A Parler client library in Python.
Authenticate to Parler using the web app. Extract the jst and mst session cookies.
Configure the session cookies as local environment variables.
export MST_COOKIE=foo
export JST_COOKIE=bar
Import the library.
from pyrler.core import pyrler
Pyrler logs JSON to stdout by default. To disable stdout logging use pyrler.Comment(log_stdout=False).
To log to a file use pyrler.Post(log_file='log.txt').
Both stdout and file handlers can be enabled together.
Pyrler pulls only the first page of results by default. To use pagination pass a truthy value to the follow argument.
Parler pages are indexed by startkey and each response returns a next value used to access the next page.
Let's look at a user's post history. First we find the user's id by fetching their profile.
profile = pyrler.Profile()
profile_response = profile.get_user_profile(username="AltCyberCommand")
user_id = profile_response.json().get("id")
Now we fetch their post history starting with the first page of data.
p = pyrler.Post()
r = p.get_user_posts(user_id=user_id, startkey=None, follow=True)
You can specify the endkey argument to stop pagination after a certain time. endkey must be formatted as a timestamp (as Parler returns for its prev/next keys). For example:
r = p.get_user_posts(user_id=user_id, startkey=None, endkey="2021-02-16T14:53:30.429Z_322497", follow=True)
Note that you may still receive results earlier than endkey if they are on a page whose last result is after endkey. endkey simply prevents pagination from continuing further back in time.
Wow. Much shitposting.
Take a look at pyrler/core/pyrler.py for the complete list of methods.
Methods return a dict by default and a list of dicts when follow=True.
Get a comment by its ID.
p = pyrler.Comment()
r = p.get_comment(comment_id="comment_id")
Upvote a comment.
p = pyrler.Comment()
r = p.vote_comment(comment_id="comment_id", upvote=True)
Get all comments from user.
p = pyrler.Comment()
r = p.get_user_comments(user_id="user_id")
Discover recommended news.
p = pyrler.Discover()
r = p.discover_news()
Discover recommended users.
p = pyrler.Discover()
r = p.discover_users()
Discover recommended posts.
p = pyrler.Discover()
r = p.discover_posts()
Get the user's feed.
p = pyrler.Feed()
r = p.get_feed(limit=10)
Get the user's followers.
p = pyrler.Follow()
r = p.get_followers(user_id="user_id")
Get the accounts a user is following.
p = pyrler.Follow()
r = p.get_following(user_id="user_id")
Follow a user.
p = pyrler.Follow()
r = p.follow_user(username="user_name")
Hashtag search.
p = pyrler.Hashtag()
r = p.search(search="#datascience")
Check if the current user is verified.
p = pyrler.Identity()
r = p.get_user_verification_status()
Get conversations.
p = pyrler.Messaging()
r = p.get_conversations()
Get a conversation.
p = pyrler.Messaging()
r = p.get_conversation(conversation_id="conversation_id")
Accept a conversation request.
p = pyrler.Messaging()
r = p.accept_conversation_request(conversation_id="conversation_id")
Accept a conversation request.
p = pyrler.Moderation()
r = p.approve_comment(conversation_id="conversation_id")
Filter words.
p = pyrler.Moderation()
r = p.filter_word(word="science", action="banUser)
Get news feed.
p = pyrler.News()
r = p.get_news()
Search news.
p = pyrler.News()
r = p.search_news(search="data science")
Get notifications.
p = pyrler.Notification()
r = p.get_notifications()
Delete all notifications.
p = pyrler.Notification()
r = p.delete_all_notifications()
Get a photo.
p = pyrler.Photo()
r = p.get_photo(photo_id="photo_id")
Get a post by its ID.
p = pyrler.Post()
r = p.get_post(post_id="post_id")
Get the comments on a post.
p = pyrler.Post()
r = p.get_post_comments(post_id="post_id")
Get the posts created by a user.
p = pyrler.Post()
r = p.get_user_posts(user_id="user_id")
Get the posts liked by a user identified by their user ID.
p = pyrler.Post()
r = p.get_liked_posts(user_id="user_id")
Search post by hashtag.
p = pyrler.Post()
r = p.search_by_hashtag(tag="datascience")
Submit a post. Multiple links can be submitted as a Python list. Emoji must be unicode escaped.
p = pyrler.Post()
r = p.post(post="Science Rules!", links="https://en.wikipedia.org/wiki/File:Bill_Nye_2017.jpg")
Upvote a post.
p = pyrler.Post()
r = p.upvote(post_id="post_id")
Delete post upvote.
p = pyrler.Post()
r = p.rescind_upvote(post_id="post_id")
Delete post upvote.
p = pyrler.Profile()
r = p.get_user_profile(user_id="user_id")
Get a User ID by searching their username.
p = pyrler.Profile()
r = p.get_user_profile(username="")
If you've collected some Parler post data, you can create a network of their mentions in Gephi with:
pyrler/utilities/network.py data.jsonl network_of_data.gexf
Create a network of their hashtags.
pyrler/utilities/network.py --hashtags data.jsonl network_of_data.gexf
Create a network of their echos.
pyrler/utilities/network.py --echo data.jsonl network_of_data.gexf
Additionally if you want to convert the network into a dynamic network with timeline enabled (i.e. nodes will appear and disappear according to their attributes), you can open up your GEXF file in Gephi and follow these instructions. Note that in network_of_data.gexf there is a column for "start_date" (which is the day the post was created) but none for "end_date" and that in the dynamic timeline, the nodes will appear on the screen at their start date and stay on screen forever after. For the "Time Interval creation options" pop-up in Gephi, the "Start time column" should be "start_date", the "End time column" should be empty, the "Parse dates" should be selected, and the Date format should be the last option, "dd/MM/yyyy HH:mm:ss".
Copyright 2020, Cognitive Security Collaborative
AGPLv3