This is a Sanity Studio v3 plugin.
Plugin for visually organizing documents as hierarchies in the Sanity studio. Applications include:
- Tables of content - such as a book's sections and chapters
- Navigational structure & menus - a website mega-menu with multiple levels, for example
- Taxonomy inheritance - "Carbs is a parent of Legumes which is a parent of Beans"
If you're looking for a way to order documents on a flat list, refer to @sanity/orderable-document-list.
# From the root of your sanity project
npm i @sanity/hierarchical-document-list
// sanity.config.js
import {defineConfig} from 'sanity'
import {hierarchicalDocumentList, hierarchyTree} from '@sanity/hierarchical-document-list'
export default defineConfig({
// ...
plugins: [hierarchicalDocumentList()],
schema: {
types: [
//...,
hierarchyTree
]
}
})
π‘ To learn about custom desk structures, refer to the Structure Builder docs.
// sanity.config.ts
import {defineConfig} from 'sanity'
import {deskTool} from 'sanity/desk'
import {createDeskHierarchy, hierarchicalDocumentList, hierarchyTree} from '@sanity/hierarchical-document-list'
export default defineConfig({
// ...
plugins: [
deskTool({
// NOTE: I'n V3 you MUST pass S and Context along to createDeskHierarchy as props
structure: (S, context) => S.list()
.title('Content')
.items([
...S.documentTypeListItems(), // or whatever other structure you have
createDeskHierarchy({
//prop drill S and context:
S,
context,
//configure plugin
title: 'Main table of contents',
// The hierarchy will be stored in this document ID π
documentId: 'main-table-of-contents',
// Document types editors should be able to include in the hierarchy
referenceTo: ['site.page', 'site.post', 'docs.article', 'social.youtubeVideo'],
// β Optional: provide filters and/or parameters for narrowing which documents can be added
referenceOptions: {
filter: 'status in $acceptedStatuses',
filterParams: {
acceptedStatuses: ['published', 'approved']
}
},
// β Optional: limit the depth of your hierarachies
maxDepth: 3,
// β Optional: subarray of referenceTo, when it should not be possible to create new types from all referenceTo types
creatableTypes: ['site.page']
})
])
}),
hierarchicalDocumentList()
],
schema: {
types: [
//...,
hierarchyTree
]
}
})
The hierarchical data is stored in a centralized document with the documentId
of your choosing. As compared to storing parent/child relationships in each individual document in the hierarchy, this makes it easier to implement different hierarchies for the same content according to the context.
This approach also simplifies querying the full structure - as you'll see in querying data below.
Keep in mind that this specified document is live-edited, meaning it has no draft and every change by editors will directly affect its published version.
Instead of requiring editors to manually add items one-by-one, the plugin will create a GROQ query that matches all documents with a _type
in the referenceTo
option you specify, that also match the optional referenceOptions.filter
. From these documents, editors are able to drag, nest and re-order them at will from the "Add more items" list.
If a document in the tree doesn't match the filters set, it'll keep existing in the data. This can happen if the document has a new, unfitting value, the configuration changed or it was deleted. Although the tree will still be publishable, editors will get a warning and won't be able to drag these faulty entries around.
The plugin stores flat arrays which represent your hierarchical data through parent
keys. Here's an example of one top-level item with one child:
[
{
"_key": "741b9edde2ba",
"_type": "hierarchy.node",
"value": {
"reference": {
"_ref": "75c47994-e6bb-487a-b8c9-b283f2436031",
"_type": "reference",
"_weak": true // This plugin includes weak references by default
},
"docType": "docs.article"
}
// no `parent`, this item is top-level
},
{
"_key": "f92eaeec96f7",
"_type": "hierarchy.node",
"value": {
"reference": {
"_ref": "7ad60a02-5d6e-47d8-92e2-6724cc130058",
"_type": "reference",
"_weak": true
},
"docType": "site.post"
},
// The `parent` property points to the _key of the parent node where this one is nested
"parent": "741b9edde2ba"
}
]
π If using GraphQL, refer to Usage with GraphQL.
