entity

Overview ¶

Package entity defines a convenient abstraction which can be used with MongoDB in order to streamline CRUD operations.

This definition also allows a centralization of security and other policies related to an entity within the app. This leads to fewer bugs, and when there are bugs, due to the modularization of policies, the origin of bugs is easier to locate.

The goal is to define an abstraction which is useful for the general entity and can make the process of writing code much more efficient.

Axis Policy ¶

An axis is defined as a eField in an Entity which can be assumed to be unique. This is important when creating collection indexes and creating Query/Update/Delete filters. The Axis Policy ensures data integrity by enforcing that all Entities within a collection have unique axis values.

This Policy is especially useful when querying elements for Read/Update/Delete operations. The client benefits through the ability to specify whether a eField is an axis, using the "axis" tag, in the struct eField. This tag can be set to "true" to enforce it as an axis eField.

Getting started ¶

To use the Entity abstraction, start by creating a struct which will define the Entity that you want to work with. Available in this step, is the "axis" tag which is useful in specifying which fields are to be treated as axis fields. For example, here is a hypothetical struct for defining a (useless) User Entity:

type User struct {
	ID    primitive.ObjectID `_id_:"user" json:"-" bson:"_id"`
	Name  string             `json:"name" _hd_:"c"`
	Email string             `json:"email" axis:"true" index:"text" _hd_:"c"`
}

Next, register this User struct as an Entity.

UserEntity := Entity{
	SchemaDefinition: TypeOf(User{}),
	PStorage:         &mongoCollection
}

Run the Optimize function to generate indexes for the axis fields:

UserEntity.Optimize()

Create a User:

u := User{
	Name:  "Jane Doe",
	Email: "jane.doe@example.com"
}

Add this user to the database:

id, err := UserEntity.Add(u)

The other Read/Update/Delete operations become as simple with this Entity definition and can lead to hundreds of lines of less code being written. The Entity package aims to (at least) provide a high level API with the basic CRUD boilerplate code already taken care of.

See github.com/navaz-alani/entity/multiplexer for information about the EntityMux which is able to manage a collection of entities for larger applications.