Alessandro Coglio
Alessandro Coglio
Nice! So it shouldn't take long to check this one off.
[Sharing the following from a Slack discussion with @damirka.] ABNF does not have an exclusion operator, but there are a couple of ways to do that: 1. When the number...
@collinc97 Since the restrictions are easy to capture in the grammar, I would go with the stricter rules. It could be a separate question whether we also change the parser...
I think we discussed this at a Wednesday sync. If I recall correctly, there was a preference towards doing all the checks on addresses post-parsing; some checks cannot be captured...
@howardwu @collinc97 What should we do for this PR?
Which types are we going to use for bits, bytes, and their sequences? A natural choice for bytes is `u8`. The closest (isomorphic, in fact) type to bits we have...
I'll make a separate RFC for type wildcards. (Not sure there's a strong use case, but it seems worth considering, maybe in the future.) Understood about the underlying Rust gadgets....
I was also wondering about endianness. Would the bits/bytes be big endian or little endian? Should both options be provided? E.g.: ``` 258:u32 --> [0 0 1 2] // big...
My first instinct would be to have "symmetry", e.g. `big_endian_bytes()` and `little_endian_bytes()`, or just `bebytes()` and `lebytes()`, but it is more verbose, especially if there is a more common case,...
By the way, having two possible conversions (big vs. little endian) reinforces the fact that this should not use a type cast syntax (against my proposal earlier in this message...