-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path2stacks-get-thread-posts.py
226 lines (206 loc) · 12.2 KB
/
2stacks-get-thread-posts.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
import requests
import random
import string
import json
import re
import config
import helpers
import boto3
import os
from bs4 import BeautifulSoup
#temp
# import logging
# logger = logging.getLogger()
# logger.setLevel(logging.INFO)
def lambda_handler(event, context):
for record in event['Records']:
callback_url = record['messageAttributes']['callback_url']['stringValue']
wd_thread_id = record['messageAttributes']['thread_id']['stringValue']
wikidot_site = record['messageAttributes']['wikidot_site']['stringValue']
data = {'t': wd_thread_id, 'moduleName': 'forum/ForumViewThreadModule'}
haystack = helpers.fetch(data, wikidot_site)
# Do some stuff with the base thread.
try:
soup = BeautifulSoup(haystack, 'html.parser')
except TypeError: # NoneType, it gone.
return False # Send this to SCUTTLE.
titleblock = soup.find("div", {"class": "forum-breadcrumbs"})
forum = int(re.search('(?:\/forum\/c-)(\d*)', str(titleblock)).group(1))
title = re.search('(?:» (?!<))(.*)', str(titleblock)).group(1)
descriptionblock = soup.find("div", {"class": "description-block well"})
# Get the subtitle, which is a surprising amount of effort.
if wikidot_site == 'scp-ru': # SCP-RU
subtitle = re.findall('(?:<\/div>)(?:\s*<div class="head">Кратко:<\/div>){0,1}([\s\S]*)(?:<\/div>)', str(descriptionblock), re.MULTILINE)
elif wikidot_site == 'lafundacionscp': # SCP-ES
subtitle = re.findall('(?:<\/div>)(?:\s*<div class="head">Resumen:<\/div>){0,1}([\s\S]*)(?:<\/div>)', str(descriptionblock), re.MULTILINE)
elif wikidot_site == 'fondationscp': # SCP-FR
subtitle = re.findall('(?:<\/div>)(?:\s*<div class="head">Résumé:<\/div>){0,1}([\s\S]*)(?:<\/div>)', str(descriptionblock), re.MULTILINE)
elif wikidot_site == 'scp-wiki-de': # SCP-DE
subtitle = re.findall('(?:<\/div>)(?:\s*<div class="head">Beschreibung:<\/div>){0,1}([\s\S]*)(?:<\/div>)', str(descriptionblock), re.MULTILINE)
else: #SCP-EN and English-speaking wikis.
subtitle = re.findall('(?:<\/div>)(?:\s*<div class="head">Summary:<\/div>){0,1}([\s\S]*)(?:<\/div>)', str(descriptionblock), re.MULTILINE)
subtitle = ''.join(subtitle)
subtitle = subtitle.replace('\n','').replace('\t','') # These are artifacts of scraping HTML and not valid in subtitles.
if len(subtitle) is 0:
subtitle = None
# Get the creation timestamp for convenience in sorting later.
created_timestamp = int(re.search('(?:odate time_)(\d*)', str(descriptionblock)).group(1))
# Get the OP of the thread. This is Wikidot for a per-page discussion thread or a user id otherwise.
attribution = descriptionblock.find("span", {"class": "printuser"})
# logger.info(attribution)
if attribution.string == "Wikidot":
op_user_id = 0
op_username = "Wikidot"
else:
try:
op_user_id = int(re.search('(?:userInfo\()(\d*)', str(attribution)).group(1))
op_username = attribution.text
except AttributeError:
try:
# Deleted Accounts
op_user_id = int(re.search('(?:data-id=\")(\d*)', str(attribution)).group(1))
op_username = "Deleted Account (" + str(op_user_id) + ")"
except AttributeError:
try:
# Anonymous Accounts
op_user_id = 0
op_username = "Anonymous User (" + str(re.search('(?:anonymousUserInfo\(\')([\d\.]*)(?:\'\); return false;\"><)', str(attribution)).group(1))
except AttributeError:
# Guest Accounts
op_user_id = 0
op_username = str(re.search('(?:</a>)([^<]*)', str(attribution)).group(1))
# What we should have back is HTML laying out a page of forum comments.
# logger.info('haystack returned:')
# logger.info(haystack)
# First, let's determine if there are multiple pages.
try:
maxpages = re.search('(?:<span class="pager-no">page \d* of )(\d*)', haystack).group(1)
maxpages = int(maxpages)
except AttributeError: # NoneType means the pager is absent, meaning there's only one page of comments. This is okay.
maxpages = 1
# else: # wtf?
# logger.info('maxpages returned:')
# logger.info(maxpages)
# raise Exception('we hit a weird thing with the maxpages, aborting')
# logger.info('maxpages returned:')
# logger.info(maxpages)
# Let's handle things the same way for one page or many.
for page in range(maxpages):
actualpage = page + 1
# logger.info('On Page ' + str(actualpage))
innerpayload = {}
haystack = get_thread_page(thread=wd_thread_id, page=actualpage, wikidot_site=wikidot_site) # I'm too lazy to not just increment this range by one to make it work.
soup = BeautifulSoup(haystack.replace("\\","")[2:], 'html.parser')
posts = soup.find_all("div", id=re.compile("(fpc-)"))
# logger.info('posts:')
# logger.info(len(posts))
for idx, post in enumerate(posts):
wd_post_id = int(re.search('(?:<div class="post" id="post-)(\d*)', str(post)).group(1))
# logger.info("Post " + str(idx) + ", ID " + str(wd_post_id))
subject = re.search('(?:<div class="title" id="post-title-\d*">\s*)([^\n]*)', str(post)).group(1)
# On a blank subject this returns as "</div>"
if subject == "</div>":
subject = None
try:
username = re.search('(?:return false;">)([^<]*)(?:<\/a><\/span>,)', str(post)).group(1)
wd_user_id = int(re.search('(?:www\.wikidot\.com\/userkarma.php\?u=)([^\)]*)', str(post)).group(1))
except AttributeError: #NoneType, deleted user.
