Closed BitcoinZavior closed 2 months ago
My big concern is why there is a
uniffi
feature separate from what integration tests appear to be using. If the feature is to distinguish between the uniffi bindings here and the flutter (and other language bindings?) Then I wonder why they live in this crate rather than where they're consumed. Perhaps this is the well reasoned best design and some repetition is ok. I don't know, it just smelled to me on first blush.Other than this this looks ok to me. I can't believe how complete these bindings seem to be 🔥
The separation of the uniffi
feature is deliberate, addressing its unique requirements—like wrapping complex structs with Arc
for shared parameters and working around the lack of direct callback support. While this approach introduces some redundancy, it isolates these limitations from the core codebase, preventing them from affecting testing or Flutter integration. I plan to further reduce code duplication in future upgrades.
Other than this this looks ok to me. I can't believe how complete these bindings seem to be 🔥
Thanks! 🧡
[0.20.0]
APIs added
v2
tov1
sends possible.APIs changed
contribute_non_nitness_input
fromv1
&v2
.payjoins
out of sweep transactions (#259).