This repository is a base project to create Telegram's bots using AWS and python-telegram-bot.
This project has two lambda functions with their Api Gateway configuration:
- /webhook: This endpoint is to catch the webhooks that come from Telegram and respond to the message.
- /register_webhook: This endpoint configures the bot in Telegram. For example, in this endpoint, the bot configures what commands it has.
The first thing that you need to do is to clone this repository and change some values in the serverles.yml and package.json files. In the future, we will have a Cookiecutter Templete
The project has the following folders:
- handlers: In this folder will have the code to handle the different requests that come from Telegram. In the project, you can see the Start command handler.
- models: In this folder, we can implement the code to access
dynamodb. Now the project has the code to manage the user table. - utils: The code for supporting and not related to the current business logic.
- It also has the two lambdas, webhook, and register_webhook.
The project already has a command to start the bot.
- Create the file
app/handlers/start.py
from string import Template
from telegram import Update
from app.utils.callback_context import CallbackContext
START_TEXT = """
Hello "$name"
"""
def start_query(update: Update, context: CallbackContext):
user_model = context.user_db # this is the logic to get or create the user from database.
update.effective_message.edit_text(text=Template(START_TEXT).substitute(name=user_model["complete_name"]))
- Next step is to register the command in the configure method in the file
app/config.py
def configure_handlers(dispatcher):
dispatcher.add_handler(CommandHandler(command="start", callback=start_command))
- Finally, to test the bot in your local machine you can run this command.
python manage.py
Note: To work the application needs to have the dynamodb in the AWS
The project has the file .devcontainer/devcontainer.json to execute the application into VScode and build the docker containers.
To work with Serverless Framework you need to install the NPM modules and after you can use the commands registered in the package.json