You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
226 lines
8.0 KiB
226 lines
8.0 KiB
__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, ResultSet, Tag
|
|
|
|
|
|
# This is the method to parse the Description Pages (one page to each topic in the Listing Pages)
|
|
def libre_description_parser(soup: Tag):
|
|
# Fields to be parsed
|
|
|
|
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
|
|
|
|
# Finding the topic (should be just one coming from the Listing Page)
|
|
|
|
topic_found = soup.find("a", {"class": "link text-xl text-zinc-300"}).text
|
|
topic = cleanString(topic_found.strip())
|
|
|
|
original_post: Tag = soup.find("div", {"class": "flex items-start"})
|
|
|
|
original_user = original_post.find("div", {"class": "info-p"}).find("a", {"class": "link"}).text
|
|
user.append(cleanString(original_user.replace("/u/", "").strip()))
|
|
|
|
original_user_statistics: ResultSet[Tag] = original_post.find("div", {"class": "info-p"}).find_all("span")
|
|
|
|
original_time = original_user_statistics[0].text[2:]
|
|
datetime_append = datetime.strptime(original_time, "%Y-%m-%d %H:%M:%S GMT")
|
|
addDate.append(datetime_append)
|
|
|
|
original_karma = original_user_statistics[1].text[2]
|
|
reputation.append(cleanString(original_karma.strip()))
|
|
|
|
original_content = soup.find("div", {"class": "content-p"}).text
|
|
post.append(cleanString(original_content.strip()))
|
|
|
|
|
|
status.append("-1")
|
|
interest.append("-1")
|
|
sign.append("-1")
|
|
feedback.append("-1")
|
|
|
|
# Finding the repeated tag that corresponds to the listing of posts
|
|
|
|
# try:
|
|
posts: ResultSet[Tag] = soup.find_all("div", {"class": "flex items-stretch"})
|
|
|
|
# For each message (post), get all the fields we are interested to:
|
|
|
|
for ipost in posts:
|
|
# Finding a first level of the HTML page
|
|
|
|
# Finding the author (user) of the post
|
|
|
|
user_name = ipost.find("p", {"class": "text-zinc-400 text-justify"}).find("a", {"class": "link"}).text
|
|
user.append(cleanString(user_name.replace("/u/", "").strip())) # Remember to clean the problematic characters
|
|
|
|
status.append("-1")
|
|
|
|
# Finding the interest of the author
|
|
# CryptBB does not have blurb
|
|
|
|
interest.append("-1")
|
|
|
|
# Finding the reputation of the user
|
|
# CryptBB does have reputation
|
|
|
|
karma = ipost.find("p", {"class": "text-zinc-400 text-justify"}).text
|
|
karma_cleaned = karma.split(" ")[6]
|
|
reputation.append(cleanString(karma_cleaned.strip()))
|
|
|
|
# Getting here another good tag to find the post date, post content and users' signature
|
|
|
|
date_posted = ipost.find("p", {"class": "text-zinc-400 text-justify"}).text
|
|
date_time_cleaned = date_posted.replace(user_name, "")[3:-12]
|
|
datetime_append = datetime.strptime(date_time_cleaned, "%Y-%m-%d %H:%M:%S GMT")
|
|
addDate.append(datetime_append)
|
|
|
|
# Finding the post
|
|
user_post = ipost.find("div", {"class": "content-c"}).text
|
|
post.append(cleanString(user_post))
|
|
|
|
# Finding the user's signature
|
|
|
|
sign.append("-1")
|
|
|
|
# As no information about user's feedback was found, just assign "-1" to the variable
|
|
|
|
feedback.append("-1")
|
|
|
|
# Populate the final variable (this should be a list with all fields scraped)
|
|
# print(topic)
|
|
# print(user)
|
|
# print(status)
|
|
# print(reputation)
|
|
# print(interest)
|
|
# print(sign)
|
|
# print(post)
|
|
# print(feedback)
|
