In Jinaga, information is stored in facts. Facts are immutable. That means that once you create one, you don't change it. Instead, you just create more facts.
Create a simple top-level fact.
const tagReact = await j.fact({
type: "Blog.Tag",
name: "React"
});
Create a fact with a single predecessor. Why is this called a predecessor? Because it comes before. We have to have a person before they can write a blog post.
const post = await j.fact({
type: "Blog.Post",
created: new Date(), // Will be converted to an ISO string, such as "2018-12-23T22:46:02.487Z".
author: person // Where person is the result of a previous j.fact.
});
Create a fact with multiple predecessors. Just put them in an array.
await j.fact({
type: "Blog.Post.Tags",
post: post,
tags: [tagReact, tagCss, tagMicroFrontends]
});
You can specify the predecessors inline. The predecessor facts will be persisted first. However, persistence is not guaranteed to be atomic.
await j.fact({
type: "Blog.Post.Tags",
post: {
type: "Blog.Post",
created: new Date(),
author: person
},
tags: [{
type: "Blog.Tag",
name: "React"
}, {
type: "Blog.Tag",
name: "CSS"
}, {
type: "Blog.Tag",
name: "Micro-Frontends"
}]
});
You may be feeling that Jinaga facts are upside down. Typically, a JSON object contains its children. But a Jinaga fact contains its parent. What's going on with that?
This all stems from the fact that Jinaga facts are immutable. You cannot change a fact. If a fact contained an array of children, then you would never be able to add another child. And so the relationship has to be flipped. A child knows its parent, because that parent relationship never changes.
To find all of the children of a fact, you need to write a query. By the way, we call these children successors, as you will soon see.