LariskaBot

LariskaBot - telegram-bot (Python, aiogram)

Made with Support Open Source

Telegram bot for chat @OldCodersClub

The bot uses priorities when searching and generating answers for users:

• filters (handler) aiogram (highest priority)

• ready-made responses for specific users (greeting at the user’s first message, once a day)

• own database of ready-made answers using FuzzyWuzzy (Levenshtein distance) for fuzzy matching of questions and answers (set to 80% match)

• OpenAI API (only works if you contact the bot by name)

Any filter located below works only if all those located above have been skipped.

Running in Docker

Create and run a container:

 $ export BOT_TOKEN=<BOT_TOKEN>$ export AI_KEY=<AI_KEY>$ export VCHAT_ID=<VCHAT_ID>$ export DCHAT_ID=<DCHAT_ID>$ export SCHAT_ID=<SCHAT_ID>$ docker-compose up -d

Stop a running container:

 $ docker-compose stop

Start a stopped container:

 $ docker-compose start

Stop and remove container and network:

 $ docker-compose down

Remove docker image:

 $ docker rmi lariska_bot

Clear logs:

 $ sudo rm -rf logs/*

Blank template for deployment

Variables

• BOT_TOKEN - Telegram bot token

• CHAT_ID - ID of the chat in which the bot is allowed to work.

• AI_KEY - OpenAI API token

• DOMAIN - URL containing the application name. After deployment, go to the project settings ( Settings ) and copy the domain from the Domains section. It should be similar to worker-production-XXXX.up.railway.app . This will be the value for the DOMAIN variable.