Skip to main content

Creating a Source Plugin

There are two types of plugins that work within Gatsby’s data system: “source” and “transformer” plugins.

  • Source plugins “source” data from remote or local locations into what Gatsby calls nodes.
  • Transformer plugins “transform” data provided by source plugins into new nodes and/or node fields.

This doc focuses on source plugins and uses gatsby-source-filesystem to explain how source plugins work.

What do source plugins do?

The gatsby-source-filesystem plugin “sources” data about files from the file system. It creates nodes with a type File, each File node corresponding to a file on the filesystem. On each node are fields like the absolutePath, extension, modifiedTime, etc.

What fields are required?

media type is not required, yet necessary to work with transformer plugins

Each node created by the filesystem source plugin includes the raw content of the file and its media type.

A media type (also MIME type and content type) are an official way to identify the format of files/content that is transmitted on the internet e.g. over HTTP or through email. You’re probably familiar with many media types such as application/javascript, application/pdf, audio/mpeg, text/html, text/plain, image/jpeg, etc.

Each source plugin is responsible for setting the media type for the nodes they create. This way, source and transformer plugins can work together easily.

This is not a required field but it’s the way for source plugins to indicate to transformers that there is “raw” data that can still be further processed. It allows plugins to remain small and focused. Source plugins don’t have to have opinions on how to transform their data. They can just set the mediaType and push that responsibility to transformer plugins.

For example, it’s quite common for services to allow you to add content as markdown. If you pull that markdown into Gatsby and create a new node, what then? How would a user of your source plugin convert that markdown into HTML they can use in their site? Luckily you don’t have to do anything. Just create a node for the markdown content and set its mediaType as text/markdown and the various Gatsby markdown transformer plugins will see your node and transform it into HTML.

This loose coupling between the data source and the transformer plugins allow Gatsby site builders to quickly assemble complex data transformation pipelines with little work on their (and your (the source plugin author)) part.

What does the code look like?

A source plugin is a normal NPM package. It has a package.json with optional dependencies as well as a gatsby-node.js where you implement Gatsby’s Node.js APIs. Gatsby’s minimum supported Node.js version is Node 8 and as it’s common to want to use more modern Node.js and JavaScript syntax, many plugins write code in a source directory and compile the code. All plugins maintained in the Gatsby repo follow this pattern.

Your gatsby-node.js should look something like:

Peruse the sourceNodes and createNode docs for detailed documentation on implementing those APIs.

What are the jobs of a source plugin?

At a high-level, these are the jobs of a source plugin:

  • Ensure local data is synced with its source and 100% accurate. If your source allows you to add an updatedSince query (or something similar) you can store the last time you fetched data using setPluginStatus.
  • Create nodes with accurate media types, human meaningful types, and accurate contentDigests.
  • “Link” nodes types you create as appropriate (see Node Link in the API specification concepts section).
  • Return either a promise or use the callback (3rd parameter) to report back to Gatsby when sourceNodes is fully executed. If a promise or callback isn’t returned, Gatsby will continue on in the build process, before nodes are finished being created. Your nodes might not end up in the generated schema at compilation, or the process will hang while waiting for an indication that it’s finished.

Getting helper functions

gatsby-node-helpers, a community-made NPM package, can help when writing source plugins. This package provides a set of helper functions to generate Node objects with the required fields. This includes automatically generating fields like node IDs and the contentDigest MD5 hash, keeping your code focused on data gathering, not boilerplate.


Two ways of adding relationships between nodes

Gatsby source plugins not only create nodes, they also create relationships between nodes that are exposed to GraphQL queries.

There are two ways of adding node relationships in Gatsby: (1) transformations (parent-child) or (2) foreign-key based.

Option 1: transformation relationships

An example of a transformation relationship is the gatsby-transformer-remark plugin, which transforms a parent File node’s markdown string into a MarkdownRemark node. The Remark transformer plugin adds its newly created child node as a child of the parent node using the action createParentChildLink. Transformation relationships are used when a new node is completely derived from a single parent node. E.g. the markdown node is derived from the parent File node and wouldn’t ever exist if the parent File node hadn’t been created.

Because all children nodes are derived from their parent, when a parent node is deleted or changed, Gatsby deletes all of the child nodes (and their child nodes, and so on) with the expectation that they’ll be recreated again by transformer plugins. This is done to ensure there are no nodes left over that were derived from older versions of data but shouldn’t exist any longer.

Creating the transformation relationship

In order to create a parent/child relationship, when calling createNode for the child node, the new node object that is passed in should have a parent key with the value set to the parent node’s id. After this, call the createParentChildLink function exported inside actions.


Here’s the above example from the gatsby-transformer-remark source plugin.

Here’s another example from the gatsby-transformer-sharp source plugin.

Option 2: foreign-key relationships

An example of a foreign-key relationship would be a Post that has an Author.

In this relationship, each object is a distinct entity that exists whether or not the other does, with independent schemas, and field(s) on each entity that reference the other entity — in this case the Post would have an Author, and the Author might have Posts. The API of a service that allows complex object modelling, for example a CMS, will often allow users to add relationships between entities and expose them through the API.

When an object node is deleted, Gatsby does not delete any referenced entities. When using foreign-key references, it’s a source plugin’s responsibility to clean up any dangling entity references.

Creating the relationship

Suppose you want to create a relationship between Posts and Authors, and you want to call the field author.

Before you pass the Post object and Author object into createNode and create the respective nodes, you need to create a field called author___NODE on the Post object to hold the relationship to Authors. The value of this field should be the node ID of the Author.

Creating the reverse relationship

It’s often convenient for querying to add to the schema backwards references. For example, you might want to query the Author of a Post but you might also want to query all the posts an author has written.

If you want to call this field on Author posts, you would create a field called posts___NODE to hold the relationship to Posts. The value of this field should be an array of Post IDs.

Here’s an example from the WordPress source plugin.

Union types

When creating fields linking to an array of nodes, if the array of IDs are all of the same type, the relationship field that is created will be of this type. If the linked nodes are of different types; the field will turn into a union type of all types that are linked. See the GraphQL documentation on how to query union types.

Edit this page on GitHub