mdaizovi / sk8_bot

Sk8DateBot backend

Sk8DateBot is a Telegram Bot used for blasting skate dates from a Telegram group chat to a Telegram Channel. It is a built with the Django web framework in Python programming language.

How to Contribute

This is an open source project. Please continue this README to get set your local env set up. You can then contribute in one of 2 ways:

• Fork this repo, and then make Pull Requests from your fork of this repo.
• Clone this repo and make Merge Requests from a feature branch.

New features will only be accepted if they include adequate tests, and testing instructions for QA (ie tell me what you did and how it should work so I can test it). Everything will be formatted by Black before being merged (I'll take care of that, don't worry).

Set Up Your Local Development Environment

From here, words that looks like this are commands you are meant to type somewhere, usually into a terminal.

• Install python on your local machine

  • Google it or go here.

• Install ngrok on your local machine

  • Google it or go here.

  NOTE this installation will only work on a mac >= 10.13, and on a mac, for me it was easier to install with brew install ngrok/ngrok/ngrok.

  • Make an account and then follow these instructions to conect my account.

• From the terminal navigate to the directory you want this code to live and clone this repo by typing the following into a terminal:

  git clone https://github.com/mdaizovi/sk8_bot

• cd sk8_bot to get into repo. cd is how you change folders from the command line.

• Copy sk8_bot/.env.example to .env. You can do this by typing cp sk8_bot/.env.example sk8_bot/.env in the terminal.

• Make a Virtual Environment:

  • If you have python 3.9 on your system, you can make a virtual environment like so:
    • python -m venv venv
    • If that doesn't work, you might need to type python3 -m venv venv

• Activate virtual environment:

  source venv/bin/activate

• Install backend requirements (make sure you are in your virtual environment first):

  pip install -r requirements.txt

  • If that doesn't work, you might need to type pip3 install -r requirements.txt NOTE TO THE FUTURE: if this does not work, check to ensure that cffi is not pinned to a specific verison, in requirements.txt As of right now cffi is causing problems between versions of python 3.9 versus 3.10

• Run initial migrations, to prepare database:

  python manage.py migrate

  • If that doesn't work, you might need to type python3 manage.py migrate

• Make yourself a super user:

  python manage.py createsuperuser (or python3 manage.py createsuperuser) so on and so forth for all other python commands

  • You will be prompted by Django to create a super user, and you can use these credentials to log into the Admin interface (more on that later. We're going to take a pause on this subject here and come back to it.)

• Import starter data into the database:

  python manage.py loaddata fixtures/db.json

• Get a Telegram API ID and Save Credenitals

  • Get a Telegram API ID here
  • From there, choose “API development tools”. If it is your first time doing so, it will ask you for an app name and a short name, you can change both of them later if you need to. Submit the form when you have completed it.
  • You will then see the api_id and api_hash for your app. These are unique to your app, and not revocable. Put those in the .env file that you copied. Replace "dummy" for TELEGRAM_APP_API_ID and TELEGRAM_APP_API_HASH with your new api_id and api_hash. Depending on your system, you might not be able to see the .env file. You will still be able to access it by typing nano sk8_bot/.env into the terminal, or by using an IDE, like PyCharm, what will show it in the directory. Don't forget to save your changes! Note, nano is a pain in the ass to exitm you need to Ctrl+X

• Create Your Own Test instance of the Bot

  • While logged in to your account on Telegram, make a bot

  • BotFather will give you an API Key. Put that in your .env file under TELEGRAM_BOT_TOKEN

  • Set the privacy of your Bot so it can read chat messages. To do this, in your conversation with BotFather type

    /setprivacy

    and then after BotFather has you choose your bot, simple type

    Disable

  • Create a group chat and a channel. Add your bot to both, make the Bot an admin of the channel, and give it access to post messages.

• Run ngrok by opening another terminal and typing:

  ngrok http 8000

  • If you're on Windows you may need to navigate in the terminal to the directory where you installed it, and typing:

  ./ngrok http 8000

• Create a Test Group and Channel for your Test Bot and save the chat ids in the database

  • Set up Telegram webhooks to your local instance with the following command, replacing "<web_hook_address>" with the Forwarding address in the ngrok terminal (it will probably be something like 'https://001f796288c4.ngrok.io' but with different numbers/letters ) Make this command from a terminal in this repo's home directory:

    python manage.py set_hook --address <web_hook_address>

  • Run the Django server:

    python manage.py runserver

  • Go back to the conversation you made with your bot. Say something. Afterwards, you should see a message in the terminal that looks something like:

    {'update_id': 506937115, 'message': {'message_id': 2, 'from': {'id': 5459064228, 'is_bot': False, 'first_name': 'Mic', 'last_name': 'Test'}, 'chat': {'id': -800841577, 'title': 'Test Bot Chat', 'type': 'group', 'all_members_are_administrators': True}, 'date': 1664634193, 'text': 'Test comment'}}
    

  • Navigate to the Admin by typing http://127.0.0.1:8000/admin/ into your browser. On the login screen, enter the username and password you made awhile go, when you typed python manage.py createsuperuser (I told you we'd come back to this.)

