In order for efficient server side parsing I am looking into a BSON solution directly for the browser javascript environment. The idea is to utilize the entire ASCII space by me
This might be incomplete but the goal of the project line up with what you want: https://github.com/muhmi/javascript-bson It doesn't look like that encodes directly to typed arrays which would be the most useful for sending over WebSocket.
For what it's worth, it appears that the MongoDB team now have a supported Javascript BSON project:
https://github.com/mongodb/js-bson
I'm no expert with the library, but the project claims to work in both Node and the browser. Below is a modified sample from their site:
<head>
<!-- Originally https://raw.github.com/mongodb/js-bson/master/browser_build/bson.js -->
<!-- But downloaded and hosted locally -->
<script src="./bson.js"></script>
</head>
<body onload="start();">
<script>
function start() {
var BSON = bson().BSON;
var Long = bson().Long;
var doc = {
oid: bson().ObjectID(),
long: Long.fromNumber(100),
date: new Date(),
string: "js-bson sample",
obj: {
string: "Object within an object"
}
}
console.log("doc %o", doc);
// Serialize a document
var data = BSON.serialize(doc, false, true, false);
console.log("data %o", data);
// De serialize it again
var doc_2 = BSON.deserialize(data);
console.log("doc_2 %o", doc_2);
}
</script>
</body>
Below are my results in Chrome:
Here in 2019 modern browsers already have embedded the functions:
btoa("string to be coded on base64 format")
atob("base64 string to be uncoded")
As you can see here: https://caniuse.com/#search=btoa All browsers listed have this feature.