xpe / q-num Goto Github PK

It would be nice to complement the generated From<f64> implementation with a TryFrom<f64> version that allows runtime handling of overflow translation rather than having the risk of panic in deployed code.

Thanks for the great library! Exactly what I needed!

I have an application using an ADC where the documentation claims the I/Q samples are each 8-bit "Q7" fixed point values, used to represent values in [-1.0, 1.0). (The documentation is a bit confusing and perhaps even inconsistent, and I'm new to this domain).

Using ARM notation, I'm assuming this means Q1.7, but when I attempt this with q-num, I get the following error

error: proc macro panicked
   --> txrx.rs:134:1
    |
134 | define_q_num!(Q7, Q1.7);
    | ^^^^^^^^^^^^^^^^^^^^^^^
    |
    = help: message: assertion failed: input.int_bits >= 2

Should this be a valid encoding, or am I misunderstanding something?