  • Go to http://127.0.0.1:8000/admin/webhooks_consumer/inputsource/. These are your input sources; basically conversations where the bot commands come from. Click on the one named "Bot Convo". Change its chat ID to whatever chat id you see in your local terminal. ... 'chat': {'id': -800841577, ...

  • Do the same for the channel. Make a test comment in the channel. Then navigate to the OutputChannels in the admin. Select the "Sk8Dates" one and replace its "Channel id" with what you see in your terminal under chat id ... 'chat': {'id': -1001845844181, ...

  • You're finally done with setup! Go to the first conversation and type in your favorite command. /doggo, /kitty, etc.

Running Your Local Development Environment While You Work

• Run ngrok by navigating in the terminal to the directory where you installed it, and typing:

`./ngrok http 8000`

• In another terminal, activate your virtual environment:

  source venv/bin/activate

• Set up Telegram webhooks to your local instance with the following command, replacing "<web_hook_address>" with the Forwarding address in the ngrok terminal (it will probably be something like 'https://001f796288c4.ngrok.io' but with different numbers/letters ) Make this command from a terminal in this repo's home directory:

  python manage.py set_hook --address <web_hook_address>

• Run the local django server

  python manage.py runserver

How Does it Work?

How Messages are Received

Posting messages works both from a Telegram chat as well as from the CiB Slack workspace. In sk8_bot/webhooks_consumer/views.py , there are 2 classes:

• TelegramBotView
• SlackBotView

I bet you can guess which one handles which.

For Telegram, every time a message is posted in a group that the skatebot is in, that message gets sent to this backend as JSON, looking something like this:

{'update_id': 944746271, 
 'message': {
   'message_id': 32726, 
   'from': 
     {'id': 531519657, 'is_bot': False, 'first_name': 'SABA', 'username': 'SABAloops'
     }, 
   'chat': {
     'id': -1001258758865, 'title': 'Berlin Quadsk8ing on Ramps/Bowls/Skateparks Chatter', 'type': 'supergroup'
    },
   'date': 1658575957, 'text': 'Who is in the parade?! :) we’re coming!! :)'
  }
}

In TelegramBotView there is a function called post, which parses this json. It checks to see if the message contains a command (if it starts with a /) and of so, if takes the word (for example, "/sk8") and checks the database to see if there is a saved commant for that command and chat combination. This will be explained a little more below, in Database Models.

In sk8_bot/webhooks_consumer/factory.py you'll see several simple functions, like

• _get_compliment
• _get_doggo etc Those are the functions that dictate what happens for their respective commands that trigger them. sk8 is not there because all it does is take input and put it somewhere else, which willl make more sense in Database Models. Thwe only thing from "/sk8" command that is in this file is the function _get_broadcast_message, which formats the message posted to the skate bulletin and CiB Slack.

Database Models

In sk8_bot/webhooks_consumer/models.py there are definitions of the database models (aka tables).

InputSource is where the messages come from, such as a Telegram chat or Slack channel. OutputChannelIs the same, but they are generally different Telegram chats/Slack channels.

For example, for a /sk8 bulletin to go from the "Berlin Quadsk8ing on Ramps/Bowls/Skateparks Chatter" to the CiB Slack channel and skate bulletin channel, the chat where the message originated is the InputSource, then the Slack channel and Sk8Date Bulletin are both OutputChannel. Because something (pictures of dogs, cats, or sloths) gets posted back to the "Berlin Quadsk8ing on Ramps/Bowls/Skateparks Chatter", it is also an OutputChannel. The "Berlin Skate Crew" chat doesn't get animal pictures, so it is not an OutputChannel, just an InputSource.

BotOutput is what the Bot actually returns. For exmample, for the simple command /doggo that returns a picture of a dog in "Berlin Quadsk8ing on Ramps/Bowls/Skateparks Chatter", it looks like this:

• output_function is _get_doggo
• output_platform is Telegram
• output_channel is the same chat the command came from, "Berlin Quadsk8ing on Ramps/Bowls/Skateparks Chatter"

BotAction brings this all together.

• command the actual text that triggers this sequence, like "sk8" or "doggo"
• content_required True or False of whether content is required. For example, /sk8 requires text or else it does nothing, but doggo doesn't.
• outputis an instance of BotOutput

FAQ

Everything should work locally but it doesn't. Why?

• Try restarting ngrok and re-running the command to set the webhook.