Script que agrupa en una serie de tiempo, los reportes diarios del gobierno sobre COVID-19. De paso deja los estados en formato compatible con el Marco Geoestadístico de INEGI (i.e. listo para mapas). Referencia coronavirus.gob.mx via https://github.com/guzmart/covid19_mex/issues/2.

covid19mx-to-jsonseries.js

(async () => {
  const assert = require('assert')
  const path = require('path')

  const aliases = {
    'ciudad de mexico': 'ciudad de méxico',
    'coahuila': 'coahuila de zaragoza',
    'michoacan': 'michoacán de ocampo',
    'michoacán': 'michoacán de ocampo',
    'queretaro': 'querétaro',
    'veracruz': 'veracruz de ignacio de la llave'
  }

  const { promisify } = require('util')
  const fs = require('fs')
  const args = process.argv.slice(2)

  let db = {}
  const dirname = args[0]
  const filenames = await promisify(fs.readdir)(dirname)
  const csvs = filenames.filter(f => f.endsWith('csv'))

  for (let csvname of csvs) {
    // Parse results and append to database
    const res = await parseFile(path.join(dirname, csvname))
    db = { ...db, ...res }
  }

  console.log('%j', db)

  async function parseFile (filename) {
    const series = {}
    const states = [
      'aguascalientes',
      'baja california',
      'baja california sur',
      'campeche',
      'coahuila de zaragoza',
      'colima',
      'chiapas',
      'chihuahua',
      'ciudad de méxico',
      'durango',
      'guanajuato',
      'guerrero',
      'hidalgo',
      'jalisco',
      'méxico',
      'michoacán de ocampo',
      'morelos',
      'nayarit',
      'nuevo león',
      'oaxaca',
      'puebla',
      'querétaro',
      'quintana roo',
      'san luis potosí',
      'sinaloa',
      'sonora',
      'tabasco',
      'tamaulipas',
      'tlaxcala',
      'veracruz de ignacio de la llave',
      'yucatán',
      'zacatecas'
    ]

    const input = await promisify(fs.readFile)(filename)
    const data = input.toString().trim()
    const entries = data
      .split('\n')                  // split by rows
      .slice(1)                     // skip header
      .filter(l => l.length)        // filter out empty lines
      .map(l => l.split(','))       // break columns

    for (let entry of entries) {
      // State name found in the data file
      let stateName = entry[1]
        .toLowerCase()
        .replace(/"/g, '')

      let match
      let matchIndex = states.indexOf(stateName)
      if (matchIndex === -1) {
        matchIndex = states.indexOf(aliases[stateName])
        if (matchIndex === -1) {
          console.log(-1, stateName)
          continue
        }
        match = aliases[stateName]
      } else {
        match = stateName
      }

      // Find the CVE_ENT code for the match
      const code = String(matchIndex + 1).padStart(2, '0')

      // Create key if it doesn't exist yet
      if (!series[code]) {
        series[code] = series[code] || {}
        series[code]['NOM_ENT'] = match
        series[code]['CVE_ENT'] = code
        series[code]['CASOS'] = 0
      }

      series[code]['CASOS'] += 1
    }

    const total = Object
      .values(series)
      .map(o => o.CASOS)
      .reduce((p, c) => p + c, 0)

    assert.strictEqual(total, entries.length, filename)

    const date = filename.substr(filename.lastIndexOf('_') + 1, 8)
    return { [date]: series }
  }
})()