parsing a formdata object with javascript

Best would be to use node-formidable, browserify and polyfill it. Here is a standalone parser, works both with string and raw responses. Make sure you use a modern browser for the raw-stuff.

/* 
 * MultiPart_parse decodes a multipart/form-data encoded response into a named-part-map.
 * The response can be a string or raw bytes.
 *
 * Usage for string response:
 *      var map = MultiPart_parse(xhr.responseText, xhr.getResponseHeader('Content-Type'));
 *
 * Usage for raw bytes:
 *      xhr.open(..);     
 *      xhr.responseType = "arraybuffer";
 *      ...
 *      var map = MultiPart_parse(xhr.response, xhr.getResponseHeader('Content-Type'));
 *
 * TODO: Can we use https://github.com/felixge/node-formidable
 * See http://stackoverflow.com/questions/6965107/converting-between-strings-and-arraybuffers
 * See http://www.w3.org/Protocols/rfc1341/7_2_Multipart.html
 *
 * Copyright@ 2013-2014 Wolfgang Kuehn, released under the MIT license.
*/
function MultiPart_parse(body, contentType) {
    // Examples for content types:
    //      multipart/form-data; boundary="----7dd322351017c"; ...
    //      multipart/form-data; boundary=----7dd322351017c; ...
    var m = contentType.match(/boundary=(?:"([^"]+)"|([^;]+))/i);

    if ( !m ) {
        throw new Error('Bad content-type header, no multipart boundary');
    }

    var boundary = m[1] || m[2];

    function Header_parse(header) {
        var headerFields = {};
        var matchResult = header.match(/^.*name="([^"]*)"$/);
        if ( matchResult ) headerFields.name = matchResult[1];
        return headerFields;
    }

    function rawStringToBuffer( str ) {
        var idx, len = str.length, arr = new Array( len );
        for ( idx = 0 ; idx < len ; ++idx ) {
            arr[ idx ] = str.charCodeAt(idx) & 0xFF;
        }
        return new Uint8Array( arr ).buffer;
    }

    // \r\n is part of the boundary.
    var boundary = '\r\n--' + boundary;

    var isRaw = typeof(body) !== 'string';

    if ( isRaw ) {
        var view = new Uint8Array( body );
        s = String.fromCharCode.apply(null, view);
    } else {
        s = body;
    }

    // Prepend what has been stripped by the body parsing mechanism.
    s = '\r\n' + s;

    var parts = s.split(new RegExp(boundary)),
        partsByName = {};

    // First part is a preamble, last part is closing '--'
    for (var i=1; i<parts.length-1; i++) {
      var subparts = parts[i].split('\r\n\r\n');
      var headers = subparts[0].split('\r\n');
      for (var j=1; j<headers.length; j++) {
        var headerFields = Header_parse(headers[j]);
        if ( headerFields.name ) {
            fieldName = headerFields.name;
        }
      }

      partsByName[fieldName] = isRaw?rawStringToBuffer(subparts[1]):subparts[1];
    }

    return partsByName;
}

In modern browsers you can use the built-in Response.formData() method:

const multipartPayload =
`------WebKitFormBoundaryU5rJUDxGnj15hIGW\r
Content-Disposition: form-data; name="fileToUpload"; filename="test.csv"\r
Content-Type: application/vnd.ms-excel\r
\r
1
2
3
4
5
...
\r
------WebKitFormBoundaryU5rJUDxGnj15hIGW--`

const boundary = multipartPayload.slice(2, multipartPayload.indexOf('\r\n'))
new Response(multipartPayload, {
  headers: {
    'Content-Type': `multipart/form-data; boundary=${boundary}`
  }
})
  .formData()
  .then(formData => {
    // The parsed data
    console.log([...formData.entries()])
    
    // In this particular example fileToUpload is a file,
    // so you need to call one more method to read its context
    formData.get('fileToUpload')
      .text()
      .then(text => console.log(text))
  })

The \r in the multipart payload are necessary, because line breaks must be \r\n, except the sub-payloads themselves. If you have a properly formed multipart/form-data payload, you don't need to add \r.

If you want to parse an HTTP response, you can use the fetch response directly:

fetch('/formdata-response')
  .then(response => response.formData())
  .then(formData => console.log([...formData.entries()]))