Convert a ginormous table of corrections data.

convert-rhu-data.js

const fs = require('fs');
const Json2csvParser = require('json2csv').Parser;

let output;

// Read in a giant CSV file.
fs.readFile('input.json', 'utf8', (err, data) => {
  if (err) throw err;
  output = JSON.parse(data);
  parse(output);
});

// Returns a new array deleting the unnecessary elements.
function parse(items) {
  const filtered = items.map(x => {
    delete x['DateOut'];
    delete x['housing_status'];
    return x;
  });
  countIncidents(filtered);
}

// Count the number of incidents per de-identified inmate ID.
function countIncidents(arr) {
  let map = {};

  // Loops over the rows and matching by ID.
  // if there's a repeat in the ID, it increases a counter
  // rather than creating a new row for the inmate/incident.
  for (let i = 0; i < arr.length; i++) {

    // If no previous entry, then assign them to the
    // map by ID.
    if (map[arr[i]['ID']] === undefined) {
      map[arr[i]['ID']] = {
        count: 1,
        data: arr[i]
      };
    } else {
      // If that particular ID has already been encountered,
      // instead, increase counter to append to column labels
      // and append properties to the original inmate/incident object.
      map[arr[i]['ID']]['count'] += 1;
      let count = map[arr[i]['ID']]['count'];
      map[arr[i]['ID']]['data'] = {
        ...map[arr[i]['ID']]['data'],
        ['DateIn' + count]: arr[i]['DateIn'],
        ['facility' + count]: arr[i]['facility'],
        ['DaysInRhu' + count]: arr[i]['DaysInRhu']
      };
    }
  }
  return stripExtras(map);
}

// Recursion === max call stack limit :\

// function countIncidents(arr, map = {}) {
//   if (arr.length !== 0) {
//     const incident = arr.shift();
//     if (map[incident['ID']] === undefined) {
//       map[incident['ID']] = {
//         count: 1,
//         data: incident
//       };
//       return countIncidents(arr, map);
//     } else {
//       map[incident['ID']]['count'] += 1;
//       let count = map[incident['ID']]['count'];
//       map[incident['ID']]['data'] = {
//         ...map[incident['ID']]['data'],
//         ['DateIn' + count]: incident['DateIn'],
//         ['facility' + count]: incident['facility'],
//         ['DaysInRhu' + count]: incident['DaysInRhu']
//       };
//     }
//     return countIncidents(arr, map);
//   }
//   return stripExtras(map);
// }

// Goes through all the newly returned objects and
// removes the "count" property and just passes
// over the contents of "data."
function stripExtras(ob) {
  const output = [];
  Object.keys(ob).forEach(key => {
    output.push(ob[key].data);
  });
  return csvify(output);
}

// function jsonify(input) {
//   fs.writeFileSync('./output.json', JSON.stringify(input));
//   console.log('All done!');
// }

// Uses Json2csv to convert resulting json file into CSV.
// Then save the resulting data to the directory using fs.
function csvify(json) {
  const transformOpts = { highWaterMark: 16384, encoding: 'utf-8' };

  try {
    const parser = new Json2csvParser(transformOpts);
    const csv = parser.parse(json);
    fs.writeFileSync('./output.csv', csv);
    console.log('Conversion complete, fucker.');
  } catch (err) {
    console.error(err);
  }
}