Christmas is coming up and I took an early vacation to spend some time on my side projects. One of my precious hobbies is home brewing which I get to do every so often. Me and my friend like to experiment with different recipes that we make ourselves. For recipes I use a tool called Brewfather.
Brewfather is a great tool for creating and managing your recipes and brew sessions. As a home brewer I have often found help in reading and learning from other people’s recipes. So I wanted a way to share my recipes with the home brewing world. There is a function in Brewfather to share individual recipes via links, but I wanted to have a place where I could share all my recipes in one place. As I already have this site I thought it would be a good idea to add a section for my recipes here.
Brewfather API
Fortunately for me Brewfather has an official API that I can use to fetch my recipes. The API is provided as a REST-like API with endpoints for fetching recipes, batches, and inventory. So I set out to build a helper script that would download each recipe and save it as a JSON file. Because I’m in love with Clojure from my old job, I decided to use babashka to build it.
The script is pretty simple, it essentially boils down to a few steps:
- Define some global variables
- Fetch the API secret using the 1Password CLI
- Fetch the list of recipes not already downloaded
- Fetch each recipe and save it as a JSON file
The main function then looks like this:
(defn main [& args]
(op-available?)
(as-> (op-secret!) $
(basic-auth client $)
(list-remote-recipes! $ (last-recipe!))
(map :_id $)
(doseq [id $]
(println "Fetcing recipe" id)
(-> (fetch-remote-recipe! (basic-auth client (op-secret!)) id)
(write-recipe!)))))
I keep the secret stored in 1Password and fetch it directly from 1Password using their CLI to avoid leaking it. Leaking the secret can happen if you pass it as an environment variable or as a command line argument. Brewfather uses basic authentication so I create a basic auth header using the hard-coded client id and the secret. Then I fetch the list of recipes that I haven’t already downloaded. Because I order the recipes by date I can use the last recipe I downloaded as a starting point. I then map the list of recipes to their ids and fetch each recipe and save it as a JSON file.
Astro
Now that I have all my recipes downloaded I need to build a page to display them.
Astro has great support for data collections.
This allows me to define a collection type for my beer recipes and strongly type them via zod schemas.
So I created a config.ts
file in src/content/config.ts
and defined my collection type:
// Sub schemas elided for brevity
const beerRecipesCollection = defineCollection({
type: "data",
schema: z.object({
name: z.string(),
slug: z.string(),
_timestamp_ms: z.number(),
abv: z.number(),
ibu: z.number(),
color: z.number(),
fg: z.number(),
og: z.number(),
batchSize: z.number(), // liters
style: style.optional().nullable(),
mash: mashAndFerment,
fermentables: z.array(fermentable),
yeasts: z.array(yeast),
hops: z.array(hop),
equipment
}),
});
export const collections = {
beers: beerRecipesCollection,
}
So now I have a collection of beer recipes that I can use in my pages. I then defined two pages, one for the list of recipes and one for the individual recipes.
List page
The list page, /src/pages/beers.astro
, is pretty simple.
It just lists all the recipes in the collection.
---
import { getCollection } from "astro:content";
import BaseLayout from "../layouts/BaseLayout.astro";
import BeerGlass from "../components/BeerGlass.astro";
import { SITE_TITLE, SITE_DESCRIPTION } from "../config";
// Use Astro.glob() to fetch all posts, and then sort them by date.
const recipes = (await getCollection("beers")).sort((a, b) => {
return b.data._timestamp_ms - a.data._timestamp_ms;
});
---
{/* Simplified layout example */}
<BaseLayout title={SITE_TITLE} description={SITE_DESCRIPTION}>
<h1>Beer recipes</h1>
<ul>
{recipes.map((recipe) => (
<li>
<a href={`/recipes/${recipe.data.slug}`}>{recipe.data.name}</a>
</li>
))}
</ul>
</BaseLayout>
Recipe page
The recipe page, /src/pages/beers/[slug].astro
, is a bit more involved.
Astro requires you to define a getStaticPaths
function that returns a list of paths to render.
For each path you return you specify any parameters and properties that should be passed to the page.
---
import { getCollection } from 'astro:content';
import { SITE_TITLE, SITE_DESCRIPTION } from '../../config';
import BaseLayout from '../../layouts/BaseLayout.astro';
import BeerGlass from '../../components/BeerGlass.astro';
export async function getStaticPaths() {
const beers = await getCollection('beers')
return beers.map(beer => {
return {
params: {
slug: beer.data.slug
},
props: {
beer: beer.data
}
}
})
}
const { slug } = Astro.params;
const { beer } = Astro.props;
---
{/* Simplified layout */}
<BaseLayout title={SITE_TITLE} description={SITE_DESCRIPTION}>
<h1>{beer.name}</h1>
<BeerGlass beer={beer} />
</BaseLayout>
Some design choices
For the beer recipe design I wanted to keep it simple like the rest of the site. However, I wanted to add some color to the page to make it stand out a bit. Also it is common to show which color the beer will have based on the recipe. I ended up using ChatGPT to help generate a simple vector like image of a tulip style beer glass. This was then vectorized using Vectorizer.ai to convert the webp raster image to an SVG. The result from that was a promising SVG with only three different colors used for the beer liquid. I then edited away artifacts using Krita which has some simple SVG support.
From the SVG I then created an Astro component which accepts a beer color (SRM). It then calculates the primary beer color in RGB and two darker shades of the same color. I do this by simply incrementing the SRM value by 2 and 4 respectively and using the same converter as with the base color.
The end result turned out to be pretty good, but don’t take my word for it, see for yourself: