Shop Rest API Microservice that provide following functionalities:

• Allow storage of shops in memory. Every time a shop is saved, the service calls the Google Maps API. The Google Maps API responds with the longitude and latitude, which allows the shop data to be updated with longitude and latitude.
• Find closest shop service: providing their current longitude and latitude (e.g. URL request params), and gets back the address, longitude and latitude of the shop nearest to them.

Java 1.8 or later.

You can consume the Rest microservice with any Rest client, I suggest to use Google Postman.

Request

POST /shops HTTP/1.1
Host: localhost:8080
Content-Type: application/json
Cache-Control: no-cache
Postman-Token: 70ec546f-8961-3816-d085-afef497bd203

{
  "shopName": "Simply Food 2",
  "shopAdress": {
    "number": "323-324",
    "street": "High Holborn",
    "postCode": "WC1V"
  }
}

Respose Body code 200 OK

{
  "shopName": "Simply Food 2",
  "shopAdress": {
    "number": "323-324",
    "street": "High Holborn",
    "postCode": "WC1V"
  },
  "geoPosition": {
    "lat": 51.51801520000001,
    "lng": -0.1123096
  },
  "_links": {
    "self": {
      "href": "http://localhost:8080/shops/Simply%20Food%202"
    }
  }
}

Request

GET /shops?latitude=51.52156840000001&longitude=-0.1123096 HTTP/1.1
Host: localhost:8080
Content-Type: application/json
Cache-Control: no-cache
Postman-Token: d71fe793-acc5-8624-0f56-9ee60ba35fd7

Respose Body code 200 OK

{
  "shopName": "Simply Food 2",
  "shopAdress": {
    "number": "323-324",
    "street": "High Holborn",
    "postCode": "WC1V"
  },
  "geoPosition": {
    "lat": 51.51801520000001,
    "lng": -0.1123096
  },
  "_links": {
    "self": {
      "href": "http://localhost:8080/shops/Simply%20Food%202"
    }
  }
}

Request

GET /shops/Simply%20Food%201 HTTP/1.1
Host: localhost:8080
Cache-Control: no-cache
Postman-Token: d9538aed-c7a1-dd12-e390-68e8a213e8c3

Respose Body code 200 OK

{
  "shopName": "Simply Food 1",
  "shopAdress": {
    "number": "323-324",
    "street": "High Holborn",
    "postCode": "WC1V"
  },
  "geoPosition": {
    "lat": 51.51801520000001,
    "lng": -0.1123096
  },
  "_links": {
    "self": {
      "href": "http://localhost:8080/shops/Simply%20Food%201"
    }
  }
}

Respose ERROR Body code 404 OK

{
  "code": "404",
  "message": "Shop not found"
}

You can add the library to your project with Gradle.

Pending to deploy in mavenCentral. It would look like this:

repositories {
    mavenCentral()
}

dependencies {
    compile 'com.db.challenge:shopsAPI:0.0.1'
    ...
}

Building the project
 $ ./gradle build
gradlew bootRun

# Compile and package the project
$ ./gradlew jar

Running the standalone service
 $ ./gradlew bootRun

Integration and unit tests has been created, the solution could be expanded with BDD, but this would required more time. Validations are performed in the controller with JSR-303 using @Validated annotation. The exceptions have been captured with a CustomHanlder.

To run the test:

# Run the tests
$  ./gradlew test

The solution could be expanded with the following features:

• Securing the service with org.springframework.boot:spring-boot-starter-security
• Persistence layer with JPA.
• JBehave BDD testing.

The application could be deployed in a CI server, such us Jenkins on TeamCicty, In the CI server we could create a task/build to launch both unit tests e integration test, after thise the service could be deployed in a VM/Host.

We could also implement BDD tests for intance with the Jbehave libray(BDD stories), after deploying with docker for instance we could create a trigger/task to launch the BDD test. Those tests should be end to end tests.

To control the build stability, we could either scheduled previous stepsin CI, or/and create a trigger, this trigger could be executed for instance every time that a colleague is pulling something into the master branch.

Enough testing coverage would make possible to avoid issues and increase the frequency we are releasing to PROD.

I recommend to deploy of the microsevices in cloud infrastructure, a private one using spring-cloud, or a public, such as Google cloud platform or AWS.

Those could also be integrated with Mesosphere that would be in charge of the service management/maintenance

We can deploy on cloud infrastructure, for instance creating containers to be run in VM/HostMachin with docker, and a system of cluster management like OpenShift.

Francisco San Roman For any question, contact me at [email protected]

This project is licensed under the GNU General Public License - see the LICENSE.md file for details