const xlsx = require('xlsx');

const fs = require('fs');

const DecisionTree = require('ml-decision-tree');

const inquirer = require('inquirer');

inquirer

  .prompt([

    {

      type: 'input',

      name: 'fileName',

      message: 'Enter the name of the Excel file:',

    },

  ])

  .then(answers => {

    // Load the Excel workbook

    const workbook = xlsx.readFile(answers.fileName);

    // Select the first sheet

    const sheet = workbook.Sheets[workbook.SheetNames[0]];

    // Get the maximum number of rows in the sheet

    const maxRow = xlsx.utils.decode_range(sheet['!ref']).e.r;

    // Initialize a list to store the data from the sheet

    const data = [];

    // Loop through the rows in the sheet

    for (let row = 2; row <= maxRow; row++) {

      const game = [];

      for (let col = 65; col < 71; col++) {

        game.push(sheet[String.fromCharCode(col) + row].v);

      }

      data.push(game);

    }

    // Read the data from the JSON file

    let storedData = {};

    if (fs.existsSync('games.json')) {

      storedData = JSON.parse(fs.readFileSync('games.json', 'utf8'));

    }

    // Train the decision tree model

    const samples = storedData.samples || data;

    const labels = Array.from({ length: samples.length }, (_, i) => i % 60 + 1);

    const decisionTree = new DecisionTree(samples, labels, { gainFunction: 'gini' });

    // Generate 10 possibilities based on the decision tree model

    const possibilities = [];

    for (let i = 0; i < 10; i++) {

      possibilities.push(decisionTree.predict(data[data.length - 1]));

    }

    // Write the updated data to the JSON file

    storedData.samples = samples.concat(data);

    fs.writeFileSync('games.json', JSON.stringify(storedData));

    console.log(possibilities);

  });