It was great to work and make this project, Tuti is now deprecated https://divyendusingh.com/farewell-tuti
Note: this was a private project, this is made public. Some CI/CD scripts might be outdated. This is only for reference and might not be deployable as is.
Built with telegraf, Telegram Bot API, AWS and ❤️
docker build . -t lingoparrot
docker tag lingoparrot:latest <docker remote>
docker run -p 3000:3000 --env-file ./.env.local.docker lingoparrot:latest
docker push <docker remote>
cd carrd-language-learners-20220623
npx serve
We use a specific dev/production setup. For all the relevant commands, yarn scripts have two variants like:
yarn run set-webhookyarn run set-webhook-production
The 1st one uses .env and the 2nd one uses .env.production. With sensible defaults, I believe that this setup is most convenient for bot development and yields least mistakes. Open to feedback.
Copy .env_sample file to .env (and .env.production for a production setup - details later) and fill the required values.
Run the following commands to start receiving @<bot-name> requests on your local machine.
yarn dev
- runs the project locally (via
sls offline start, supports hot realoading). - creates a tunnel (languagelearnersclub.localtunnel.me) from local to the internet.
- points the bot to local development version (uses
.envand localtunnel url).
yarn run watch- to watch and compile TS to JS.
Generally, I have two versions of the bot i.e. development and production. I point my dev bot to my local setup using the above steps.
yarn run deploy - deploys to Lambda using parameters from .env
yarn run deploy-production - deploys to Lambda using parameters from .env.production
- https://telegraf.js.org/#/
- https://core.telegram.org/bots/api#inlinequeryresult
- https://aws.amazon.com/translate/pricing/
- https://docs.aws.amazon.com/translate/latest/dg/API_TranslateText.html
- https://docs.aws.amazon.com/translate/latest/dg/pairs.html
- https://ibb.co/rx3tsGD (Pricing for a conversational app)
- https://pair-code.github.io/lit/