# NoSQL Concepts: Document Databases

## Document Databases
- Schemaless
- Store data in documents and these documents are grouped into collection
- If you are familiar with relational databases we can say documents are analogous to rows and collections are analogous to table


## Documents
- Set of key-value pairs
- Key: string
- Value: number, string, booleans, arrays, objects
- Polymorphic model it's means that documents within the same collection don't needt to have the same structure
- Formats: JSON, BSON, YAML, XML


## JSON document format
````
{  
  "user_id": 512,
  "name": "Carol", 
  "last_name": "Harper",
  "email": "carolharper@datazy.com",
  "address": { 
  "street": "123 Sesame Street",
  "city": "New York City", 
  "state": "New York",  
  "country": "USA"  },
  "hobbies": [   
  "hiking", "painting"
  ]
}

````

The value of "address" key is another document, notice that is related information is embedded in the main document, so don't need for searching about address in another document


## Collections
- Set of documents
- Store **the same type of entities**



## Advantages of document databases
- Flexibility
  - Don't need to predefine the schema
  - Documents can vary over time => avoid schema migrations 
  - Embedded documents => avoid joins
- Intuitive for developers
  - Natural way to work
  - JSON is human-readable
  - **Documents map objects** in code

- Horizontal scalability


## Limitations of document databases
- More responsibility
  - Care about data in the application code
  - Care about redundant data


The most popular document databases is **"Mongo DB"**