Page Builder > Extending Functionality
Extend GraphQL API
Learn how to extend the Page Builder-related GraphQL types and operations.
- how to extend the Page Builder-related GraphQL types and operations
Use the webiny watch
command to continuously deploy application code changes into the cloud and instantly see them in action. For quick (manual) testing, you can use the built-in API Playground.
Adding New Page Fields
In this example, we’ll add a new special
boolean field to the central PbPage
GraphQL type. As the name suggests, the field will tell us whether a page is special or not.
It all starts with the GraphQLSchemaPlugin
, which we’ll need to register within our GraphQL API’s application code. Once we have that, optionally, we might want to register the IndexPageDataPlugin
plugin, which will enable us to get the value of the new special
field also while listing pages.
The code above can be placed in the api/graphql/src/plugins/pages.ts
file, which doesn’t exist by default, so you will have to create it manually. Furthermore, once the file is created, make sure that it’s actually imported and registered in the api/graphql/src/index.ts
entrypoint file.
With all the changes in place, we should be able to update an existing Page Builder page and mark it as special, with the following mutation:
For example:
Running the above mutation should mark the page with the 60f903881f76a1000820068e#0001
ID as special, which we should be able to see afterwards while performing queries:
Modifying GraphQL Queries
If needed, existing pages-related GraphQL queries can be modified too.
Continuing from the previous example, let’s say we also wanted to be able to list special pages only. We can do that with the help of the SearchLatestPagesPlugin
and SearchPublishedPagesPlugin
plugins (both extending SearchPagesPlugin
):
With all the changes in place, we should be able to run the following GraphQL query:
For example:
Note that because we’ve created both the SearchLatestPagesPlugin
and SearchPublishedPagesPlugin
plugins, we can also apply the same special: true
filter within the listPublishedPages
GraphQL query.
The difference between the listPages
and listPublishedPages
is in the returned results. The former will always return latest revisions of a pages, which is more useful while listing pages inside the Admin Area application. The latter always returns published revisions of pages, which is more suitable for public applications and websites.
Custom GraphQL Mutations
Let’s say we wanted to extend our GraphQL schema with the custom duplicatePage
mutation, which, as the name suggests, would enable us to make copies of pages.
We can achieve this with a single GraphQLSchemaPlugin
plugin.
The code above can be placed in the api/graphql/src/plugins/pages.ts
file, which doesn’t exist by default, so you will have to create it manually. Furthermore, once the file is created, make sure that it’s actually imported and registered in the api/graphql/src/index.ts
entrypoint file.
With all the changes in place, we should be able to run the following GraphQL mutation:
For example:
After the mutation has been executed, we should be able to see the created copy in the list of pages:
FAQ
What Is thecontext
Object and Where Are All of Its Properties Coming From?
In the shown examples, you may have noticed we were using the context
object in GraphQL resolver functions. This object contains multiple different properties, mainly being defined from different Webiny applications that were imported in the GraphQL API’s api/graphql/src/index.ts
entrypoint file.
That’s why, for example, we were able to utilize the context.pageBuilder.pages.get
and context.pageBuilder.pages.update
methods, in the Custom Mutations section.
For easier discovery and type safety, we suggest a type is always assigned to the context
object in your GraphQL resolver functions.