Epf-parser NPM

epf-parser

Usage

'use strict';

const fs = require('fs');
const parse = require('epf-parser');

fs.createReadStream('/tmp/application')
  .pipe(parse((meta, rows) => {
    console.log('meta', meta);
    rows.on('data', row => console.log('row', row));
  }));

Will call back with meta like this, with all the meta k/v pairs in one object, and columns and column types merged:

meta { columns:
   [ { name: 'export_date', type: 'BIGINT' },
     { name: 'application_id', type: 'INTEGER' },
     { name: 'title', type: 'VARCHAR(1000)' },
     { name: 'recommended_age', type: 'VARCHAR(20)' },
     { name: 'artist_name', type: 'VARCHAR(1000)' },
     { name: 'seller_name', type: 'VARCHAR(1000)' },
     { name: 'company_url', type: 'VARCHAR(1000)' },
     { name: 'support_url', type: 'VARCHAR(1000)' },
     { name: 'view_url', type: 'VARCHAR(1000)' },
     { name: 'artwork_url_large', type: 'VARCHAR(1000)' },
     { name: 'artwork_url_small', type: 'VARCHAR(1000)' },
     { name: 'itunes_release_date', type: 'DATETIME' },
     { name: 'copyright', type: 'VARCHAR(4000)' },
     { name: 'description', type: 'LONGTEXT' },
     { name: 'version', type: 'VARCHAR(100)' },
     { name: 'itunes_version', type: 'VARCHAR(100)' },
     { name: 'download_size', type: 'BIGINT' } ],
  primaryKey: 'application_id',
  exportMode: 'INCREMENTAL' }

Will emit rows as data events like this (real data hidden for reasons):

row [ Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string'),
  Buffer('some string') ]

For debugging purposes, every row will have the original raw line attached as .raw.