feat: first compression iteration #639
Conversation
📝 WalkthroughWalkthroughAdds compressed-account (Photon) support and a new workspace crate Suggested reviewers
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
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 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 56
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (12)
magicblock-committor-program/src/instruction_builder/close_buffer.rs (1)
1-46: Revertnew_with_bytespattern tonew_with_borsh; the refactor introduces unnecessary panic points.All four instruction builders (
realloc_buffer.rs,init_buffer.rs,write_buffer.rs,close_buffer.rs) changed fromInstruction::new_with_borsh(program_id, &ix, accounts)toInstruction::new_with_bytes(program_id, &ix.try_to_vec().unwrap(), accounts).Verification confirms
new_with_borshis not deprecated in current solana-program and remains active elsewhere in the codebase. The refactor:
- Removes built-in error handling
- Introduces
.unwrap()panic points- Creates inconsistency with existing code patterns
Revert these four files to use
new_with_borsh.test-integration/programs/flexi-counter/src/instruction.rs (1)
229-242: ConsistentInstruction::new_with_bytes+ Borsh pattern looks good, but consider avoiding bareunwrap()The switch to
Instruction::new_with_bytesplusborsh::to_vec(&FlexiCounterInstruction::...)is consistent and makes payload encoding uniform across all constructors. The only downside is the pervasive.unwrap(), which will panic on serialization failure and make client-side debugging harder.Since these helpers run off-chain in tests, this is acceptable, but if you want a slightly safer pattern you could:
- Use
.expect("borsh serialization of FlexiCounterInstruction should never fail")to give clearer panic messages, or- Return a
Result<Instruction, borsh::mayfail::Error>from the helpers and propagate errors explicitly.Not a blocker—more of a polish consideration.
Also applies to: 244-265, 267-279, 281-303, 304-314, 316-343, 345-367, 369-386, 388-419, 421-463, 465-496, 498-518
test-integration/test-runner/bin/run_tests.rs (1)
244-301: Change cleanup calls to usecleanup_light_validatorfor Light validators.Lines 275, 292, and 301 call
cleanup_devnet_only(&mut devnet_validator), but the validator is started asValidatorCluster::Light(line 246). Thecleanup_devnet_onlyfunction only kills the process and runs generic validator cleanup; it does not execute thelight test-validator --stopcommand thatcleanup_light_validatorprovides.This causes incomplete cleanup of Light validators. Update these three calls to:
cleanup_light_validator(&mut devnet_validator, "light");This matches the pattern used in
cleanup_validators_with_lightand ensures the light validator is properly stopped.magicblock-api/src/genesis_utils.rs (1)
107-113: Normalize AccountSharedData imports for consistency, but no functional issue identifiedBased on code inspection,
solana_account::AccountSharedDataandsolana_sdk::account::AccountSharedDataare compatible types (the code compiles and the push at line 114–115 succeeds without casting). The concern in the review is valid from a consistency and clarity perspective but not a correctness issue.The codebase currently mixes:
- Bare
AccountSharedData::new()(imported fromsolana_sdk::account) at lines 95, 99- Fully qualified
solana_account::AccountSharedData::from()at line 107This inconsistency makes the code harder to follow. The review's suggestion to pick one source of truth and use it consistently throughout the function is sound and improves readability going forward.
magicblock-chainlink/tests/utils/accounts.rs (1)
2-47: Align compressed helper with non‑compressed helper’s slot semanticsThe compressed helper builds a plausible
CompressedDelegationRecordand wraps it in anAccountSharedData, which is good for test fixtures. However, unlikeaccount_shared_with_owner_and_slot, it never callsset_remote_slot(slot)on the resultingAccountSharedData; theslotargument is only stored asdelegation_slotinside the record.If your compressed‑account tests (or
RemoteAccountProvider) rely onAccountSharedData::remote_slotfor slot matching, this discrepancy could lead to subtle differences vs. the non‑compressed helper.Consider aligning the behavior:
pub fn compressed_account_shared_with_owner_and_slot( @@ -) -> AccountSharedData { +) -> AccountSharedData { @@ - let mut acc = Account::new( + let mut acc = Account::new( 0, delegation_record_bytes.len(), &compressed_delegation_client::ID, ); acc.data = delegation_record_bytes; - AccountSharedData::from(acc) + let mut acc_shared = AccountSharedData::from(acc); + acc_shared.set_remote_slot(slot); + acc_shared }Or, if compressed accounts deliberately shouldn’t use
remote_slot, it’d help to document that distinction in the helper’s comment or name.magicblock-committor-service/src/tasks/args_task.rs (1)
1-15: Critical runtime panic:CompressedUndelegateinstruction builder incomplete;reset_commit_idskipsCompressedCommit
CompressedUndelegateinstruction will panic at runtime (args_task.rs:110–125)
UndelegateBuilder::instruction()calls.expect()onpayer,owner_program,system_program, andargs(undelegate.rs:167–184).- Current builder only sets
.delegated_account()and remaining accounts; all required fields are missing.CompressedUndelegateTaskstruct lackspayerandsystem_programfields entirely, so they cannot be passed to the builder.- Either add these fields to
CompressedUndelegateTaskand populate them in task creation, or document why compressed undelegate does not require them.- The commented-out
.owner_program()and.rent_reimbursement()are misleading (latter is not even a builder method); remove or clarify intent.
