rust-bitcoin

2026-06-21 automated rustfmt nightly

Fix typo in p2p decoding error docs

Signed-off-by: haoshengzhen <haoshengzhen@outlook.com>

Update API files

Add and move tests to kill mutants

With the move of Builder and the script functions to primitives, new
tests are needed to ensure sufficient coverage of the code and mutants.

Introduce tests to builder and owned modules and move tests from
bitcoin to primitives to cover new logic and mutants.

Assisted by: Claude Opus 4.8 for new tests.

Remove From impls on FromStrError

Since the error type is being stabilised, we want to eliminate needless
From impls to minimise the API surface. These From impls can be
trivially replaced by map_err calls, and so should be removed.

Remove From<ParseIntError> and From<TryFromError> for FromStrError.

Change match to *self in Display and Error

The std::error::Error and Display impls should match on a deref'ed copy
of the error, rather than on self.

Adjust docs on WitnessVersion

primitives: Update API files

Add tests for WitnessVersion and Opcode

The new WitnessVersion logic introduces mutants in primitives. In order
to both test the logic and kill the mutants, new tests are needed.

Add tests to WitnessVersion module in primitives.

Fix lint errors in primitives

The move of bitcoin code to primitives has revealed some lint errors.

Fix lint errors in the new code in primitives.

Move Builder to primitives

With the BuilderExt split complete, the Builder type can now be cleanly
moved to primitives.

Move Builder to primitives and re-export in bitcoin.

Move WitnessVersion to primitives

As part of moving Address to addresses and the script functionality to
primitives, WitnessVersion will both also need to be available to both.

Move WitnessVersion and associated errors to primitives, re-exporting
from bitcoin.

Move ScriptBuf extension trait functions to primitives

Following the move of Opcode to primitives, a handful of ScriptBufExt
functions can now be moved to primitives on the main type. These
functions will be used in Builder.

Move push_slice, push_slice_non_minimal, push_opcode to ScriptBuf.
Duplicate private extension functions on ScriptBuf and copy
ScriptBufAsVec to primitives.

Remove access to private fields of Builder

In preparation for the upcoming split of Builder into a BuilderExt
trait, various functions which rely on internal access to the private
fields of Builder to function need to change.

Adjust functions to be left behind to not rely on internal field access
of Builder.

Move push_int_non_minimal to BuilderExtPriv

The push_int_non_minimal function on Builder is not supposed to be
public, nor should it be moved to primitives. As such, it should be
split to a private extension trait to be used in crate::blockdata.

Add BuilderExtPriv trait and move push_int_non_minimal to it.

Split Builder into BuilderExt

The Builder type has various functions that rely on unstable types or
provide an unclean interface. Since we don't want these in primitives,
they should be split to an extension trait to be left behind.

Split Builder into BuilderExt

Remove the second field of Builder

The second field in the Builder struct, the Option<Opcode>, is
exclusively used to optimise the push_verify function. Since retaining
it greatly complicates the process of moving Opcode to primitives, it
should be removed.

Remove the Option<Opcode> field of Builder.

fee_rate: fix mul_by_weight to use ceil not floor

fee_rate: add test for mul_by_weight rounding

fix: InstructionIndices::nth byte position bug

Remove the custom `nth` implementation for `InstructionIndices` because
it captures position via `next_with` and returns as byte offset instead of advancing it in later next_fn calls.

Fixes #6377

test: adding test for InstructionIndices::nth byte position bug

Remove the custom `nth` implementation for `InstructionIndices` because
it captures position via `next_with` and returns as byte offset instead of advancing it in later next_fn calls.

Adding test to show that nth position is not in sync with returned opcode value, no real fix in the commit

Fixes #6377

Use core::cmp in units kani verification

In the units::amount::verification kani tests make use of cmp for min
and max calculations. These are imported from std, even though they
could be imported from core. In some unpredictable cases, this causes
kani to choke on a lack of std. Since there is no functionality loss,
it's better to use core::cmp in place of std::cmp and avoid the problem
entirely.

Replace use std::cmp with use core::cmp in verification.rs of
units::amount.

Add regression test for from_base64 bug

To prevent changes to the from_base64 function reintroducing the non-65
byte invalid behaviour, a regression test should be added to catch any
changes.

Introduce regression test for
sign_message::MessageSignature::from_base64.

Fix base64 panic in from_base64

In sign_message, the from_base64 function can panic if the string has
both a length of 88 bytes, and decodes to an array of bytes > 65. This
can happen for a string of 88 "A" characters, for example. Further, the
function incorrectly parses base64 strings that produce 64 byte arrays.
While the latter is likely caught by the underlying secp parsing, both
should be checked to provide a more meaningful error return.

Add checks for non-65 byte base64 decoding in from_base64, returning
an InvalidLength error for 64 or 66 byte decodes. Also prevent panic on
66 byte decode.

Fix buffer bug in encoding::decode_from_read_unbuffered_with

In decode_from_read_unbuffered_with, the clamped buffer is pushed into
the decoder after a successful read. Some decoders do not completely
consume all provided bytes in a single push_bytes call, which can lead
to parts of the buffer being dropped. Like encoding::decode_from_hex,
the push_bytes should instead be called in a loop until the buffer is
empty, or the error case occurs.

Call push_bytes in a loop to prevent buffer bug in
encoding::decode_from_read_unbuffered_with.