less than 1 minute read

By default, when you find documents on mongo it returns the entire document, similar to what SELECT * FROM TABLE; would do in a relational world.

> db.user.find().pretty()
{
	"_id" : ObjectId("5217705cc2e6fc761c0b5371"),
	"userId" : "james",
	"age" : 27,
	"favouriteVideoGame" : "Skyrim"
}

You can retrieve partial document info by using something known as a projection (see mongo docs here). Use find() with an empty query, but specify the fields you want from the document like so:

> db.user.find({}, {userId:true})
{ "_id" : ObjectId("5217705cc2e6fc761c0b5371"), "userId" : "james" }

> db.user.find({}, {userId:true, age:true})
{ "_id" : ObjectId("5217705cc2e6fc761c0b5371"), "userId" : "james", "age" : 27 }

For the Java devs using Spring data, you can do a similar thing using the Query APIs, notably the fields().include() (docs here):

Query q = Query.query(Criteria.where("userId").is(userId));
q.fields().include("age");

User score = mongoTemplate.findOne(q, User.class, "user");