser(soup)
        elif marketPlace == "Ares":
            rmm = ares_description_parser(soup)
        elif marketPlace == "DarkBazar":
            rmm = darkbazar_description_parser(soup)
        elif marketPlace == "Sonanza":
            rmm = sonanza_description_parser(soup)
        elif marketPlace == "Kingdom":
            rmm = kingdom_description_parser(soup)
        elif marketPlace == "BlackPyramid":
            rmm = blackpyramid_description_parser(soup)
        elif marketPlace == "Quest":
            rmm = quest_description_parser(soup)
        elif marketPlace == "CypherMarketplace":
            rmm = cyphermarketplace_description_parser(soup)
        elif marketPlace == "TheDarkMarket":
            rmm = darkmarket_description_parser(soup)
        elif marketPlace == "WeTheNorth":
            rmm = wethenorth_description_parser(soup)
        elif marketPlace == "GoFish":
            rmm = gofish_description_parser(soup)
        elif marketPlace == "ZeroDay":
            rmm = zeroday_description_parser(soup)
        elif marketPlace == "Torzon":
            rmm = torzon_description_parser(soup)
        else:
            print("MISSING CALL TO DESCRIPTION PARSER IN PREPARE_PARSER.PY!")
            raise Exception
        return rmm

    except:

        incrementError()
        print("There was a problem to parse the file " + descriptionFile + " in the Description section!")
        traceback.print_exc()
        if createLog:
            logFile.write(
                str(nError) + ". There was a problem to parse the file " + descriptionFile + " in the Description section.\n"
                + traceback.format_exc() + "\n")
        return None


def persist_record(url, rec, cur, con, createLog, logFile, listingFile, descriptionFile):
    try:
        persist_data(url, tuple(rec), cur)
        con.commit()
        return True
    except:

        con.rollback()

        incrementError()
        print(f"There was a problem to persist the files ({listingFile} + {descriptionFile}) in the database!")
        traceback.print_exc()
        if createLog:
            logFile.write(
                str(nError) + f". There was a problem to persist the files ({listingFile} + {descriptionFile}) in the database!\n"
                + traceback.format_exc() + "\n")
        return False


def move_file(filePath, createLog, logFile):

    source = filePath
    destination = filePath.replace(os.path.basename(filePath), "") + 'Read\\' + os.path.basename(filePath)

    try:
        shutil.move(source, destination, shutil.copy2)
        return True
    except:

        try:
            shutil.move(source, destination, shutil.copytree)
            return True
        except:
            incrementError()
            print("There was a problem to move the file " + filePath)
            traceback.print_exc()
            if createLog:
                logFile.write(
                    str(nError) + ". There was a problem to move the file " + filePath + "\n" + traceback.format_exc() + "\n")
            return False


def new_parse(marketPlace, url, createLog):

    from MarketPlaces.Initialization.markets_mining import config, CURRENT_DATE

    global nError
    nError = 0

    print("Parsing the " + marketPlace + " market and conduct data classification to store the information in the database.")

    # Connecting to the database
    con = connectDataBase()
    cur = con.cursor(cursor_factory=RealDictCursor)

    # Creating the tables (The database should be created manually)
    create_database(cur, con)

    mainDir = os.path.join(config.get('Project', 'shared_folder'), "MarketPlaces\\" + marketPlace + "\\HTML_Pages")

    # Creating the log file for each Forum
    if createLog:
        try:
            logFile = open(mainDir + f"/{CURRENT_DATE}/" + marketPlace + "_" + CURRENT_DATE + ".log", "w")
        except:
            print("Could not open log file!")
            createLog = False
            logFile = None
            # raise SystemExit
    else:
        logFile = None

    # Reading the Listing Html Pages
    listings = glob.glob(os.path.join(mainDir, CURRENT_DATE + "\\Listing", '*.html'))

    listings.sort(key=os.path.getmtime)

    for listingIndex, listingFile in enumerate(listings):

        print("Reading listing folder of '" + marketPlace + "', file '" + os.path.basename(listingFile) + "', index= " + str(
            listingIndex + 1) + " ... " + str(len(listings)))

        listingSoup = read_file(listingFile, createLog, logFile)

        # listing flags
        doParseListing = listingSoup is not None
        doDescription = False

        readDescriptionError = False
        parseDescriptionError = False
        persistDescriptionError = False
        moveDescriptionError = False
        findDescriptionError = False

        rw = []       

        if doParseListing:

            rw = parse_listing(marketPlace, listingFile, listingSoup, createLog, logFile)
            
            doDescription = rw is not None

        if doDescription:

            nFound = 0

            for rec in rw:

                rec = rec.split(',')
                
                descriptionPattern = cleanLink(rec[22]) + ".html"

                # Reading the associated description Html Pages
                descriptions = glob.glob(os.path.join(mainDir, CURRENT_DATE + "\\Description", descriptionPattern))

                descriptions.sort(key=os.path.getmtime)

                nFound += len(descriptions)

                for descriptionIndex, descriptionFile in enumerate(descriptions):

                    print("Reading description folder of '" + marketPlace + "', file '" + os.path.basename(
                        descriptionFile) + "', index= " + str(descriptionIndex + 1) + " ... " + str(len(descriptions)))

                    descriptionSoup = read_file(descriptionFile, createLog, logFile)

                    # description flags
                    doParseDescription = descriptionSoup is not None
                    doPersistRecord = False
                    doMoveDescription = False

                    rmm = []

                    if doParseDescription:

                        rmm = parse_description(marketPlace, descriptionFile, descriptionSoup, createLog, logFile)

                        doPersistRecord = rmm is not None

                    else:
                        readDescriptionError = True
                        parseDescriptionError = True

                    if doPersistRecord:

                        # Combining the information from Listing and Description Pages
                        rec = mergePages(rmm, rec)

                        # Append to the list the classification of the topic
                        rec.append(str(predict(rec[4], rec[5], language='sup_english')))

                        # Persisting the information in the database
                        persistSuccess = persist_record(url, rec, cur, con, createLog, logFile, listingFile,
                                                        descriptionFile)

                        doMoveDescription = persistSuccess

                    else:
                        parseDescriptionError = True

                    if doMoveDescription:

                        # move description files of completed folder
                        moveSuccess = move_file(descriptionFile, createLog, logFile)

                        if not moveSuccess:
                            moveDescriptionError = True

                    else:
                        moveDescriptionError = True

            if not (nFound > 0):

                findDescriptionError = True

                incrementError()
                print(f"There was a problem to locate the file(s) for {listingFile} in the Description section!")
                if createLog:
                    logFile.write(
                        str(nError) + f". There was a problem to locate the file(s) for {listingFile}"
                                      f" in the Description section!\n\n")

            if not (readDescriptionError or parseDescriptionError or persistDescriptionError
                    or moveDescriptionError or findDescriptionError):
                # move listing files of completed folder
                move_file(listingFile, createLog, logFile)

    # registering the current forum status (up/down) and the number of scraped pages in the database
    marketId = verifyMarketPlace(cur, marketPlace)
    if (marketId > 0):

        readListings = glob.glob(os.path.join(mainDir, CURRENT_DATE + "\\Listing\\read", '*.html'))
        readDescriptions = glob.glob(os.path.join(mainDir, CURRENT_DATE + "\\Description\\read", '*.html'))

        create_status(cur, marketId, CURRENT_DATE, len(readListings), len(readDescriptions), '1' if len(listings) > 0 else '0')
        con.commit()

    if createLog:
        logFile.close()

    cur.close()
    con.close()

    print("Parsing the " + marketPlace + " market and data classification done.")