You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Merge #849: several improvements and bugfixes to uncompressed key handling and sortedmulti
b5db340 sortedmulti: take Thresh as input to constructors (Andrew Poelstra)
276cf94 sortedmulti: simplify constructor and improve unit test (Andrew Poelstra)
3250e40 miniscript: take pubkey size into account in multi() type extdata (Andrew Poelstra)
8874640 miniscript: fix ExtData to correctly account for uncompressed keys (Andrew Poelstra)
040c29d miniscript: remove MalleablePkH script context rule (Andrew Poelstra)
a5c477b clippy: fix all nightly lints (Andrew Poelstra)
Pull request description:
Removes the weird and incorrect `MalleablePkH` script context rule; fixes several bugs related to key size estimation in pkh and multisig fragments; improves tests of sortedmulti; uses the `Thresh` type in sortedmulti constructors.
These bugs are obscure and only visible when using uncompressed pubkeys and I don't think they're worth backporting.
This PR is starting to make "real" changes toward unifying our validation parameters. In particular we separate out validation errors in the `sortedmulti` constructors from threshold-construction errors (which the user is forced to deal with before calling a `sortedmulti` constructor). The PR is bigger than you might think because when testing this separation I found some bugs.
This PR also includes a clippy fix which should unstick #805 (update nightly version) which has been stalled for months because the weekly retries don't trigger Github notifications and I didn't notice it.
ACKs for top commit:
sanket1729:
utACK b5db340
Tree-SHA512: 4c221f4efe1014001d4bbb99fd89728f5b56348c09d34df9163ff01766cd11e5d13834ecb3bc7ceb49ae89a99d6149587d83b0d901f2b73c64cc91c9a47684fc
0 commit comments