processors/token_transfer: accept a Void-encoded to_muxed_id in V4 event data by tamirms · Pull Request #5983 · stellar/go-stellar-sdk

tamirms · 2026-08-19T08:22:17Z

The bug

parseV4MapDataForTokenEvents rejects a to_muxed_id bound to Void:

default:
    return amt, nil, fmt.Errorf("invalid to_muxed_id type for data: %s", entry.Val.Type)

and contractEventsFromOperation discards an event whose parse fails rather than surfacing the error:

// You dont bail on error here, since error here means that it is not a sep-41 compliant token event.
if err == nil {
    events = append(events, ev)
}

So the entire transfer event vanishes from the stream, silently, over a field that carries no information.

Why a contract emits `Void`

A Soroban UDT struct (#[contracttype]) is a map with Symbol keys, and an Option field holding None is encoded as a key bound to Void. A SEP-41 token that publishes its event data as

#[contracttype]
pub struct TransferData { pub amount: i128, pub to_muxed_id: Option<u64> }

therefore emits to_muxed_id -> Void for a transfer with no muxed destination — the natural way to write it, and reachable from any SEP-41 contract, since those are built with soroban-sdk.

CAP-0067 instead specifies the key as absent in that case:

Non-muxed accounts are represented as a singular from/to value in the event topics and the to_muxed_id key won't be present in the data map.

That form already parsed. CAP-0086 (protocol 28) is what closes the gap going forward: it adds sparse_map_new_from_linear_memory / sparse_map_unpack_to_linear_memory, which omit a key holding None, so the same struct built with the v28 SDK produces the compliant encoding. Contracts built before that keep emitting Void, so both forms coexist in ledger history.

The fix

Treat Void as "no muxed destination" — the same result as an absent key. Three lines and a test. There is no XDR change and nothing to regenerate.

Scope

Deliberately limited to processors/token_transfer, which decodes event data from arbitrary SEP-41 contracts.

The Stellar Asset Contract needs no change, on three independent grounds:

It is not built with soroban-sdk. storage_types.rs and public_types.rs in rs-soroban-env use soroban_builtin_sdk_macros::contracttype — a builtin macro crate compiled into the host — so a soroban-sdk release does not reach it.
Its storage types have no Option fields: BalanceValue { amount: i128, authorized: bool, clawback: bool }, AllowanceValue { amount: i128, live_until_ledger: u32 }, AlphaNum{4,12}AssetInfo { asset_code: String, issuer: BytesN<32> }. Sparse encoding omits only Void, so there would be nothing to omit.
Its event data map is built by get_amount_data_maybe_muxed via map_new plus explicit puts, never through the linear-memory host functions CAP-0086 touches.

So ingest/sac is untouched here, as is Horizon's own SAC event parser.

Testing

processors/token_transfer and xdr pass. gofmt clean, go vet clean under the flags govet.sh uses.

🤖 Generated with Claude Code

…ent data A Soroban UDT struct (`#[contracttype]`) encodes an `Option` field holding `None` as a map key bound to `Void`. CAP-0067 instead specifies that `to_muxed_id` is simply absent when a transfer has no muxed destination, and that form already parsed. The `Void` form did not. It fell through to the `default:` arm and failed the parse, and callers discard an event that fails to parse rather than surfacing the error, so the entire transfer event vanished from the stream over a field carrying no information. A contract reaches that path by publishing its event data as a struct with an `Option` field, which is the natural way to write it. CAP-0086 (protocol 28) adds host functions that omit such a key instead, so the same struct built with the v28 SDK now produces the compliant encoding. Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>

Copilot

Pull request overview

Allows historical SEP-41 V4 token events with Void-encoded to_muxed_id values to remain in the event stream.

Changes:

Treats Void as no muxed destination.
Adds regression coverage for the encoding.
Documents the fix in the changelog.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
`processors/token_transfer/contract_events.go`	Accepts `Void` muxed IDs.
`processors/token_transfer/contract_events_test.go`	Tests `Void` parsing behavior.
`CHANGELOG.md`	Records the bug fix.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

leighmcculloch

🎉

…sdk#5983) Fixes silent drop of SEP-41 events with a Void-bound to_muxed_id from token_transfers_raw. Co-Authored-By: Claude Fable 5 <[email protected]>

A CAP that changes no XDR can still change what existing code sees. Both CAP-0086 runs of go-stellar-sdk concluded "nothing to do" while the repo's V4 token-event parser rejected a Void-encoded to_muxed_id and its caller silently dropped the event (fixed by hand in stellar/go-stellar-sdk#5983). _base.md gains a section on the outgoing encoding a CAP replaces, which survives in ledger history and is the side consumers actually break on. go-stellar-sdk moves to opus/high both ways: at sonnet/medium it spent 3-4 turns and missed this, at high effort it found it.

