Warning: GraphQL Slice Machine Limitations

Developing with Prismic
, ,
1

So after this update you advise against using GrapQL because it does not work 100%?
I would like to start a new project using GrapQL. The main reason is because of this post (the site is not pre-generated).
If using GrapQL will have general disadvantages when using it with Prismic, I will continue to use REST

Thank you

2

Yes, exactly that; by continuing to use GraphQL, you will lose the latest group field developments with Slice Machine.

You can still use GraphQuery in the rest API for this use case:

Thanks.

3

Hi! Thanks for the answer!
The problem is that grapQuery doesn't allow to fetch only meta data
ES: This doesn't work:
graphQuery: {} not even this:graphQuery: { blog { } }

PS: I suggest you to add a warning banner in the graphQL documentation because no one can know the limitation unless they view this release note

Thank you!

4

Hi Lorenzo,

While it's correct that GraphQL is no longer a good option for querying linked Slice data since these updates, you can still query your linked data on the page level with the Rest API and GraphQuery, then build your sitemap separately with GraphQL.

Adding a warning banner to the GraphQL docs is a great idea. I've added it to our backlog.

Thanks.

6

I spent a whole day trying to get GraphQL working inside Nuxt because of the 403 error due to prismic-ref only to find out that you added an undocumented utility in the client to facilitate this mechanism. client.graphQLFetch(). Maybe you should include this function in the documentation too.
Once the fetch is done, I noticed that the all{Type_id}s can fetch a maximum of 100 items. Maybe you should notify this as well. I prefer to do a single fetch to retrive all blogs rather than do 10 different ones.

The best solution would be to improve graphQuery by allowing only metadata to be fetched

Thank you.

7

Thank you for taking the time to share your experience and feedback. We apologize for the challenges you faced while working with GraphQL in Nuxt, and we appreciate you bringing this to our attention.

The client.graphQLFetch() utility was indeed introduced to streamline the process, and you’re right—it should have been clearly documented to avoid confusion. We’ll work on updating the documentation to make it easier for others to find and implement.

Regarding the item limit in all{Type_id}s, we understand the frustration with the 100-item cap. We’ll make sure this limitation is better highlighted in our docs.

Your suggestion about enhancing graphQuery to fetch only metadata is a valuable one. We’ll take this into consideration for future improvements.

Thanks again for your constructive feedback, and please don’t hesitate to reach out if you encounter any further issues or have additional suggestions.