nsa-yoda/emx-test

EMX Code Test

Thank you for the opportunity to work on and submit this code test.

Path to Solution

Initially, I thought that the solution was based on rules - for instance a < in one cell would dictate which characters would be in the surrounding cells.

However, upon testing, I quickly realized that this was not the case. So, I took to pen and paper. I translated each character to a numeric value, as follows:

=   2
>   1
<  -1
-   0

Where 0 would be considered an empty cell. I then translated a few puzzle problems as follows:

Initial translations:

Further translations:

This led me to see that the = were dividers, and each cell's lower mirror was the opposite value. Additionally, the character in the cell on each row that was pre-filled dictated which characters were to be on that line (if >, then left would be < and to the right would be > in most cases, for instance).

The rest was simply a matter of coding those rules into the matrix manipulation function.

Structure

There is an index.php which acts as the gateway. It's a simple service, so I saw no need for an overkill application based on a framework like Laravel.

Additionally, I wrote helper.php, which contains two classes - Log, and Helper (which extends Log for ease of use).

All of the data is stored in a data.json file, though in a broader application it would be stored in a database (sample schema below), though the treatment would be the same.

Possible MySQL Schema for data.json

TABLE STRUCTURE FOR: users

CREATE TABLE `users` (
  `id` int(11) unsigned NOT NULL AUTO_INCREMENT,
  `uuid` varchar(64) NOT NULL,
  `username` varchar(100) NOT NULL,
  `password` varchar(255) NOT NULL,
  `created_at` datetime NOT NULL,
  PRIMARY KEY (`id`),
  KEY `username` (`username`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

TABLE STRUCTURE FOR: answers

CREATE TABLE `answers` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `user_id` int(11) NOT NULL,
  `question_id` int(11) unsigned NOT NULL,
  `answer` varchar(255) NULL,
  `created_at` datetime NOT NULL,
  UNIQUE KEY `id` (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

TABLE STRUCTURE FOR: questions

CREATE TABLE `questions` (
  `id` int(9) unsigned NOT NULL AUTO_INCREMENT,
  `key` varchar(32) NOT NULL
  `question` varchar(255) NULL,
  `created_at` datetime NOT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

Where a query for selecting an answer based on a q value of 'Years', might go as follows:

SELECT
    a.answer
FROM
    users u
    INNER JOIN answers a on a.user_id = u.id
    LEFT JOIN questions q on q.id = a.question_id
WHERE
    u.uuid = ?
    AND q.key = ?

Where the first placeholder would be the user ID for a specific user ( assuming all answers were publicly accessible if the UUID is passed in), and the second placeholder would be the q portion of the query (e.g. 'Years')