cingicanga-npwp-match

index.js

// Usage : use Node 8.x, pipe the output to empty csv file.
var fs = require('fs');
var async = require('async');
var byline = require('byline');
var stringSimilarity = require('string-similarity');
var sqlite3 = require('sqlite3').verbose();  

var db = new sqlite3.Database('data.db');  
var stream = fs.createReadStream('A2914.csv');
db.serialize(function() { 
  stream = byline.createStream(stream);
  stream.on('data', function(line){
    var date = line.toString().split(';')[1];
    var name = line.toString().split(';')[2];
    var q = 'select * from MF where birth_date=\'' + date + '\';';
    //console.log(line.toString());
    //console.log(date);
    //console.log(q);
    db.all(q, function(err, rows) {
      if (rows.length < 1) {
        console.log(line.toString() + ';-;-;-');
        return;
      }
      var bestMatchScore = 0;
      var bestMatch = "";
      var bestMatchNPWP = "";
      async.eachSeries(rows, function(row, cb){
        //console.log(row.name);
        var score = stringSimilarity.compareTwoStrings(name, row.name);
        if (bestMatchScore < score) {
          bestMatchScore = score;
          bestMatch = row.name;
          bestMatchNPWP = row.id;
        }
        cb();
      }, function(){
        console.log(line.toString() + ';' + bestMatch + ';' + bestMatchScore + ';' + bestMatchNPWP);
      })
    });
  });
});

package.json

{
  "name": "cingicanga-npwp-match",
  "version": "1.0.0",
  "description": "",
  "main": "index.js",
  "scripts": {
    "test": "echo \"Error: no test specified\" && exit 1"
  },
  "author": "",
  "license": "ISC",
  "dependencies": {
    "async": "^2.6.0",
    "byline": "^5.0.0",
    "csv2sql-lite": "0.0.6",
    "sqlite3": "^3.1.13",
    "string-similarity": "^1.2.0"
  }
}