Dex-parse is a node.js library for parsing raw vending machine audit data in the DEX/UCS format into plain JS objects. Currently supports basic product sales and machine data, with support for remaining fields planned as needed.

DEX data is new line (\n) delimited text where each line contains 2 or 3 character prefix denoting the type of record. Fields within a record are delimited by '*', and the interpretation depends on the type of record.

DXS*WTN4213952*VA*V0/6*1
ST*001*0001
ID1*WTN11082110074*GVC1        *8207***
ID4*2*001*5
CB1*11082110074*GVC1        *8207
PA1*010*50*****
PA2*6*350*6*350*0*0*0*0*0*0*0*0
PA3*2*150*2*150
PA4*0*0*0*0
PA5*20120301*125320*0

Output is a single JS object containing an array of products and machine data properties. Data is typed where possible; for example, price information is returned as Number.

Output for the example given above would be:

{
  "products": [
    {
      "name": "010",
      "price": 0.5,
      "sold": 6,
      "revenue": 3.5,
      "testVendCount": "2",
      "soldOutDate": "20120301",
      "soldOutTime": "125320",
      "soldOutCount": "0"
    }
  ],
  "machine": {
    "serialNumber": "WTN11082110074",
    "modelNumber": "GVC1",
    "buildStandard": "8207",
    "assetNumber": "",
    "controlBoard": {
      "serialNumber": "11082110074",
      "modelNumber": "GVC1",
      "softwareRevision": "8207"
    }
  }
}

Input data can be passed directly as string or read from a file.

var dexparse = require('dex-parse')

var mytext = "..."
dexparse.readText(mytext, function(err, data) {
    console.log(data.products[0].name)
})

dexparse.readFile('myfile.dex', function(err, data) {
    console.log(data.products[0].name)
})

Dex-parse is licensed under the MIT license. See LICENSE.txt.