getLookupPipeline()

getLookupPipeline(description: Description, options?: BuildLookupPipelineOptions) => Document[]

This function builts a MongoDB aggregation pipeline using the internal reference resolution API. It can be used to iterate documents lazily with .next() instead of returning paginated arrays.

Example

The example below gets the aggregation pipeline for retrieving documents of the person collection with references being populated with objects instead of ObjectIds. The value of pipeline will be an array containing MongoDB aggregation stages.

Instead of retrieving all documents in a single call using getAll() (that also deep-populates references), the code in this example iterates documents one by one, and can be used to traverse all the collection in a memory efficient way.

const pipeline = await getLookupPipeline(context.collections.person.description, {
  memoize: context.collections.person.description.$id,
  project: payload.populate || project,
})

const it = context.collection.model.aggregate(pipeline)

for await ( const document of it ) {
  // {
  //   name: 'John',
  //   pet: {
  //     _id: ObjectId('...'),
  //     name: 'Thor'
  //   }
  // }
  console.log(document)
}