travel_finder

TRAVEL_FINDER

Travel Finder is a simple application where the goal is to find the cheapest trip between two points doesn't matter how many hops. The solution implemented is based in theShortest path algorytm.

Project structure

The project follows the Clean architecture where deliveries and repositories are the external interfaces, services are the use cases and entity is the business rules.

Prerequisites

There are two ways to build the project: one is using Golang version >= 1.14 and the second is using Docker.

How to use it

For your convenience you can use make commands to run the tests and the projects.

Running

You can execute the app with the following commands:

make run database=database.csv
or
make docker-run database=database.csv

If the binary was already built, you can run using it:

./travel_finder database.csv

Build

If you would like to only build and not run the application, you can use:

make build
make docker-build

Tests

The same way from other commands, you can run the tests using the make commands:

make tests
make docker-tests

You can also see the code coverage info typing:

make coverage

It will open a webpage with the coverage data.

Rest API

The API communication is given through port number 8080. There are two routes for this API.

Insert new route:

To insert a new route you should use the request below:

POST /origins/:origin/destinations/:destination/values/:value

Sample:

curl -X POST -v http://localhost:8080/origins/gru/destinations/fln/values/10

Response codes:

• 201
• 400
• 500

Find the cheapest route:

To find the cheapest route between two points, you should use the request:

GET /origins/:origin/destinations/:destination

Sample:

curl -v http://localhost:8080/origins/gru/destinations/orl

Response codes:

• 200
• 400
• 404
• 500

Room for improvements

Some improvements points to be implemented in future versions:

• Create a proper API documentation using swagger;
• Refactor the package command to be able to implement unit test;
• Handle with the current routes instead of duplicate them;
• Use a database instead of a .csv file;