Open yaquawa opened 3 years ago
@yaquawa to make sure we are clear on the feature request itself, could you please add more information about the scenario you want to solve, example schemas, expected behavior, how could you imagine defining the projection attributes, is it for builtin @model
operations or only for @key
defined queries?
For example, if I have a schema like this
type Post {
id: ID!
content: String
title: String
}
even I query with only title
field, at the backend, content
is still fetched, think about what will happen if the content
contains 10MB byte data.
query {
getPost(id){
title
}
}
Currently,
GetItem
andQuery
operation on the VTL are not using Projection Expression according to the requested GraphQL field.I have items with large data in specific attributes, this causes a serious problem on both performance and running cost.