fix(abi): encode zero-width static aggregates as empty bytes

[spokodev]

Summary

An empty fixed array (T[0]) or empty tuple (()) is a zero-width static type. In encodeAbiParameters.ts, the static branches of encodeArray and encodeTuple build their encoding with concat(preparedParams.map(p => p.encoded)) over an empty array.

concat([]) (in utils/data/concat.ts) checks typeof values[0] === 'string'. For an empty input values[0] is undefined, so it routes to concatBytes([]) and returns an empty Uint8Array rather than the hex string '0x'. That Uint8Array then leaks back into encodeParams, poisoning the string-vs-bytes detection and producing either wrong bytes or an uncaught TypeError.

The function's documented return type is Hex, never Uint8Array.

Reproduction

encodeAbiParameters([{ type: 'uint256[0]' }, { type: 'uint256' }], [[], 3n])
// before: Uint8Array(66)            (33 leading zero bytes + the uint256)
// after:  the 32-byte uint256 0x..0003

encodeAbiParameters([{ type: 'uint256' }, { type: 'tuple', components: [] }], [5n, []])
// before: throws TypeError: x.replace is not a function   (concatHex on a Uint8Array)
// after:  the 32-byte uint256 0x..0005

Oracle

Per the ABI spec, enc(T[0]) and enc(()) are zero-length: an empty fixed array or empty tuple contributes no bytes. ethers agrees:

AbiCoder.defaultAbiCoder().encode(['uint256[0]', 'uint256'], [[], 3n])
// '0x0000000000000000000000000000000000000000000000000000000000000003'
AbiCoder.defaultAbiCoder().encode(['uint256', 'tuple()'], [5n, []])
// '0x0000000000000000000000000000000000000000000000000000000000000005'

Fix

Emit '0x' directly when preparedParams is empty in the static branch of encodeArray and encodeTuple, instead of relying on concat([]). The change is local to the two aggregate encoders; concat's behaviour is untouched (it has other callers).

Added tests cover both repros against the ethers/ABI-spec oracle.

This is a separate root cause from #921 (empty dynamic bytes[]) and from the RLP trailing-bytes fix in a sibling PR.

[changeset-bot]

⚠️ No Changeset found

Latest commit: 5ae4a93

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[vercel]

@spokodev is attempting to deploy a commit to the Wevm Team on Vercel.

A member of the Team first needs to authorize it.

[spokodev]

Sibling PR for a separate root cause: #4778 (fromRlp trailing bytes).