|
# print(addDate)
|
|
# print(len(user))
|
|
# print(len(status))
|
|
# print(len(reputation))
|
|
# print(len(interest))
|
|
# print(len(sign))
|
|
# print(len(feedback))
|
|
# print(len(addDate))
|
|
|
|
row = (topic, user, status, reputation, interest, sign, post, feedback, addDate)
|
|
|
|
# Sending the results
|
|
|
|
return row
|
|
|
|
|
|
# This is the method to parse the Listing Pages (one page with many posts)
|
|
def libre_listing_parser(soup):
|
|
nm = 0 # *this variable should receive the number of topics
|
|
forum = "Libre" # 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)
|
|
|
|
# Finding the board (should be just one)
|
|
|
|
board = soup.find('div', {"class": "title"}).find("h1").text
|
|
board = cleanString(board.strip())
|
|
|
|
# Finding the repeated tag that corresponds to the listing of topics
|
|
|
|
itopics = soup.find("div", {"class", "space-y-2 mt-4"}).find_all('div', {"class": "flex box"})
|
|
|
|
nm = 0
|
|
for itopic in itopics:
|
|
nm += 1
|
|
# For each topic found, the structure to get the rest of the information can be of two types. Testing all of them
|
|
# to don't miss any topic
|
|
|
|
# Adding the topic to the topic list
|
|
topic_string = itopic.find("a", {"class": "link text-xl text-zinc-300"}).text
|
|
cleaned_topic_string = cleanString(topic_string.strip())
|
|
topic.append(cleaned_topic_string)
|
|
|
|
# Adding the url to the list of urls
|
|
link_to_clean = itopic.find("a", {"class": "link text-xl text-zinc-300"}).get("href")
|
|
|
|
href.append(link_to_clean)
|
|
|
|
# Finding the author of the topic
|
|
username_not_cleaned = itopic.find('div', {"class": "flex-grow p-2 text-justify"}).find('a').text
|
|
username_cleaned = username_not_cleaned.split("/")[-1]
|
|
author.append(cleanString(username_cleaned))
|
|
|
|
# Finding the number of views
|
|
num_views = itopic.find_all("div", {"class": "flex items-center"})[0].find("p").text
|
|
views.append(cleanString(num_views))
|
|
|
|
# Finding the number of replies
|
|
num_replies = itopic.find_all("div", {"class": "flex items-center"})[1].find("p").text
|
|
posts.append(cleanString(num_replies))
|
|
|
|
# If no information about when the topic was added, just assign "-1" to the variable
|
|
|
|
date_time_concatenated = itopic.find("p", {"class": "text-sm text-zinc-400 italic"}).text
|
|
date_time_cleaned = date_time_concatenated.replace(username_not_cleaned, "")
|
|
# creating the datetime object
|
|
date_time_array = date_time_cleaned[3:]
|
|
datetime_append = datetime.strptime(date_time_array, "%Y-%m-%d %H:%M:%S GMT")
|
|
addDate.append(datetime_append)
|
|
|
|
# print(forum)
|
|
# print(nm)
|
|
# print(board)
|
|
# print(author)
|
|
# print(topic)
|
|
# print(views)
|
|
# print(href)
|
|
# print(addDate)
|
|
# print(len(author))
|
|
# print(len(topic))
|
|
# print(len(views))
|
|
# print(len(href))
|
|
# print(len(addDate))
|
|
|
|
return organizeTopics(
|
|
forum=forum,
|
|
nm=nm,
|
|
board=board,
|
|
author=author,
|
|
topic=topic,
|
|
views=views,
|
|
posts=posts,
|
|
href=href,
|
|
addDate=addDate
|
|
)
|
|
|
|
|
|
def libre_links_parser(soup):
|
|
# Returning all links that should be visited by the Crawler
|
|
href = []
|
|
listing = soup.find_all('div', {"class": "flex-grow p-2 text-justify"})
|
|
|
|
for a in listing:
|
|
link = a.find('div', {'class': 'flex space-x-2 items-center'}).find('a').get('href')
|
|
|
|
href.append(link)
|
|
|
|
return href
|