Object Relational Mapping

The javaxt-orm library is a handy tool for generating Java code and DDL from a set of models defined in a Javascript or JSON document. The generated classes are used to persist data in a relational database via the javaxt.sql.Model class.

Key Features

• Auto-generate database schema and Java classes
• Command line interface
• Supports spatial geometry types with JTS
• Plug and play with javaxt-core and javaxt-express
• Supports common RDMBS (PostgreSQL, H2, etc)

Note that this is not intended to be a full fledged ORM framework. Instead, the goal is to help jumpstart new projects by providing a simple utility for stubbing out code and SQL.

java -jar javaxt-orm.jar /path/to/model.js /output

After you run the script, the output directory will contain a "schema.sql" and *.java files with source code.

Example Model

var package = "com.example.models";
var models = {
    Address: {
        fields: [
            {name: 'street',        type: 'string'},
            {name: 'city',          type: 'string'},
            {name: 'state',         type: 'string'},
            {name: 'postalCode',    type: 'string'},
            {name: 'coordinates',   type: 'geo'}
        ]
    }
}

• package: Name of the Java package that will contain the source code.
• models: JSON with model specs.

Model Specs

Individual models are defined using one or more fields. In addition, models can have constraints and default values, as well as other settings. Here's a full list of all the parameters.

• fields: An array of field names and field type. Models must contain at least 1 field. The full range of supported field types is listed below.
• constraints: An array of database constraints for individual fields. This parameter is optional.
• defaults: An array of default values for individual fields. This parameter is optional.
• hasMany: An array of Model names. This parameter is optional.
• implements: An array of Java classes that the model implements (e.g. java.security.Principal). This parameter is optional.

Supported Field Types

Notes:

• IDs are automatically added so you don't have to explicitly define one in the model.
• lastUpdate fields in the model are automatically assigned a trigger in the database.

Model Field Type

In addition to the standard field types, you can specify another model as a `type`. In the example below, the address field in the Contact model is a `Address` type.

{
  Contact: {
     fields: [
        {name: 'name', type: 'string'},
        {name: 'address', type: 'Address'}
     ]
  },
  Address: {
      ...
  }
}

When a model is used as a type, a foreign key is created in the database.

Supported Field Constraints

Field Defaults

• name: Name of the field.
• value: Default value (e.g. number, text, or boolean).

Example Model With Constraints and Defaults

The following is an example of a "User" model with constraints defined for several fields (e.g. username, password, active):

{
    User: {
        fields: [
            {name: 'username',    type: 'string'},
            {name: 'password',    type: 'password'},
            {name: 'accessLevel', type: 'int'},
            {name: 'active',      type: 'boolean'},
            {name: 'contact',     type: 'Contact'},
            {name: 'auth',        type: 'json'}
        ],
        constraints: [
            {field: 'username',   required: true,  length: 255,  unique: true},
            {field: 'password',   required: true},
            {field: 'active',     required: true}
        ],
        defaults: [
            {name: 'active',    value: true}
        ]
    }
}

You can also define constraints in the field definitions like this:

{
    User: {
        fields: [
            {name: 'username',    type: 'string',      required: true,  length: 255,  unique: true},
            {name: 'password',    type: 'password',    required: true},
            {name: 'accessLevel', type: 'int'},
            {name: 'active',      type: 'boolean',     required: true},
            {name: 'contact',     type: 'Contact'},
            {name: 'auth',        type: 'json'}
        ]
    }
}

Indexes

Indexes are created automatically for any foreign keys associated with models and any "geo: field type. You can add additional indexes using the "indexes" array like this:

{
    Contact: {
        fields: [
            {name: 'firstName',   type: 'string'},
            {name: 'lastName',    type: 'string'},
            {name: 'birthDay',    type: 'int'},
            {name: 'address',     type: 'Address'} //see address model example
        ],
        indexes: [
            {field: 'firstName'},
            {field: 'lastName'},
            {field: 'address'}
        ]
    }
}

You can simplify the index definitions further like this:

{
    Contact: {
        fields: [
            ...
        ],
        indexes: [ 'firstName','lastName','address' ]
    }
}

Or you can create more complex indexes like this:

{
    Contact: {
        fields: [
            ...
        ],
        indexes: [ 
            {name: 'idx_contact_name', type: 'unique', field: ['firstName','lastName']},
            {name: 'idx_contact_address', field: 'address'}
        ]
    }
}

Advanced Options

• tablespace: Name of the tablespace to use for the database object (tables, indexes, etc).
• jts: Which JTS package to use when using geometry types. Options are "org.locationtech.jts" and "com.vividsolutions.jts"

Dependencies

• Nashorn for parsing model inputs in Javascript format
• javaxt-core for JSON and basic file I/O

Generated Code Dependencies

License

Maven Support

  <repositories>
    <repository>
      <id>javaxt.com</id>
      <url>https://www.javaxt.com/maven</url>
    </repository>
  </repositories>

  <dependencies>
    <dependency>
      <groupId>javaxt</groupId>
      <artifactId>javaxt-orm</artifactId>
      <version>1.2.3</version>
    </dependency>
  </dependencies>