Imager (Backend)

Imager is a simple web-application that allows users to browse and share places of interests.

Background

This repository holds the backend source-code.

Imager backend is an Express application built on NodeJS using Vite, and serves as a REST API for the frontend to communicate places and users data back and forth the MongoDB storage.

• Express Validator provides a means to perform field validation as a middleware in routes, as well as aggregate validation outcomes in the controllers.
• Multer facilitates handling of image payloads from incoming POST requests, as well as persistence to disk storage within hosted space.
• Mongoose caters methods to perform CRUD operations for places and users records in MongoDB collections.
• Authentication is carried out via issuance and verification of tokens to and from the frontend, especially for save/update/delete actions that require protection against unauthorized changes, and is done so with the jsonwebtoken library.
• Other dependencies include uuid for identifier-generation, bcryptjs for password hashing and verification, axios for querying Google reverse-geocoding API, dotenv for handling of environment variables, and nodemon aiding hot-reloading of development activities.

Imager (frontend) repository can be accessed here.

Routes

Places

Routes	Method	Description	Protected
"/api/places/placeId"	GET	Retrieves an existing Place record from MongoDB of the targeted place identifier in the URL.	No
"/api/places/user/:userId"	GET	Retrieves a list of all Place records that are associated with a user of the provided user identity in the URL.	No
"/api/places"	POST	Creates a new Place record based on multi-part form data consisting of text fields and image payload. Text data is save to MongoDB, while image is written to disk storage. Record is associated with user identity encrypted within JWT.	Yes
"/api/places/:placeId"	PATCH	Updates a Place record within MongoDB of the targeted place identifier in the URL. Verifies that the targeted place is authored by the same user identity that is encrypted within the JWT. Only title and description can be updated via this method.	Yes
"/api/places/:placeId"	DELETE	Removes a place record from MongoDB with the provided place identifier in the URL. Subsequently removes the place identifier from the places array field within User record. Verifies that the targeted place is authored by the same user identity that is encrypted within the JWT.	Yes

Users

Routes	Methods	Description	Protected
"api/users"	GET	Retrieves all User records.	No
"api/users/signup"	POST	Creates a new User record based on the multi-part form data consisting of name, email, password and avatar image. Text data is saved into MongoDB, while the avatar image is written to disk storage. Upon successful creation of the new User record, a token is generated and returned in the response to immediately sign in the new user.	No
"api/users/login"	POST	Verifies user credentials against existing User record. Upon successful verification, a token is generated and returned in the response to sign in the user	No

Data Dictionary

Place

Column	Type	Description	Mandatory
title	String	Name of place.	Yes
description	String	Plain text description/review of the place that was visited. Minimum length of 20 characters.	Yes
imageUrl	String	Relative path to the place image file that is stored on disk.	Yes
address	String	Full address of the place. This value would be used for submission to Google Maps API for reverse-geocoding, so as a derive the latitude/longitude coordinates.	Yes
location	Object	Holds 2 fields, lat and lng	Yes
location.lat	Number	Latitude value of place, returned from Google Maps API	Yes
location.lng	Number	Longitude value of place, returned from Google Maps API	Yes
places	Array (ObjectId)	List of Document identifier values of Place records belonging to this User	Yes
creator	ObjectId	Document identifier value of User who created this place	Yes

User

Column	Type	Description	Mandatory
name	String	Name of User	Yes
email	String	Email address of User	Yes
password	String	Hashed value of login password	Yes
imageUrl	String	Relative path to the user avatar file that is stored on disk.	Yes
places	Array (ObjectId)	List of Document identifier values of Place records belonging to this User	Yes

Setup

Pre-requisite:

• MongoDB has been fully set up, i.e. a database with 2 collections, namely places and users. Note down all the database access credentials.
• A valid API key has been obtained from Google Maps Platform for usage within Imager backend. This key would be used for reverse-geocoding of place address to derive lat/lng coordinates.
• A private key that would serve as a salt value for signing and verifying JWTs.

Steps:

1. Clone this repository in terminal.

git clone git@github.com:eugeneyan84/imager-backend.git

2. Navigate into the project folder in terminal, install all dependencies.

npm install

3. Rename .env_template file in project root directory to .env, and provide the necessary values as captured from the Pre-requisite section.

# .env file
GOOGLE_API_KEY=<google api key>

PORT=<port value of this backend>

# MongoDb access credentials
IMAGER_APP_HOSTNAME=<MongoDB hostname>
IMAGER_APP_USERNAME=<MongoDB username>
IMAGER_APP_KEY=<MongoDB user password>
IMAGER_APP_DBNAME=<MongoDB database name>

#private key for JWT signing
BACKEND_P_KEY=<key string>

From this point on, there are 2 options to serving the frontend:

Option A: Serve directly from code server (e.g Visual Studio Code)

4. Navigate to project folder in terminal, run the project:

npm run start

Option B: Serve as a Dockerized application

4. Navigate to project folder in terminal and use the following build command:

docker build -f Dockerfile.dev -t imager-backend .

5. After verifying that Docker image has been successfully built, use the following run command in terminal to spin up a container in detached mode.

docker run -d --env-file ./.env -p 3000:3000 imager-backend

Note: Edit the Docker commands accordingly if you choose to perform additional actions like tagging. Ensure that the path to the .env file is indicated correctly during running.

Standard tools like Postman can be used to test the accessibility of the endpoints documented in the Routes section above, to ensure if the backend is correctly served.

Written with StackEdit.