Persisted Queries

[houdinigraphql]

Persisted Queries

[AlecAivazis]

Its definitely something that I would love to see in houdini, just forgot to create the issue 😅

[pixelmund]

Thinking about this some more, we could probably use a relay like approach where we create a output file queryMap.json customizable with an cli flag e.g. --persist-output ./src/some_folder/queryMap.json with the artifact hash as key and the query as value. We could then pass the artifact or some form to the environments fetchFunction instead of only text and variables. From there some could simply pass documentId (the artifact hash) instead of the full query.

export default new Environment(async function (operation) {
	// send the request to the api
	const result = await this.fetch('/graphql', {
		method: 'POST',
		headers: {
			'Content-Type': 'application/json',
		},
		body: JSON.stringify({
			// query: operation.text,
			variables: operation.variables,
                        documentId: operation.hash		
            }),
	})

	// parse the result as json
	return await result.json()
})

We could then probably set the raw field in generated artifacts to null if we're persisting queries, but i'm not sure.

• Where are queries getting hashed? hashDocument doesn't seem to be used.

[AlecAivazis]

Awesome - in general the approach seems like the right one. I think it's probably more likely that the user is going to follow the standard that urql and apollo both follow which can be found here but I don't think that really matters from our perspective. As long as a hash gets generated and attached to the artifact in a way that the environment has access to when making queries, the user can make that decision for themselves.

Also, hashDocument isn't used any more but I left it in the codebase in case we needed it. It's a remnant from when the preprocessor import the artifacts but that's no longer the case.