dark_web_forums/Forums/Altenens/parser.py

__author__ = 'DarkWeb'

# Here, we are importing the auxiliary functions to clean or convert data
from Forums.Utilities.utilities import *
from datetime import date
from datetime import timedelta
import re

# 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 topic in the Listing Pages)
def altenens_description_parser(soup):

    topic = "-1"            # 0 *topic name
    user = []               # 1 *all users of each post
    status = []             # 2 all user's authority in each post such as (adm, member, dangerous)
    reputation = []         # 3 all user's karma in each post (usually found as a number)
    interest = []           # 4 all user's interest in each post
    sign = []               # 5 all user's signature in each post (usually a standard message after the content of the post)
    post = []               # 6 all messages of each post
    feedback = []           # 7 all feedbacks of each vendor (this was found in just one Forum and with a number format)
    addDate = []            # 8 all dates of each post
    image_user = []         # 9 all user avatars of each post
    image_post = []         # 10 all first images of each post

    etopic = soup.find("h1", {"class": "p-title-value"})
    if etopic is not None:
        topic = etopic.text
        topic = cleanString(topic.strip())

    body = soup.find('div', {"class": "block-container lbContainer"})
    iposts = body.find_all('article', {"class": "message message--post js-post js-inlineModContainer"})

    for ipost in iposts:

        author = ipost.find('h4', {"class": "message-name"}).text
        user.append(cleanString(author.strip()))

        stat = ipost.find('h5', {"class": "userTitle message-userTitle"}).text
        status.append(cleanString(stat.strip()))

        bar = ipost.find('div', {"class": "xtr-progress-bar"})
        if bar is not None:
            rep = bar.find('p').get('data-value')
        else:
            rep = "-1"
        reputation.append(cleanString(rep))

        interest.append("-1")

        signature = ipost.find('aside', {"class": "message-signature"})
        if signature is not None:
            signature = signature.text.strip()
        else:
            signature = "-1"
        sign.append(cleanString(signature))

        inner = ipost.find('div', {"class": "bbWrapper"}).find(text=True, recursive=False)
        if inner is not None:
            inner = inner.strip()
        else:
            inner = ""  # cannot use -1 because the post is hidden unless you reply
        post.append(cleanString(inner))

        feedback.append("-1")

        dt = ipost.find('time', {"class": "u-dt"}).get('datetime')
        date_time_obj = datetime.strptime(dt, '%Y-%m-%dT%H:%M:%S%z')
        addDate.append(date_time_obj)

        img = ipost.find('div', {"class": "message-avatar-wrapper"}).find('img')
        if img is not None:
            img = img.get('src').split('base64,')[-1]
        else:
            img = "-1"
        image_user.append(img)

        image_post.append("-1")

    # Populate the final variable (this should be a list with all fields scraped)

    row = (topic, user, status, reputation, interest, sign, post, feedback, addDate, image_user, image_post)

    # Sending the results

    return row


# This is the method to parse the Listing Pages (one page with many posts)
def altenens_listing_parser(soup):

    nm = 0              # *this variable should receive the number of topics
    forum = "Altenens"  # 0 *forum name
    board = "-1"        # 1 *board name (the previous level of the topic in the Forum categorization tree.
                        # For instance: Security/Malware/Tools to hack Facebook. The board here should be Malware)
    author = []         # 2 *all authors of each topic
    topic = []          # 3 *all topics
    views = []          # 4 number of views of each topic
    posts = []          # 5 number of posts of each topic
    href = []           # 6 this variable should receive all cleaned urls (we will use this to do the marge between
                        # Listing and Description pages)
    addDate = []        # 7 when the topic was created (difficult to find)
    image_author = []   # 8 all author avatars used in each topic

    board = soup.find('h1', {"class": "p-title-value"}).text
    board = cleanString(board.strip())

    regex = re.compile('structItem structItem--thread.*')
    itopics = soup.find_all('div', {"class": regex})

    nm = len(itopics)

    for itopic in itopics:

        topics = itopic.find('div', {"class": "structItem-title"}).text
        topic.append(cleanString(topics.strip()))

        author_icon = itopic.find('a', {"class": "avatar avatar--s"})
        if author_icon != None:
            author_icon = author_icon.find('img')
            author_icon = author_icon.get('src')
            author_icon = author_icon.split('base64,')[-1]
        else:
            author_icon = "-1"
        image_author.append(author_icon)

        link = itopic.find('div', {"class": "structItem-title"}).find('a').get('href')
        href.append(link)

        user = itopic.find('ul', {"class": "structItem-parts"}).find('a').text
        author.append(cleanString(user.strip()))

        dt = itopic.find('time', {"class": "u-dt"}).get('datetime')
        date_time_obj = datetime.strptime(dt, '%Y-%m-%dT%H:%M:%S%z')
        addDate.append(date_time_obj)

        nposts = itopic.find('dl', {"class": "pairs pairs--justified"}).text
        nposts = nposts.replace('Replies', '')
        nposts = nposts.replace('K', '000')
        posts.append(cleanString(nposts))

        nviews = itopic.find('dl', {"class": "pairs pairs--justified structItem-minor"}).text
        nviews = nviews.replace('Views', '')
        nviews = nviews.replace('K', '000')
        views.append(cleanString(nviews))

    return organizeTopics(forum, nm, board, author, topic, views, posts, href, addDate, image_author)


def altenens_links_parser(soup):

    # Returning all links that should be visited by the Crawler

    href = []

    listing = soup.find_all('div', {"class": "structItem-cell structItem-cell--main"})

    for a in listing:
        link = a.find('a', {"class": ""}).get('href')

        href.append(link)

    return href