DIfference between sub docs and population in Mongoose and doubts about database schema

Question

I am new to MongoDB and MongooseJS. I am also quite inept at creating databases, sorry.

First question

what is the difference between sub docs and population? Looking at the docs example, a Parent-Children subdoc seems very similar to a Person-Stories population.

Database schema question

Scenario: users can create a number of Canvas. Each Canvas can host a number of Shapes. Each Shape can be of 2 types: Simple, or Complex. A Simple shape can be a Square or a Circle (objects). A Complex shape is made of a Frame and a Material.

            Canvas
               |
             Shape
            /     \       
      Simple       Complex: Frame, Material 
      /     \
Square       Circle

Now, a Shape can be assigned to only 1 Canvas but can be moved between Canvases or live alone outside a Canvas. Frame and Material can be created independently (live alone) and added to a Complex Shape (each Complex Shape can only have 1 frame and 1 material).

Some of the queries I would have to implement are asking for all the shapes in a specific Canvas, or all the shapes that are using a certain material, or again all the Frames that are being used by shapes.

Since the difference between sub-doc and population is not clear to me, I am not even sure how to begin... any help or example would be appreciated (I am not expecting a full and working database schema).

Answer 1

In answer to part one, the difference is that in a Document-Subdocument scenario the subdocument is stored with the document (as a child document, accessible through the use of dot notation).

In the Mongoose populate scenario, a reference is held to a document in another collection. When you tell mongoose to populate what it does is make another query to MongoDB to get this 'subdocument'.

The main difference I see is that with the mongoose populate method you cannot query for documents using the 'subdocument' properties, whereas you can with the embedded subdocument model.

---

Edit, subdocument example

So in MongoDB you can store the data as nested subdocuments like so:

{
    _id: acbdbd,
    "property": "value",
    "subdocumentproperty" : {
        "param1": 1,
        "param2": 2
    }
}

This data is only stored with the master document and your querying would be done through the master document, e.g.

db.collection.find({"subdocumentproperty.param1": 1})