SQL Relations Filter for User Permissions

The sqlRelationsFilter function provides an efficient way to apply row-level filters across related entities directly within SQL, streamlining access control for related rows. In this lesson, we’ll use sqlRelationsFilter to apply permissions based on user access to Task entities in the TimeEntry entity.

Defining User Permission Filters

In this example, we want to ensure that users can only access TimeEntry rows associated with Task entities they have permission to view.

TimeEntry Entity Setup

import { Entity, Fields, Relations } from 'remult'
import { sqlRelationsFilter } from 'remult/internals'
import { Task } from './Task'

@Entity<TimeEntry>('timeEntries', {
  allowApiCrud: true,
  apiPrefilter: () =>
    sqlRelationsFilter(TimeEntry).task.some(Task.allowedTasks()),
})
export class TimeEntry {
15 collapsed lines
  @Fields.id()
  id = ''

  @Fields.string({ required: true })
  taskId = ''

  @Relations.toOne<TimeEntry, Task>(() => Task, 'taskId')
  task?: Task

  @Fields.date()
  startTime!: Date

  @Fields.date()
  endTime!: Date
}

Explanation of the Code

1. Using sqlRelationsFilter:

   • The sqlRelationsFilter function is designed to apply row-level filtering logic directly within the database.
   • Here, we use it to filter TimeEntry rows based on the associated Task rows that meet certain user permissions.

2. Relation-Specific Filtering with .some():

   • The .some() method is used to match TimeEntry rows with related Task rows that satisfy the filter in Task.allowedTasks().
   • By passing Task.allowedTasks() to .some(), we enforce permissions on TimeEntry rows linked to tasks the user is authorized to view.

3. Simplified Filtering Logic:

   • sqlRelationsFilter enables us to express complex filtering conditions for related entities with a clear and concise API.
   • This approach is not only efficient but also significantly reduces code complexity by handling filtering at the SQL level.

Try It Out

1. Sign in as Different Users: Test with various users (e.g., users with and without admin roles) to observe how access to TimeEntry records changes based on the user’s permissions for related Task entities.
2. Experiment with Permissions: Modify the allowedTasks filter logic in Task to see how different rules impact the visibility of TimeEntry entries.

By leveraging sqlRelationsFilter, you can create highly performant and intuitive access control that directly uses SQL to enforce row-level permissions across related entities.