The Transaction Size and Fees section says a P2MR witness is always 32
bytes smaller than an equivalent P2TR script path spend witness. The
control blocks are 1 + 32*m and 33 + 32*m bytes, each serialised with
a compact size prefix, so at m = 7 the P2TR control block is 257 bytes
and takes a 3-byte prefix while the P2MR one is 225 bytes and takes 1.
The saving there is 34 bytes, not 32. Every other depth up to 128
gives 32.
The section already footnotes this boundary for the P2MR side a few
paragraphs earlier.
- The Test Vectors section linked P2MR_construction.json; the file on
disk is lowercase p2mr_construction.json, so the link 404s.
- The same section still linked the rust implementation under
bip-0360/ref-impl/rust, which #2202 moved out of the repository.
- The header said Version: 0.12.0 while the changelog's latest entry
is 0.12.1.
- The 135-byte depth-1 witness example labeled the merkle path
"(empty)" while counting 32 bytes for it (a depth-1 path has one
32-byte node).
In p2mr_pqc_construction.json, the three-leaf vectors
(p2mr_three_leaf_complex, p2mr_three_leaf_alternative) listed
intermediary.leafHashes with entries 0 and 2 swapped relative to the
depth-first order of the script tree, while the merkle root,
scriptPubKey, address and control blocks all follow depth-first
order (as do the same trees in p2mr_construction.json).
In p2mr_different_version_leaves, scriptPathControlBlocks[1] started
with byte 0xc1 although the leaf it commits to has leafVersion 0xfa;
the control byte carries the leaf version in its upper 7 bits with
the low bit set, so it must be 0xfb. The vector's leafHashes[1] and
merkle root already correspond to the 0xfa leaf.
bip47:
- Change BIP type from Informational to Specification
- Move "recent changes" from above preamble to Changelog section
- Drop Comments headers from Preamble
Clarifies that the commutativity only applies if the original PSBTs do not have conflicting fields.
---------
Co-authored-by: Murch <murch@murch.one>
Co-authored-by: Ava Chow <github@achow101.com>
