brysgo / graphql-bookshelf Goto Github PK

I have created a small example where the model is like the following

 Student --- hasMany ---> Homework --- hasMany -----> Page

Now, when I load the following (note: viewer is the student) the there is one query for the student, one for the homework, but multiple for the pages. There are as many queries as there are pages. How can i optimize it so that there is only one query for the all the pages.

Best case scenario would be to load the whole thing at once, but i'd be happy for any pointers on how to optimize this.

{ 
 viewer(id: 3) {
    name
    homeworks{
      subject
      pages{
        content
      }
    }
  }
}

The code i used to test this is available at https://gist.github.com/hardfire/3ff69d3a602bf3c30dcbbca71d67d605

Hey there,

I'm interested in this library because I'm currently using bookshelf in an app and am curious about graphql. I've peeked at the source code and tests a few times now, but don't quite grok what the point of this library is.. Is the idea that you can use the bookshelf models you've already defined within your graphql types?

I think it would be nice if a clearer explanation existed on the readme describing the problem this library is trying to solve as well as what its philosophy is.

Thanks very much,
Thomas

Similar to the improvement to eager load relationships requested in the initial graphQL query we should also look at only fetching the attributes requested in the original graphQL query as well instead of loading the entire list of attributes of the underlying model.

Hello, great library! I've been evaluating it and had a question about using the hasMany helper function. If you specify that a BookshelfType hasMany of something else, it looks like here in the code that it loads the relation as a second query. Is that correct?

Also, instead of using this.constructor.collection to call the static method on L31 I believe you could use BookshelfType.collection

Just wondering if this library work with bookshelf that is connecting to an oracle db?

Thanks!

The devDependency graphql was updated from 14.3.0 to 14.3.1.

🚨 View failing branch.

This version is covered by your current version range and after updating it in your project the build failed.

graphql is a devDependency of this project. It might not break your production code or affect downstream projects, but probably breaks your build or test tools, which may prevent deploying or publishing.

Your Greenkeeper Bot 🌴

I think we should try and build caching and or batching into bookshelf. This framework will not scale very well until this optimization is made.

Version 0.14.2 of knex was just published.

This version is covered by your current version range and after updating it in your project the build failed.

knex is a devDependency of this project. It might not break your production code or affect downstream projects, but probably breaks your build or test tools, which may prevent deploying or publishing.

Your Greenkeeper Bot 🌴

Hey, thanks for the great work, but I'm curious, how would you go about using this library to do mutations?

Hello! I'm trying to wire up a User model with two GraphQL query fields.

The type:

const userType = new GraphQLObjectType(BookshelfType({
  name: 'User',
  description: 'A user',
  fields: model => ({
    id: globalIdField('User'), // Relay stuff
    email: model.attr({
      type: GraphQLString,
    }),
    role: model.attr({
      type: GraphQLString,
    }),
  }),
  interfaces: [nodeInterface], // Relay stuff
}));

The schema:

const queryType = new GraphQLObjectType({
  name: 'Query',
  fields: () => ({
    node: nodeField,
    users: {
      type: new GraphQLList(userType),
      resolve: () => User.fetchAll(),
    },
    user: {
      args: {
        id: { type: GraphQLInt },
      },
      type: userType,
      resolve: (_, args) => User.forge({ id: args.id }).fetch(),
    },
  }),
});

export default new GraphQLSchema({
  query: queryType,
});

I can query one user just fine with { user(id: 1) { id, email, role } }.

However, if I query the all users field { users { id, email, role } } I get this error: User Error: expected iterable, but did not find one for field Query.users.

I guess it's complaining about type: new GraphlQLList(userType) as return type? It's expecting an iterable but gets a Bookshelf Collection, right?

If I force it to return an array instead of a collection it works:

// I change
resolve: () => User.fetchAll()
// to:
resolve: () => User.fetchAll().then(users => users.reduce((arr, user) => { arr.push(user); return arr; }, [])),

What am I doing wrong here?

Last changes in GraphQL break graphql-bookshelf.

The model.* wrappers that define the resolve functions make use of info.fieldName.
However, as you can see from:
https://github.com/graphql/graphql-js/blob/master/src/execution/execute.js#L593
the signature of the resolve function has changed and info is now the 4th parameter.

I have included a fix in #12.

It is still unclear to me what the best approach for building a relay compatible schema with graphql bookshelf is.

Take advantage of this bad boy: https://www.graphile.org/graphile-build