BIP 0340 followups #253
BIP 0340 followups #253
Conversation
apoelstra
force-pushed
the
2020-11--schnorrsig-followup
branch
from
fa26333
to
37049d7
Compare
| let err = ffi::secp256k1_xonly_pubkey_tweak_add_check( | ||
| secp.ctx, | ||
| tweaked_ser.as_c_ptr(), | ||
| if tweaked_parity { 1 } else { 0 }, |
There was a problem hiding this comment.
You can do tweaked_parity as u8
There was a problem hiding this comment.
Oh nice. Though I'd have to cast to a c_int ... with the export, the if statement is actually shorter, and in any case it's more explicit.
src/schnorrsig.rs
Outdated
| secp.ctx, | ||
| tweaked_ser.as_c_ptr(), | ||
| if tweaked_parity { 1 } else { 0 }, | ||
| &self.0 as *const _, |
There was a problem hiding this comment.
Can you replace with self.0.as_c_ptr()?
There was a problem hiding this comment.
I don't think so, this is a FFI type not a byte array
There was a problem hiding this comment.
References coerce to pointers, so you wouldn't actually need the cast: &self.0.
There was a problem hiding this comment.
lol I had no idea. Added a commit which fixes this across the library.
src/schnorrsig.rs
Outdated
| tweaked_key: &Self, | ||
| tweaked_parity: bool, | ||
| tweak: &[u8], | ||
| ) -> Result<(), Error> { |
There was a problem hiding this comment.
What do you think about passing tweak: [u8; 32] and then returning bool instead of a Result?
In general I feel like it would be nicer to use the type system to require sizes instead of adding dozens of results/errors (especially with TryFrom that let's you slice.try_into()? for converting to array, but that's 1.34)
There was a problem hiding this comment.
if for ergonomic pre 1.34 reasons you want to keep it as a slice then that's ok
There was a problem hiding this comment.
Yeah I think it'd be reasonable to use arrays here.
There was a problem hiding this comment.
Welll, for what it's worth we already have an error return here if the user passes an invalid tweak. If you know you're giving a hash then it can't fail, but that's still an error you have to .unwrap away.
I'm not sure there's any additonal ergonomics loss to using a slice.
Couple things on #237 that weren't done in the original PR.