chore(genvm): update runner hashes and rename accepted to decided ⬆️ - #105
chore(genvm): update runner hashes and rename accepted to decided ⬆️#105kp2pml30 wants to merge 2 commits into
Conversation
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
cc0fa46 to
485af1b
Compare
Bump the v0.3 runner pins to the new ids, rename the raw fee-params aliases leader/validator_timeunits_allocation to leader/validator_time_units_allocation, and keep the wait_until rename of accepted to decided.
485af1b to
86c1db1
Compare
The pinned embeddings runner's `VecDB` takes the distance metric as a fourth type parameter. Three arguments now fail at contract import, which leaves the deploy without state rather than reporting a type error. `log_indexer_testnet.py` is deliberately left alone: it targets Asimov, which runs the older SDK.
MuncleUscles
left a comment
MuncleUscles
left a comment
There was a problem hiding this comment.
Reviewed the fee-field/lifecycle migration and explicit VecDB metric update; native CI and exact cross-repo E2E are green. No blocking findings.
2 participants
Description
Two GenVM v0.3.0-rc7 changes:
_decode_pending_transactionsemitson="decided"instead ofon="accepted"— the pre-finalization state was renamed on the wire with no back-compat alias, and the node parses exactly these stringstests/e2e/contracts/simple_time_contract.pyrunner pins bumped to the rc7 hashesUpstream: genvm-manager #24.
Depends-On: genlayerlabs/genvm-manager#24
Motivation and Context
latest_non_final/acceptednamed an implementation queue rather than the state-view contract. The wire values change with no alias, so hosts and SDKs must move together or every post-message/deploy with an explicitonbreaks.How Has This Been Tested?
uv run pytest tests/unit -q— 141 passed. No unit fixture asserts the changed string (allpending_transactionssample payloads are empty lists), so CI is the real check.Decisions Made
onwire value moved.TransactionStatus.ACCEPTED,TransactionAccepted(bytes32)andstates: {accepted, finalized}are consensus-contract concepts and are untouched.TransactionHashVariant.LATEST_FINAL/LATEST_NONFINAL("latest-final"/"latest-nonfinal") is the Studio-onlytransaction_hash_variantRPC selector, not GenVMStorageType— also untouchedsimple_time_contract.pypreviously pinned v0.2-line hashes while being written in v0.3 SDK style. That looked stale rather than intentional, so it is bumped to rc7 and CI will confirmPre-Existing, Not Addressed
RESULT_CODESingenlayer_py/utils/jsonifier.pyalready diverges from v0.3ResultCode0 → accepted,1 → finalized, while the node'sonAcceptancebool istruefor accepted — possible polarity bug, orthogonal to this renameChangeset
v0.6-devv0.6-devv0.123-devv0.19-devv0.30-devmainmainNo PR for genlayer-js or genlayer-consensus — neither needed a change.
Depends-On:is declared only against genvm-manager#24 here. genlayerlabs/genlayer-node#1753 is the hub: it carriesDepends-Onfor every repo in the set, so one E2E run from there resolves the whole closure and fans checks out to each open PR. Sibling rows above are deliberately plain links — a fully-connected mesh would be a dependency cycle, which the E2E dispatcher rejects.