Skip to content
gqlxj1987's Blog
Go back

Medium GraphQL server design

Edit page

原文链接

It should be easy to alter the shape of the data

We currently use protocol buffers as a schema for data that comes from our backend

It should be clear what data is for the client

Within our GraphQL server, data is being passed around and exists in different stages of “readiness” for the client

It should be easy to add new data sources

Fetchers, Repositories (Repos), and the GraphQL Schema.

Fetchers should correspond to a REST or preferably a gRPC endpoint. Fetchers require a protobuf. This means that any data that is being fetched by a Fetcher must follow the schema defined by the protobuf

data flow

The data’s shape becomes more like what the client needs as it passes through each of the distinct layers. It’s clear where the data comes from at each step and we know what each piece of the server is responsible for

关注数据shape


Edit page
Share this post on:

Previous Post
kubeflow pipline deploy
Next Post
python command-line interfaces