Represent line levels as a bool

[fpagliughi]

I know the char driver(s) expose bits as u8 values, but it might be easier to manipulate the values as booleans. Should we consider:

impl LineHandle {
    pub fn get_value(&self) -> Result<bool> { ... }
    pub fn set_value(&self, value: bool) -> Result<()> { ... }
    ...

[posborne]

The actual value (0, non-zero) is modeling "active" state for the line which can be inverted in the kernel via the active-low setting. I could see is_active and set_active as an API that modeled what this is in the kernel being something that make sense. This could be in addition to the value or as a full replacement.

[fpagliughi]

Whenever I wrap a C library or API in a language that has a bool type (C++, Rust, etc), I usually prefer to convert the C integer boolean (0, non-zero) to a bool to be a little more explicit about the possible values, and remove ambiguities. But it's no big deal either way.

[rubberduck203]

Possibly relevant. I've been working on a reusable abstraction for logic levels.
The idea is to abstract over whether an I/O line is active high or low and only expose whether or not it is currently asserted.

https://github.com/rubberduck203/switch-hal

[ryankurte]

i can't find it right now but i believe there is discussion about introducing a LogicLevel (or similar) enum along with some more gpio accessors somewhere amongst our issues and PRs.

[ajmull]

I do think that it would have advantages as a bool, since the returned value will virtually never not be 0 or 1. It would allow things like if input.get_value(), and one could always just cast it to an integer if one was so inclined. However, I do also see why it may be better as an integer.