Math API Web Service

This repository is a golang web service that exposes REST HTTP APIs to perform math operations like min, max, average, median, percentile against the list of numbers. It uses Gorilla Mux package to implement a request router and dispatcher for matching incoming requests to their handlers. The web service can be connected through port 8080 on localhost.

API endpoints

Get min number(s) based on the quantifier and the list of numbers

• Method: POST
• URI: /min
• Content-Type: application/json
• Request Body:

Paramter	Description
numList	Array of numbers
quantifier	Number, (how many)

Get max number(s) based on the quantifier and the list of numbers

• Method: POST
• URI: /max
• Content-Type: application/json
• Request Body:

Paramter	Description
numList	Array of numbers
quantifier	Number, (how many)

Get average on the list of numbers

• Method: POST
• URI: /avg
• Content-Type: application/json
• Request Body:

Paramter	Description
numList	Array of numbers

Get median on the list of numbers

• Method: POST
• URI: /median
• Content-Type: application/json
• Request Body:

Paramter	Description
numList	Array of numbers

Get qth percentile based on the quantifier 'q' and the list of numbers

• Method: POST
• URI: /percentile
• Content-Type: application/json
• Request Body:

Paramter	Description
numList	Array of numbers
quantifier	Number, (q)

Assumptions

This utility assumes that it contains enough memory space to store all the resources. The percentile is calculated using Nearest-Rank method.

Prerequisites

Before you start, please make sure you have Go installed in your system. If not, please use the following link to install Golang: https://golang.org/doc/install

Installation

Clone the git repository in your system and then cd into project root directory

$ git clone https://github.com/RashikaAggarwal/math-api-webservice.git
$ cd math-api-webservice

Build your tool by executing the following steps

$ cd application
$ go run .

The server runs on http://localhost:8080 and can be hit through any client like Postman.

Examples

This utility takes input request in the form of json body. See below examples

Get min number(s) based on the quantifier and the list of numbers

http://localhost:8080/min

Request Body

{
    "numList": [1,5,9,3,7.9],
    "quantifier": 2
}

Success Response

[1 3]

Failure Response

For example, In case of Quantifier passes greater than the length of array

Quantifier must be less than equal to the length of list of numbers.

Get max number(s) based on the quantifier and the list of numbers

http://localhost:8080/max

Request Body

{
    "numList": [1,5,9,3,7.9],
    "quantifier": 2
}

Success Response

[9 7.9]

Failure Response

For example, In case of empty quantifier

Incomplete input request parameters.

Get average against the list of numbers

http://localhost:8080/avg

Request Body

{
    "numList": [1,5,9,3,7.9]
}

Success Response

5.18

Failure Response

For example, In case of empty array of numbers

Incomplete input request parameters.

Get median against the list of numbers

http://localhost:8080/median

Request Body

{
    "numList": [1,5,9,3,7.9]
}

Success Response

5

Failure Response

For example, In case of invalid numbers of array

Invalid input request.

Get qth percentile based on the quantifier(q) and the list of numbers

http://localhost:8080/percentile

Request Body

{
    "numList": [1,5,9,3,7.9],
    "quantifier": 75
}

Success Response

7.9

Failure Response

For example, In case of quantifier greater than 100th percentile

Quantifier must be less than equal to the 100th percentile.

Unit Testing

This repository contains unit test cases which provide the industry standard code coverage. The test cases can be executed using following commands:

$ cd math-api-webservice/application
$ go test

To get the code coverage, use the following command:

$ go test -cover

RoadMap

• Handling of invalid routing requests.
• Response Body in a formatted structure
• Response codes and messages maintained in configuration file for success and failure scenarios.
• Include Logging mechanism using packages like "log" to trace the entire API call for monitoring and debugging purposes.
• Secure the APIs through tokens and authorization purposes.
• Handling of concurrent user requests.
• Deploy the application using Nginx server to balance the load and handle server crashing.
• Include Docker to deploy the application on Kubernetes cluster to handle vertical and horizontal scaling effectively.