@giortzisg

) * Bump go-stellar-sdk to protocol-next for Protocol 28 Co-Authored-By: Claude Fable 5 <[email protected]> * Bump go-stellar-sdk to protocol-next tip with GA P28 XDR pin Co-Authored-By: Claude Fable 5 <[email protected]> * Bump go-stellar-xdr-json to P28 release branch commit Co-Authored-By: Claude Fable 5 <[email protected]> * Swap go-stellar-sdk pin to tagged release v0.7.2 Co-Authored-By: Claude Fable 5 <[email protected]> * Repin go-stellar-xdr-json to merged main commit Co-Authored-By: Claude Fable 5 <[email protected]> * Bump go-stellar-sdk for TTP void to_muxed_id fix (stellar/go-stellar-sdk#5983) Fixes silent drop of SEP-41 events with a Void-bound to_muxed_id from token_transfers_raw. Co-Authored-By: Claude Fable 5 <[email protected]> * Swap go-stellar-sdk pin to tagged release v0.7.3 Co-Authored-By: Claude Fable 5 <[email protected]> --------- Co-authored-by: Claude Fable 5 <[email protected]>

tamirms mentioned this pull request Aug 19, 2026

ingest/contractevents: tolerate CAP-86 sparse maps in V4 SAC event data stellar/stellar-horizon#223

Closed

tamirms force-pushed the cap-86-sparse-udt-maps branch from 1f4fdd4 to 75f2bcb Compare August 19, 2026 08:51

tamirms changed the title ~~xdr, ingest/sac, processors/token_transfer: tolerate CAP-86 sparse UDT maps~~ xdr, processors/token_transfer: accept Void-encoded optional UDT fields Aug 19, 2026

tamirms force-pushed the cap-86-sparse-udt-maps branch 3 times, most recently from 1f4fdd4 to a427ecf Compare August 19, 2026 09:04

tamirms changed the title ~~xdr, processors/token_transfer: accept Void-encoded optional UDT fields~~ xdr, ingest/sac, processors/token_transfer: tolerate CAP-86 sparse UDT maps Aug 19, 2026

tamirms force-pushed the cap-86-sparse-udt-maps branch from a427ecf to 2b3858a Compare August 19, 2026 09:08

tamirms changed the title ~~xdr, ingest/sac, processors/token_transfer: tolerate CAP-86 sparse UDT maps~~ xdr, processors/token_transfer: accept Void-encoded optional UDT fields Aug 19, 2026

tamirms force-pushed the cap-86-sparse-udt-maps branch from 2b3858a to 03ff491 Compare August 19, 2026 09:13

tamirms force-pushed the cap-86-sparse-udt-maps branch from 03ff491 to 80588db Compare August 19, 2026 09:25

tamirms changed the title ~~xdr, processors/token_transfer: accept Void-encoded optional UDT fields~~ processors/token_transfer: accept a Void-encoded to_muxed_id in V4 event data Aug 19, 2026

tamirms requested a balanced review from Copilot August 19, 2026 09:28

Copilot started reviewing on behalf of tamirms August 19, 2026 09:29 View session

Copilot AI reviewed Aug 19, 2026

View reviewed changes

tamirms marked this pull request as ready for review August 19, 2026 09:37

tamirms requested a review from a team August 19, 2026 09:43

leighmcculloch approved these changes Aug 19, 2026

View reviewed changes

karthikiyer56 approved these changes Aug 19, 2026

View reviewed changes

tamirms merged commit a7921dd into main Aug 19, 2026
12 checks passed

tamirms deleted the cap-86-sparse-udt-maps branch August 19, 2026 15:42

Navigation Menu

Uh oh!

processors/token_transfer: accept a Void-encoded to_muxed_id in V4 event data - #5983

processors/token_transfer: accept a Void-encoded to_muxed_id in V4 event data#5983
tamirms merged 1 commit into
mainfrom
cap-86-sparse-udt-maps

tamirms commented Aug 19, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

leighmcculloch left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Uh oh!

Conversation

tamirms commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

The bug

Why a contract emits Void

The fix

Scope

Testing

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

leighmcculloch left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

tamirms commented Aug 19, 2026 •

edited

Loading

Why a contract emits `Void`