gatsby-source-strapi
Source plugin for pulling documents into Gatsby from a Strapi API.
⚠️ This version of
gatsby-source-strapi
is only compatible with Strapi v5 and v4. For v3 use this release
This plugin is now maintained publicly by the Gatsby User Collective. Join us in maintaining this piece of software.
Table of contents
- gatsby-source-strapi
Installing the plugin
Using yarn
yarn add gatsby-source-strapi gatsby-plugin-image gatsby-plugin-sharp gatsby-source-filesystem gatsby-transformer-remark gatsby-transformer-sharp
Or using NPM
npm install --save gatsby-source-strapi gatsby-plugin-image gatsby-plugin-sharp gatsby-source-filesystem gatsby-transformer-remark gatsby-transformer-sharp
Setting up the plugin
You can enable and configure this plugin in your gatsby-config.js
file.
Basic usage
First, you need to configure the STRAPI_API_URL
and the STRAPI_TOKEN
environment variables. We recommend using dotenv
to expose these variables.
Make sure to create a full-access API token in Strapi.
Path: ./.env.development
STRAPI_API_URL=http://127.0.0.1:1337
STRAPI_TOKEN=<my-development-api-token-for-gatsby>
Path: ./gatsby.config.js
require("dotenv").config({
path: `.env.${process.env.NODE_ENV}`,
});
const strapiConfig = {
version: 4, // Strapi version 4 or 5
apiURL: process.env.STRAPI_API_URL,
accessToken: process.env.STRAPI_TOKEN,
collectionTypes: ["article", "company", "author"],
singleTypes: [],
maxParallelRequests: 5, // (Optional) Default: Number.POSITIVE_INFINITY
remoteFileHeaders: {
/**
* Customized request headers
* For http request with a image or other files need authorization
* For expamle: Fetch a CDN file which has a security config when gatsby building needs
*/
Referer: "https://your-site-domain/",
// Authorization: "Bearer eyJhabcdefg_replace_it_with_your_own_token",
},
};
module.exports = {
plugins: [
{
resolve: `gatsby-source-strapi`,
options: strapiConfig,
},
],
};
Advanced usage
Deep queries populate
const strapiConfig = {
// ...
collectionTypes: [
{
singularName: "article",
queryParams: {
// Populate media and relations
// Make sure to not specify the fields key so the api always returns the updatedAt
populate: {
image: "*",
images: "*",
author: {
populate: {
avatar: "*",
company: {
populate: {
image: "*",
},
},
},
},
},
},
},
],
// ...
};
Draft content
Strapi now supports Draft and publish, which allows you to save your content as a draft and publish it later. By default, this plugin will only fetch the published content.
Draft content should only be pulled when previewing content in the Gatsby preview environment to enable the preview you need to fetch content only when the GATSBY_IS_PREVIEW
environment variable is truthy.
Refer to the Gatsby cloud and preview environment setup section to see how to set it up.
Path: ./gatsby.config.js
const strapiConfig = {
// ...
collectionTypes: [
{
singularName: "article",
queryParams: {
publicationState: process.env.GATSBY_IS_PREVIEW === "true" ? "preview" : "live",
// or the v5 format,
// we will automatically rewrite publicationState to status:
// status: process.env.GATSBY_IS_PREVIEW === "true" ? "draft" : "published",
populate: {
category: { populate: "*" },
cover: "*",
blocks: {
populate: "*",
},
},
},
},
],
// ...
};
Then in your GraphQL query you should be able to display published content by using the following query:
{
allStrapiArticle(filter: { publishedAt: { ne: null } }) {
nodes {
id
}
}
}
`
Image optimisation
By default all medias are downloaded in the Gatsby file system. To query your asset use the following query:
{
allStrapiArticle {
nodes {
cover {
localFile {
childImageSharp {
gatsbyImageData
}
}
}
}
}
}
If you do not want this plugin to download your media (images, videos, and file attachments), you can use the skipFileDownloads
configuration option. This may be useful for a site that is using a CDN such as Cloudinary: since the CDN provides image transformations, Gatsby’s transformation layer is not needed, and opting out makes the builds faster.