From the the above, we know how to expand referenced documents in GROQ:
*[_id == "main-table-of-contents"][0]{
tree[] {
// Make sure you include each item's _key and parent
_key,
parent,
// "Expand" the reference to the node
value {
reference->{
// Get whatever property you need from your documents
title,
slug,
}
}
}
}
The query above will then need to be converted from flat data to a tree. Refer to Using the data.
From the flat data queried, you'll need to convert it to a nested tree with flatDataToTree
:
import {flatDataToTree} from '@sanity/hierarchical-document-list'
const hierarchyDocument = await client.fetch(`*[_id == "book-v3-review-a"][0]{
tree[] {
// Make sure you include each item's _key and parent
_key,
parent,
value {
reference->{
title,
slug,
content,
}
}
}
}`)
const tree = flatDataToTree(data.tree)
/* Results in a recursively nested structure. Using the example data above:
{
"_key": "741b9edde2ba",
"_type": "hierarchy.node",
"value": {
"reference": {
"_ref": "75c47994-e6bb-487a-b8c9-b283f2436031",
"_type": "reference",
"_weak": true
},
"docType": "docs.article"
},
"parent": null,
"children": [
{
"_key": "f92eaeec96f7",
"_type": "hierarchy.node",
"value": {
"reference": {
"_ref": "7ad60a02-5d6e-47d8-92e2-6724cc130058",
"_type": "reference",
"_weak": true
},
"docType": "site.post"
},
"parent": "741b9edde2ba"
}
]
}
*/
After the transformation above, nodes with nested entries will include a children
array. This data structure is recursive.
By default, this plugin will create and update documents of _type: hierarchy.tree
, with a tree
field holding the hierarchical data. When deploying a GraphQL Sanity endpoint, however, you'll need an explicit document type in your schema so that you get the proper types for querying.
To add this document type, create a set of schemas with the createHierarchicalSchemas
:
// hierarchicalSchemas.js
import {createHierarchicalSchemas} from '@sanity/hierarchical-document-list'
export const hierarchicalOptions = {
// choose the document type name that suits you best
documentType: 'myCustomHierarchicalType',
// key for the tree field in the document - "tree" by default
fieldKeyInDocument: 'customTreeDataKey',
// Document types editors should be able to include in the hierarchy
referenceTo: ['site.page', 'site.post', 'docs.article', 'social.youtubeVideo'],
// β Optional: provide filters and/or parameters for narrowing which documents can be added
referenceOptions: {
filter: 'status in $acceptedStatuses',
filterParams: {
acceptedStatuses: ['published', 'approved']
}
},
// β Optional: limit the depth of your hierarachies
maxDept: 3
}
export default createHierarchicalSchemas(hierarchicalOptions)
And add these schemas to your studio:
import createSchema from 'part:@sanity/base/schema-creator'
import schemaTypes from 'all:part:@sanity/base/schema-type'
import hierarchicalSchemas from './hierarchicalSchemas'
export default createSchema({
name: 'default',
types: schemaTypes.concat([
// ...Other schemas
...hierarchicalSchemas // add all items in the array of hierarchical schemas
])
})
Then, in your desk structure where you added the hierarchical document(s), include the right documentType
and fieldKeyInDocument
properties:
createDeskHierarchy({
// Include whatever values you defined in your schema in the step above
documentType: 'myCustomHierarchicalType', // the name of your document type
fieldKeyInDocument: 'customTreeDataKey' // the name of the hierarchical field
// ...
})
// Ideally, use the same configuration object you defined in your schemas:
import {hierarchicalOptions} from './hierarchicalSchemas'
createDeskHierarchy({
...hierarchicalOptions
// ...
})
π Note: you can also use the method above to add hierarchies inside the schema of documents and objects, which would be editable outside the desk structure.
We're considering adapting this input to support any type of nest-able data, not only references. Until then, avoid using the generated schemas in nested schemas as, in these contexts, it lacks the necessary affordances for a good editing experience.
MIT-licensed. See LICENSE.
MIT Β© Sanity
This plugin uses @sanity/plugin-kit with default configuration for build & watch scripts.
See Testing a plugin in Sanity Studio on how to run this plugin with hotreload in the studio.
Run "CI & Release" workflow. Make sure to select the main branch and check "Release new version".
Semantic release will only release on configured branches, so it is safe to run release on any branch.