Consider the following code:
const db = mongodb.getDb();
// Results count
let count = (await db.collection('sec').aggregate(aggregateQuery)
.group({ _id: null, count: { $sum: 1 } })
.project({ _id: 0, count: 1 })
.toArray());
console.dir(count)
// test
let test = await db.collection('sec').aggregate(aggregateQuery).toArray();
console.dir(test)
The aggregate(aggregateQuery) filters the collection ad performs a lookup with another collection.
The first query counts all documents resulting from the aggregate, and it outputs [ { count: 2086 } ] as expected.
In my mind, test should contain a bunch of documents filtered from the collection (according to the aggregate operations), but it is just [ { count: 2086 } ], as in the first query!
If in the second query instead of aggregate(aggregateQuery) I put find(), it outputs all the collection's documents, as expected, so I guess there is something I am missing in the aggregate() function...
The object passed as a parameter inside aggregate (aggregateQuery in this case) is modified by the following functions (group and project), so when it is used again in the second aggregate call, it is different and yields different results.