feat(hive): serialize limit_order_create / limit_order_cancel#373
Merged
Conversation
Unblocks Hive internal-market swaps. Firmware clear-signs both ops (keepkey-firmware #315); the vault had no serializer, so hive_broadcast failed before the request reached the device. Byte layout mirrors the firmware parser and its unit-test vectors (Hive.LimitOrderCreateRetainsEveryDisplayedField / LimitOrderCancelParses). The firmware's degenerate-order rejections — same symbol on both sides, zero amount, non-HIVE/HBD symbol, non-0/1 fill_or_kill — are mirrored host-side so the user gets a clear error instead of a device reject. Requires firmware >= the release cut from #315. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
BitHighlander
added a commit
that referenced
this pull request
Jul 20, 2026
…sion tests keepkey-sdk had zero Hive coverage — no address getter, no sign methods — despite the vault exposing /addresses/hive and /hive/sign-operations since 7.15.0. Adds address.hiveGetAddress and hive.hiveSignOperations, mirroring the existing osmosis/tron/ton client blocks exactly. Scoped to what the new tests need; sign-transfer/sign-message coverage is a separate follow-on (see handoff-keepkey-sdk-clearsign-coverage.md). Adds two test suites, following the tests/thorchain/ruji.js pattern (live Vault + attached device/emulator, no mocks): - tests/hive/phase3-ops.js — limit_order_create/cancel (fw #315, vault #373) sign successfully, plus the host-side degenerate-order rejections (same-symbol, zero-amount, VESTS-on-market, mixed-tier) fire without a device round-trip. Requires firmware >= rc15, the vault's hive_enabled setting, and live Pioneer for TaPoS header data — the endpoint is explicitly STAGED in rest-api.ts pending this smoke pass. - tests/osmosis/format-amount-regression.js — send/delegate/undelegate still sign with amounts beyond a float's ~7 significant digits, proving the atof()+"%.6f" removal (fw #315) didn't break signing. Does not (and cannot, from the SDK) verify the OLED renders the exact digits — that's Gate-3 OLED proof, still owed. lib/*.js and openapi/swagger.json are the rebuilt output (npm run build); src/ is the source of truth.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Last link in the Hive internal-market swap chain. Firmware clear-signs both ops (keepkey-firmware #315), but the vault had no serializer for them, so a dApp swap died at
hive_broadcastwith:Wire format
Mirrors the firmware parser byte-for-byte:
Host-side rejections
The firmware rejects these; mirrored here so the user sees a real message instead of a device reject: same symbol on both sides, either amount zero, a symbol other than HIVE/HBD (VESTS never trades on the internal market),
fill_or_killthat isn't a boolean. Active tier, so the existing mixed-tier guard already blocks sharing a tx with posting ops.expirationis deliberately unbounded — the device has no RTC, and a stale or far-future value costs an on-chain rejection, not a bad signature.Tests
11 new cases in
hive-ops.test.ts, vectors taken from the firmware unit tests (Hive.LimitOrderCreateRetainsEveryDisplayedField,Hive.LimitOrderRejectsDegenerateOrders,Hive.LimitOrderCancelParses). Expected bytes are hand-built in the test from the condenser layout, so a serializer bug can't cancel out against itself. 41/41 pass.Requires firmware >= the release cut from #315.
🤖 Generated with Claude Code