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.
166 lines
6.2 KiB
166 lines
6.2 KiB
__author__ = 'Helium'
|
|
|
|
# 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 abyssForums_description_parser(soup):
|
|
|
|
# Fields to be parsed
|
|
|
|
topic = "-1" # 0 topic name
|
|
user = [] # 1 all users of each post
|
|
addDate = [] # 2 all dated of each post
|
|
feedback = [] # 3 all feedbacks of each vendor (this was found in just one Forum and with a number format)
|
|
status = [] # 4 all user's authority in each post such as (adm, member, dangerous)
|
|
reputation = [] # 5 all users's karma in each post (usually found as a number)
|
|
sign = [] # 6 all user's signature in each post (usually a standard message after the content of the post)
|
|
post = [] # 7 all messages of each post
|
|
interest = [] # 8 all user's interest in each post
|
|
image_user = [] # 9 all user avatars of each post
|
|
image_post = [] # 10 all first images of each post
|
|
|
|
# Finding the topic (should be just one coming from the Listing Page)
|
|
|
|
li = soup.find("div", {"class": "page-body"}).find("h2", {"class": "topic-title"})
|
|
topic = li.text.replace(",","")
|
|
topic = topic.replace("\n","")
|
|
topic = cleanString(topic.strip())
|
|
|
|
regex = re.compile('post has-profile.*')
|
|
posts = soup.find_all('div', {"class": regex})
|
|
# print(len(posts))
|
|
|
|
# For each message (post), get all the fields we are interested to:
|
|
|
|
for ipost in posts:
|
|
|
|
# Finding the author (user) of the post
|
|
author = ipost.find('a', {"class": "username"}).text
|
|
user.append(cleanString(author)) # Remember to clean the problematic characters
|
|
|
|
status.append("-1")
|
|
reputation.append("-1")
|
|
interest.append("-1")
|
|
sign.append("-1")
|
|
feedback.append("-1")
|
|
image_post.append("-1")
|
|
|
|
img = ipost.find('dl', {"class": "postprofile"}).find('img')
|
|
if img is not None:
|
|
img = img.get('src').split('base64,')[-1]
|
|
else:
|
|
img = "-1"
|
|
image_user.append(img)
|
|
|
|
image_user.append("-1")
|
|
|
|
date_time_obj = ipost.find('time').attrs
|
|
date = date_time_obj['datetime'][0:10]
|
|
time = date_time_obj['datetime'][11:19]
|
|
date_time_obj = datetime.strptime(date + " " + time, '%Y-%m-%d %H:%M:%S')
|
|
addDate.append(date_time_obj)
|
|
|
|
# Finding the post
|
|
|
|
inner = ipost.find('div', {"class": "content"})
|
|
inner = inner.text.strip()
|
|
post.append(cleanString(inner))
|
|
|
|
# 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 abyssForums_listing_parser(soup: BeautifulSoup):
|
|
|
|
|
|
nm = 0 # this variable should receive the number of topics
|
|
forum = "AbyssForum" # 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
|
|
addDate = [] # when the topic was created (difficult to find)
|
|
image_author = [] # 8 all author avatars used in each topic
|
|
|
|
# Listing and Description pages)
|
|
#finding the board
|
|
|
|
board = soup.find("h2", {"class": "forum-title"}).text
|
|
board = cleanString(board.strip())
|
|
|
|
type_of_posts = soup.find_all("li", {"class": re.compile("row bg\d")} )
|
|
for literature in type_of_posts:
|
|
title_of_post = literature.find("a", {"class": "topictitle"}).text
|
|
title_of_post = cleanString(title_of_post)
|
|
topic.append(title_of_post)
|
|
user = literature.find("div", {"class": "topic-poster responsive-hide left-box"}).find("a", {"class": "username"}).text
|
|
author.append(user)
|
|
num_post = literature.find("dd", {"class": "posts"}).text.replace("Replies","").strip()
|
|
posts.append(num_post)
|
|
num_view = literature.find("dd", {"class": "views"}).text.replace("Views","").strip()
|
|
views.append(num_view)
|
|
#if int(num_post) != 0: join the last user who posted with the author?
|
|
# reply = literature.find("dd", {"class": "lastpost"}).find("a", {"class": "username"}).text
|
|
# user.append(reply)
|
|
|
|
date_time_obj = literature.find('time').attrs
|
|
date = date_time_obj['datetime'][0:10]
|
|
time = date_time_obj['datetime'][11:19]
|
|
date_added = datetime.strptime(date + " " + time, '%Y-%m-%d %H:%M:%S')
|
|
|
|
addDate.append(date_added)
|
|
|
|
listing_href = literature.find("a", {"class": "topictitle"}).get("href")
|
|
href.append(listing_href)
|
|
|
|
image_author.append("-1")
|
|
|
|
nm = len(topic)
|
|
|
|
return organizeTopics(
|
|
forum=forum,
|
|
nm=nm,
|
|
board=board,
|
|
author=author,
|
|
topic=topic,
|
|
views=views,
|
|
posts=posts,
|
|
href=href,
|
|
addDate=addDate,
|
|
image_author=image_author
|
|
)
|
|
|
|
|
|
|
|
|
|
def abyssForum_links_parser(soup):
|
|
|
|
# Returning all links that should be visited by the Crawler
|
|
|
|
href = []
|
|
#print(soup.find('table', {"class": "tborder clear"}).find(
|
|
# 'tbody').find_all('tr', {"class": "inline_row"}))
|
|
listing = soup.find_all('dl', {"class": "row-item topic_read"})
|
|
|
|
for a in listing:
|
|
link = a.find('div', {"class": "list-inner"}).find('a').get('href')
|
|
|
|
href.append(link)
|
|
|
|
return href
|