HapiJS Boilerplate

Hapify

This boilerplate is meant to be used with Hapify. To get more info about Hapify setup, please refer to https://www.hapify.io/get-started.

Stack

This boilerplate provides an API built with HapiJS, MongoDB and Docker.

Get Started

1. Clone repository

• Option 1: Clone and configure this boilerplate using command hpf new --boilerplate hapijs_tractr.
• Option 2: You can clone this repository and change the project id in file hapify.json by running command hpf use.

2. Generate code

Then you need to generate code from your Hapify project using hpf generate.

⚠️ Important: For development purpose, generated files are ignored in the .gitignore. You should edit this file and remove the last lines before committing.

3. Run the API

This API should be used with docker and docker-compose.

3.1 Installation

Run installation scripts to create MongoDB indexes, S3 bucket and insert an admin:

docker-compose run --rm api npm install

docker-compose run --rm api npm run cmd setup

docker-compose run --rm api npm run cmd insert-admin

Or run all in one command:

docker-compose run --rm api bash -c "npm install && npm run cmd setup && npm run cmd insert-admin"

The login and password of the admin user is defined in file cmd/insert-admin/admin.js ([email protected] / admin).

3.2 Start server

To start the API, run this command

docker-compose up api

Now the API is available on http://localhost:3000.

3.3 Insert development data

To insert randomized data into the database, run this command

docker-compose run --rm api npm run cmd populate

3.4 Use a front-end boilerplate

This boilerplate can be used with those front-end boilerplates:

• Admin dashboard built with Angular
• Components library built with Angular

3.5 Updates

If you need to update you data models and re-generate code (using Hapify, you should:

• 1. Run this command docker-compose run --rm api npm run cmd setup to update the MongoDb indexes.
• 2. Restart the API.

Please refer to Hapify Best Practices to learn more about Git patches within Hapify Context.

Documentation

Once the API is started, you can access http://localhost:3000/docs to browse the documentation.

Advanced Integration

This boilerplate includes the following modules

• user sessions (sessions are stored in Redis)
• users accesses management
• file storage with S3 (Minio for local)
• email sender via Mailjet
• geo-coding and reverse geo-coding service

Models interpretation

This boilerplate interprets Hapify data-models fields properties as described bellow:

• Primary: Represent the MongoDB Id.
• Unique: Creates an unique index for MongoDB and throw a 409 in case of conflict.
• Label: Creates a text index for MongoDB and allow partial match search for this field.
• Nullable: Allow null to be send for POST and PATCH endpoints.
• Multiple: Only used for entity relation for One-to-Many relation. If also searchable, it performs search using operator $all.
• Embedded: Only used for entity relation. It joins related entities in search results. Related entities are always joined in a read response.
• Searchable: Allows this field in query params for search and count endpoints. If the field is also a DateTime or a Number, it adds min and max query params. It also creates an index for MongoDB.
• Sortable: allows this field's name as value in the query param _sort of the endpoint search. Also create an index for MongoDB.
• Hidden: Adds projection when getting the data from MongoDB to remove this field. On create route, it deletes the fields before returning the response.
• Internal: This field won't be allowed in POST and PATCH endpoints. It tries to guess a suitable value for this field. You may need to edit this default value after code generation.
• Restricted: This field is allowed in POST and PATCH endpoints only for admins. An admin is a user with the field role='admin'.
• Ownership: This field is used to allow the request when the access of the action is made by an owner. The value of the field is compared to the connected user id. For search and count endpoints, if also searchable, it forces to perform the lookup in the owner's documents.