Data model Reference and schema

Question

I just started to learn MongoDB database and for my project.

I am trying to implement a MongoDB reference schema.

In my Menu table I would have a food ingredients attributes where it will store the food ingredients.

I am confused about designing my schema.

How am I going to build the schema with the food ingredients in Menu table? Because it might be 100 ingredients, do I need to create an attribute for each of them ? Or I can make one string attribute where I am going to store everything?

Here is an example of it.

var Menu = new Schema({
  foodTitle: String,
  foodIng: {
    title: String,
    ingredients: {
       rice: String,
       beans: String,
       meat: String,
       cheese: String,
       // etc...

    }      
  },
});

Answer 1

As far as I have understood your question, you are trying to create a model for dishes and list their ingredients.

You should never, ever use values as keys. With a relational database, would you create a table which has a column for each imaginable ingredient? Hardly.

Furthermore, while redundancy isn't necessarily a bad thing, it is in keys. In a collection called "dishes" or "food", it is natural that an ingredient does not refer to a chemical compound, right?

So, here is how I'd model a dish:

{
  "_id": new ObjectId(),
  "name": "Katsuo no Tataki",
  "desc": "Lightly Broiled Bonito",
  "tags": ["fish","jp"]
  "ingredients": [
    { "name": "Bonito Filet", qty: "1/2" }
    { "name": "onion", qty: "1/2"},
    { "name": "garlic", qty: "2 cloves"}
  ]
}

When displaying a dish, you simply iterate over the ingredients array, displaying both name and qty.

In order to find all dishes which contain garlic, you simply do

db.dishes.find({"ingredients.name":/garlic/i})

or, to find recipes containing the fish tag:

db.dishes.find({"tag":/fish/i})

Note that I used a case insensitive regex search.