Path: ./gatsby.config.js
const strapiConfig = {
// ...
skipFileDownloads: true,
// ...
};
Rich text field
Rich text fields can now be processed using the gatsby-transformer-remark
plugin.
It only works if the content of the rich text field saved in the database is in markdown format. So if you customized the WYSIWYG in your Strapi Administration panel make sure that it is saved in a markdown format.
Files that are added in the richtext field can now also be processed by the gatsby-plugin-image
plugin.
To do so, according the restrictons and limitations of the plugin you need to make sure that at least one of your content types entities has a file uploaded in the richtext field.
To query markdown local fields use the following query:
{
allStrapiArticle {
nodes {
# richtext field content
body {
# object to access the markdown node
data {
# unprocessed data from Strapi
body
# processed markdown
childMarkdownRemark {
html
rawMarkdownBody
}
}
# files from the markdown that are processed using the gatsby-image-plugin
medias {
# alternative text saved in the markdown
alternativeText
# file from the media library
file {
# alternative text of the file in the media library
# it can be different from the one set in your markdown
alternativeText
}
# file processed with gatsby-plugin-image
localFile {
childImageSharp {
gatsbyImageData
}
}
# src set in your markdown field (ex: [alternativeText](src))
src
# prefixed url with the Strapi api endpoint of the file
# when using a provider the src field value is equal to the url field value
url
}
}
}
}
}
Components
Strapi components creates unique Gatsby nodes to be able to query a single component in GraphQL using the Gatsby id. To query a specific component use the following query:
{
strapiComponentSharedRichText(id: { eq: "id" }) {
id
}
}
Dynamic zones
To query dynamic zones, write a query using inline GraphQL fragments.
You can use the following query:
{
allStrapiArticle {
nodes {
blocks {
... on STRAPI__COMPONENT_SHARED_RICH_TEXT {
id
# Since __component is forbidden in gatsby this field is prefixed by strapi_
strapi_component
}
... on STRAPI__COMPONENT_SHARED_QUOTE {
id
strapi_component
}
... on STRAPI__COMPONENT_SHARED_MEDIA {
id
strapi_component
}
... on STRAPI__COMPONENT_SHARED_SLIDER {
id
strapi_component
}
}
}
}
}
Internationalization
Content types in Strapi can be localized with the i18n plugin. But by default, gatsby-source-strapi will only fetch data in the default locale of your Strapi app. To specify which locale should be fetched, an i18n
object can be provided in the content type’s pluginOptions
. You can also set the locale to all
to get all available localizations of a content type:
const strapiConfig = {
// ...
collectionTypes: [
{
singularName: "article",
pluginOptions: {
i18n: {
locale: "fr", // Only fetch a specific locale
},
},
},
],
singleTypes: [
{
singularName: "global",
pluginOptions: {
i18n: {
locale: "all", // Fetch all localizations
},
},
},
],
// ...
};
Then use the one of the following queries to fetch a localized content type:
{
# Get content in all available localizations
allStrapiGlobal {
nodes {
locale
}
}
# Get a single type in a specific locale
strapiGlobal(locale: { eq: "fr" }) {
locale
}
# Get a collection type in a specific locale
allStrapiArticle(filter: { locale: { eq: "fr" } }) {
nodes {
locale
}
}
}
Restrictions and limitations
This plugin has several limitations, please be aware of these:
-
At the moment, fields that do not have at least one populated instance will not be created in the GraphQL schema. This can break your site when field values get removed. You may workaround with an extra content entry with all fields filled out.
-
When using relational fields, be aware that this source plugin will automatically create the reverse reference for the first level of relation. It is advised to query both
articles
andcategories
if you want to link the properly and be able to navigate properly in the GraphQL schema. -
In Gatsby, some field names are restricted therefore these fields will be prefixed by
strapi_
. Here’s the list of the restricted field names:
children
fields
internal
parent