# logger.info('thread:')
# logger.info(wd_thread_id)
# logger.info('post:')
# logger.info(wd_post_id)
try:
wd_user_id = int(re.search('(?:data-id=")(\d*)', str(post)).group(1))
username = "Deleted Account " + str(wd_user_id)
except AttributeError: #NoneType, anonymous user!
try:
wd_user_id = 0
username = "Anonymous User (" + str(re.search('(?:anonymousUserInfo\(\\\')([\d\.]*)', str(post)).group(1)) + ")"
except AttributeError: # One last NoneType, GUEST user holy crap.
# logger.info(str(post))
try:
username = re.search('(?:alt=""/></a>)([^>]*)(?:</span>,)', str(post)).group(1)
wd_user_id = 0
except AttributeError: # This is getting ridiculous. More guest account types.
try:
# logger.info(str(post))
username = re.search('(?:&default=http:\/\/www.wikidot.com/common--images/avatars/default/a16.png&size=16"\/><\/a>)([^>]*)(?:<\/span>,)', str(post)).group(1)
wd_user_id = 0
except AttributeError:
# Guest with a URL in their name
wd_user_id = 0
tempusername = re.search('(?:rel=\"nofollow\">)([^<]*)(?:<\/a> \(guest\))', str(post)).group(1)
username = tempusername + " (guest"
post_created_at = int(re.search('(?:<span class="odate time_)([^\s]*)', str(post)).group(1))
content = post.find("div", {"class": "content"})
body = ''.join(str(item) for item in content.contents)
body = body[1:-1] # Wikidot pads the text with a \n on both sides, which the author didn't write.
try:
if post.parent['id'] == 'thread-container-posts':
# Top-level response
parent = 0
else:
# 'id' will look like fpc-12345678, take a slice of the string
# logger.info('parent:' + post.parent['id'])
parent = int(post.parent['id'][4:])
except KeyError: # We're at the root.
parent = 0
changespresent = post.find("div", {"class": "revisions"})
if changespresent is not None:
# This post was edited, send along a list of revisions and let those get picked up in a different routine.
# We're guaranteed at least two entries in here.
changes = re.findall('(?:showRevision\(event, )(\d*)', str(changespresent))
else:
changes = False
innerpayload[idx]={"wd_post_id": wd_post_id, "wd_user_id": wd_user_id,
"parent_id": parent, "subject": subject, "username": username, "timestamp": post_created_at,
"changes": changes, "text": body}
# logger.info('wd_post_id is a: ')
# logger.info(type(wd_post_id))
# logger.info('wd_user_id is a ')
# logger.info(type(wd_user_id))
# logger.info('parent_id is a ')
# logger.info(type(parent))
# logger.info('subject is a ')
# logger.info(type(subject))
# logger.info('username is a ')
# logger.info(type(username))
# logger.info('timestamp is a ')
# logger.info(type(post_created_at))
# logger.info('changes is a ')
# logger.info(type(changes))
# logger.info('text is a ')
# logger.info(type(body))
# While we could wait and send one big payload, that's a risky proposition on threads with lots of posts so let's not.
# logger.info('out of the loop for a single page')
# Wrap the payload and send it, SCUTTLE can sort out posts it already has.
outerpayload = {"wd_thread_id": int(wd_thread_id), "wd_forum_id": forum,
"wd_user_id": op_user_id, "wd_username": op_username, "title": title,
"subtitle": subtitle, "created_at": created_timestamp, "posts": innerpayload}
# logger.info('wd_thread_id is a: ')
# logger.info(type(wd_thread_id))
# logger.info('wd_forum_id is a ')
# logger.info(type(forum))
# logger.info('wd_user_id is a ')
# logger.info(type(wd_user_id))
# logger.info('wd_username is a ')
# logger.info(type(op_username))
# logger.info('title is a ')
# logger.info(type(title))
# logger.info('subtitle is a ')
# logger.info(type(subtitle))
# logger.info('created_at is a ')
# logger.info(type(created_timestamp))
# logger.info('posts is a ')
# logger.info(type(innerpayload))
# Send everything to SCUTTLE
output = json.dumps(outerpayload)
headers = {"Authorization": "Bearer " + config.scuttle_token, "Content-Type": "application/json"}
r = requests.put(callback_url + '/2stacks/thread/posts', data=output, headers=headers)
# logger.info('Made a SCUTTLE Request!')
# logger.info('DATA: ')
# logger.info(outerpayload)
return {"job": "complete"}
def get_thread_page(thread: int, page: int, wikidot_site: str):
data = {'t': thread, 'moduleName': 'forum/ForumViewThreadPostsModule', 'pageNo': page}
return helpers.fetch(data, wikidot_site)