typhon VS decimal

I've created StormRPC an abstraction or wrapper on NATS Request/Reply messaging capabilities. I'm a fan of the NATS project and wanted to see what it would look like to build a higher level RPC mechanism on top of it. This project is inspired by gRPC and monzo's Typhon.

The communicate via protobuf RPCs, which integrates with our own RPC framework typhon.

Try it in JS with console.log(4.225 * 100);, you'll see the same unexpected result. Depending on your needs, it's actually fine to just be off by some amount. If it does really matter, then use an arbitrary precision library like https://github.com/shopspring/decimal to get the results you need.

Most programming languages ( JavaScript, PHP, Go, Python, Java, C# ) have built-in support or 3rd-party libraries for handling this data type, even though the internal implementations can differ:

while working with money, dont use floats, use something like https://github.com/shopspring/decimal (go std lib doesnt have its own decimals yet)

Integers, or something like this: https://github.com/shopspring/decimal

I make heavy use of caching, and was caching both in-memory and in Redis using JSON to convert the data in to a string. The struct itself isn't super complicated, one level nested, and some fields use the shopspring/decimal library.

I actually use this to represent monetary values btw :)

Check out https://github.com/shopspring/decimal for doing money related calculations.

We personally use https://github.com/shopspring/decimal and it works well.

decimal, one of the first Go library that created to work with decimal and monetary value.

https://github.com/shopspring/decimal also.