Project Deep Dive: Tails

Published See discussion on Twitter

I took about a week and a half off of work starting just before Christmas and ending right after New Years, I usually use this time to dig into a meaty project (sometimes related to this blog) and this year was no exception!

I had a few other projects in mind coming into this break but didn't feel too motivated to work on them. Instead I took the break slow, mainly focusing on my own working process (revamping my Notion setup) and also getting some reading in.

However, about halfway into my time off, I found an itch that I just had to scratch.

My dad and I share new cocktail recipes back and forth fairly often[1], usually in the form of TikTok clips. We've also been selecting usually a single (or a couple) specific cocktails that we want to make for the holidays.

However, we always ask the same questions:

What ingredients do we need for that one cocktail you wanted to make?

What was the recipe for that special cocktail again?

This year also, we made one of our favorites from last year (The Polar Empress) for my grandparents as well, and the next day my Grandma wanted to get the recipe for the drink.

Since she had asked about it, and since both of the above questions are common enough, I felt like it'd be interesting to build a quick and dirty web app to store cocktail recipes that we can then refer to later[2].

And so, I started hacking on Tails.

๐Ÿšง Be warned, there's still a ton of things I need to get properly setup with this version of the service, but feel free to mess around and report any bugs here. ๐Ÿšง

What is Tails?

Tails is a very basic CRUD[3] web app for saving and sharing cocktail recipes. That's it at the moment, no social features, no fancy whiz-bang features, just saving recipes!

It has a very basic version of auth/user accounts at the moment, allows you to add a recipe (with a title, description, steps, and ingredients), and also supports sharing those recipes with universal links (viewable to those without an account on Tails).

How does it work?

Similar to microfibre, Tails is built with two primary layers:

  • API
  • Frontend Web Client

The API is built on Cloudflare Workers (using Cloudflare D1 as the storage solution, honojs to handle routing, and lucia for auth), and the frontend web client is built with Next.js and deployed to Vercel.

The client interactions are all for the most part plain old forms using React Server Components and Server Actions. The Next.js server then interacts with the API hosted on Cloudflare.

The code is fully open source and available here: hamlim/tails, feel free to peruse the source and open issues for questions/comments/contributions.

If you read my microfibre deep dive, this tech stack probably sounds pretty familiar! I've been really enjoying Cloudflare's platform offerings, and I'm particularly interested in exploring Cloudflare R2 storage soon as well!

The Future

There's still a ton of things I want to build into the service, the first of which might be building up user profiles a bit more (so you can see who shared what cocktails) - however I don't think there's that much future for a social aspect built into the site!


Footnotes:

[1] - Ok ok, it's usually me sending him cocktail recipes, but still
[2] - Now I know you're saying "but matt, there's a dozen such services out there for this use case, why build yet another one?" and to that I say - because I wanted to!
[3] - At the moment, it doesn't even support updating or deleting recipes! I'll have to add that in the future!