Script to process all baptiste data

processingBaptisteData.js

(function() {
  "use strict";

  var fs = require('fs');
  var firebase = require("firebase");
  var ref = new firebase("https://xyz.firebaseio.com/baptiste");
  var request = require('request').defaults({ encoding: null });

  // Grab all teacher data
  var obj = JSON.parse(fs.readFileSync('teachers-part1.json', 'utf8'));
  var teachers = obj.articles;
  obj = JSON.parse(fs.readFileSync('teachers-part2.json', 'utf8'));

  teachers = teachers.concat(obj.articles);

  // now add in metafields
  obj = JSON.parse(fs.readFileSync('metafieldsTeachersv2.json', 'utf8'));
  var metafieldTeacherData = {};

  obj.forEach(function(metadata) {
    if(!metafieldTeacherData[metadata.resource_id]) { // create the obj if it doesn't exist yet
    metafieldTeacherData[metadata.resource_id] = {};
  }

  if(metadata.key == 'state-prov') {
    if(metadata.value) {
      metafieldTeacherData[metadata.resource_id].state = metadata.value;
    }
  } else if(metadata.key == 'studio-link') {
    if(metadata.value) {
      var studioName = metadata.value.match(/>(.*)</);

      if(studioName) { // if there is a URL embedded
        metafieldTeacherData[metadata.resource_id].studioName = studioName[1];
        metafieldTeacherData[metadata.resource_id].studioURL = metadata.value.match(/href=?\+? ?("|“)(.*)("|“|”)/)[2];
      }
      else {
        metafieldTeacherData[metadata.resource_id].studioName = metadata.value;
      }
    }
  } else if(metadata.key == 'city') {
    if(metadata.value) {
      metafieldTeacherData[metadata.resource_id].city = metadata.value;
    }
  } else if(metadata.key == 'country') {
    if(metadata.value) {
      metafieldTeacherData[metadata.resource_id].country = metadata.value;
    }
  }
});

function urlToBase64(url) {
  return new Promise((resolve, reject) => {
    request.get(`http:${url}`, function (error, response, body) {
      if (!error && response.statusCode == 200) {
        resolve("data:" + response.headers["content-type"] + ";base64," + new Buffer(body).toString('base64'));
      }
      else {
        resolve('todoManually');
      }
    });
  });
}

let promises = teachers.map(teacher => {
  return urlToBase64(teacher.summary_html.match(/src="(.*?)"/)[1])
  .then(base64Image => {

    let names = teacher.title.split(' ');

    let firebaseObj = {
      firstName: names.splice(0,1).toString(),
      lastName: names.splice(-1,1).toString(),
      middleNames: names.join(' '),
      description: teacher.body_html,
      image: base64Image,
      city: metafieldTeacherData[teacher.id].city,
      country: metafieldTeacherData[teacher.id].country,
      state: metafieldTeacherData[teacher.id].state,
      studioName: metafieldTeacherData[teacher.id].studioName,
      studioURL: metafieldTeacherData[teacher.id].studioURL,
      listInDirectory: {
        baptiste: true
      }
    };

    let pushRef = ref.push(firebaseObj);

    if(base64Image === 'todoManually') {
      console.log(`Need to manually add base64 image for ${teacher.title}`);
      //console.log(teacher);
      console.log(pushRef.key());
    }

    return firebaseObj;
  })
});

Promise.all(promises)
.then(results => {
//  console.log(results);
  console.log(`Created ${results.length} teacher profiles.`);
})
}());