Images manager

This repository contains the necessary code to deploy a server which exposes a rest api that handles two types of image operations: loading a new image and getting an image given its name.

Project structure

├── cmd
│   └── server
│       └── main.go       // To run our rest server
├── pkg
│   └── model
│       └── image.go      // Model for our application
│   └── rest
│       └── handler.go    // Request handlers
│   └── storage
│       └── device.go     // Implementation when work on device storage
│       └── handler.go    // Implementation of an interface to work with different storage systems
│       └── memory.go     // Implementation when work on memory storage

API

The api exposes two endpoints with which you can interact with it.

Endpoint	HTTP Method	Possible http status codes	Note
/images/<img-id>	GET	200, 400, 404, 500	img-id = image name (extension included). Ex: my_image.png
/images	POST	201, 400, 409, 500	Set Header Content-Type: multipart/form-data

How to consume API

The following are examples of how to consume the api using cURL.

Get image by ID

• cURL

curl --request GET 'http://localhost:8080/images/person.jpeg' --output <full-new-file-name>

Upload new image

• cURL

curl --request POST 'http://localhost:8080/images' \
--form 'image=@"<img-path>"'

How to run the application locally

The application can be deployed in two ways: locally or using docker. But before deploying the application, it is necessary to configure the following environment variables:

Environment variable	Possible value	Default value	Description
IMG_STORAGE_DEVICE	memory / device	device	type of storage used by the app
IMG_STORAGE_DIR	path to folder	generated in app	directory where the images are stored when the storage is 'device'

Locally

To execute the application locally, it is enough to generate the binary that contains our application. To do so, inside the cmd/server directory of our project, execute::

go build -o cmd/server ./...

Once the binary has been generated and environment variables has been set, we can deploy our server locally by running:

./cmd/server/server

Using docker

Another alternative to deploy the server is through docker. For this it is necessary to have the tool installed. Instructions for this can be found here. Once docker is installed, follow the next steps:

Build docker image (inside repository folder)

docker build -t <image-name> .

Run the docker image

docker run -p 8080:8080 --env-file=./.env <image-name>

NOTE: Fill .env file (attached .env.sample) with the desired values.

Observations

NOTE: Tests not implemented due to the limited business logic of the application itself.