Writing CRUD layers is a pain, maintaining generated CRUD layers is a pain, you've defined CRUD APIs a million times, and you want to get on with the interesting parts of your project - you should be able to describe your data, the relationships between your data, and rules for accessing your data succinctly, in one place, and all of the infrastructure and machinery implied by that declaration should just exist.
You shouldn't have to choose between tools that work for your web-app and tools that work for massive scale - online or offline - data processing and analysis. You shouldn't be constrained to one implementation, or trapped by one cloud provider.
Basestar solves a relatively simple problem, but it solves it comprehensively and gives you simplicity and power without taking away flexibility or control.
Pick the parts you need, for example:
<dependencies>
<dependency>
<groupId>io.basestar</groupId>
<artifactId>basestar-database-server</artifactId>
<version>${basestar.version}</version>
</dependency>
<dependency>
<groupId>io.basestar</groupId>
<artifactId>basestar-storage-dynamodb</artifactId>
<version>${basestar.version}</version>
</dependency>
</dependencies>
See: Basestar schema (basestar-schema)
See: Basestar expressions (basestar-expression)
Write me
Write me
Principal:
type: object
User:
type: object
extend: Principal
links:
groups:
schema: Group
expression: member.id == this.id for any member in members
Group:
type: object
extend: Principal
properties:
members:
array: User
indexes:
over:
member: members
partition:
- member.id
Project:
type: object
properties:
owner:
type: Principal
permissions:
read:
expression: this.principal.id in {caller.id} + {group.id for group in caller.groups}
expand:
- caller.groups
Thing:
type: object
properties:
project:
type: Project
permissions:
read:
inherit:
- project.read