Quickstart

This page gives you a brief overview of some of the core concepts in Nevermore. It's based on

Nevermore is designed to map documents. What's a document? Well, for our purposes it's a class. Here's an example of a document we'll be working with. The only assumption is that you'll provide an "Id" property which is a string (though it doesn't have to be a string).

class Person
{
    public string Id { get; private set; }
    public string FirstName { get; set; }
    public string LastName { get; set; }
    public string Email { get; set; }
    
    // Documents can have all kinds of things on them, including arrays, 
    // nested properties, and more. 
    // All these properties will be stored in the JSON blob.
    public HashSet<string> Tags { get; } = new HashSet<string>();    
    public int[] LuckyNumbers { get; set; }
}

Note that the class has some properties with arrays or hashsets - things that don't get stored in columns in SQL very well. In Nevermore, anything that isn't written to a column is serialized as JSON and stored in a JSON column.

Every document type gets a SQL table. Here's a script that creates the SQL table we'll use for this document.

CREATE TABLE [Person] (
  [Id] NVARCHAR(50) NOT NULL CONSTRAINT [PK_Person_Id] PRIMARY KEY CLUSTERED, 

  -- These are columns we might need to query against
  [FirstName] NVARCHAR(20) NOT NULL, 
  [LastName] NVARCHAR(200) NULL, 
  [Email] NVARCHAR(200) NOT NULL, 

  -- This is where we store everything else
  [JSON] NVARCHAR(MAX) NOT NULL
)
ALTER TABLE [Person] ADD CONSTRAINT [UQ_UniquePersonEmail] UNIQUE([Email])

ℹ️ It's a good practice to name your constraints (primary key and unique constraints above, for example) in case you need to drop or disable them later.