reset_commit_iddoes not handleCompressedCommit(args_task.rs:196–202)
CompressedCommitTaskhas acommit_idfield that is used in instruction generation (line 75:update_nonce: value.commit_id).- Current implementation only matches
Commitvariant, leavingCompressedCommitunchanged.- Extend the match to also update
CompressedCommitTask::commit_idfor consistency.magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (1)
61-90: PreserveTaskBuilderErrorcontext instead of collapsing toCompressedDataNotFound(lines 161-167)The
PreparationTask::Compressedbranch discards the underlying error details by collapsing to a generic variant:let Ok(compressed_data) = get_compressed_data(&delegated_account, photon_client).await else { error!("Failed to get compressed data"); return Err(InternalError::CompressedDataNotFound); };However,
InternalErroralready hasTaskBuilderError(#[from] TaskBuilderError), which preserves concrete error reasons (CompressedDataFetchError,MissingStateTrees, etc.). Use the propagation pattern instead:let compressed_data = get_compressed_data(&delegated_account, photon_client) .await .map_err(InternalError::TaskBuilderError)?;This retains full error context for debugging while eliminating unnecessary wrapping logic.
magicblock-committor-service/src/tasks/mod.rs (1)
173-187: BufferPreparationTask logging may be overly verbose in hot paths
BufferPreparationTask::init_instructionnow emits multipledebug!lines per call, including sizes and counts. This is useful while validating the newChunks::struct_size(self.chunks.count())usage, but if this path is hit frequently in production it will significantly increase log volume.Once you’re confident in the sizing, consider either removing most of these logs or gating them behind a more specific target/feature flag so normal runs aren’t spammed. The struct_size call itself looks correct given
Chunks::struct_size(count).Also applies to: 189-201
magicblock-chainlink/src/chainlink/fetch_cloner.rs (3)
303-470: Compressed delegation handling in subscription path is mostly correct but lacks fallback loggingThe new
owned_by_compressed_delegation_programbranch correctly:
- Parses
CompressedDelegationRecordfrom the fresh account data.- Falls back to re‑fetching via
remote_account_provider.try_getwhen direct parsing fails.- Marks the account as compressed, sets owner/data/lamports from the record, and sets
delegatedonly whenauthority == validator_pubkey.Two improvements to consider:
- When both direct and Photon fetch parsing fail, we silently return
None. Emitting at least awarn!in that case would make diagnosing bad compressed records easier.- The fallback uses the data from the fetched account only to parse the record but still mutates the original subscription account. That’s probably fine, but if slots can differ you might want to pick the fresher of the two sources explicitly.
1750-1776: Buggy test assertions for compressed accounts (test_multiple_mixed_accounts_some_missing)
test_multiple_mixed_accounts_some_missinghas contradictory expectations:let cacc3 = provider.try_get(cpk3).await.unwrap(); assert_compressed_account!(cacc3, 777, 3); assert!(!cacc3.is_found());
assert_compressed_account!already assertsis_found(), so the subsequentassert!(!cacc3.is_found())will always fail. Additionally, sincecpk3is not part of thecompressed_pubkeyspassed intosetup_with_mixed_accounts, it’s not clear thatcpk3should be found at all unless the mock is updated elsewhere.Fixing the test likely means:
- Keeping
cacc3as not found in the initialtry_get_multicall, and- Either removing the final
assert!(!cacc3.is_found())(since the macro already asserts found) or changing the setup to actually addcpk3to the Photon mock before expecting it to be found.For example, the minimal change is:
- let cacc3 = provider.try_get(cpk3).await.unwrap(); - assert_compressed_account!(cacc3, 777, 3); - assert!(!cacc3.is_found()); + let cacc3 = provider.try_get(cpk3).await.unwrap(); + assert_compressed_account!(cacc3, 777, 3);Also applies to: 1778-1815, 1817-1859
535-607: Compressed accounts have inconsistent lamports between subscription and fetch pathsThe review comment is correct. The subscription path at line 460 calls
account.set_lamports(delegation_record.lamports)for compressed accounts, but the fetch path at lines 746-775 processes compressed accounts without setting lamports.The code locations match exactly:
- Subscription path (lines ~455-461): Sets all four fields including
set_lamports(delegation_record.lamports)- Fetch path (lines 746-775): Sets
set_dataandset_ownerbut omitsset_lamportsThe
CompressedDelegationRecordstruct has apub lamports: u64field available in both paths. This inconsistency means accounts will have different lamport values depending on whether they arrive via subscription updates or fetches, creating subtle state divergence in the bank.The suggested fix is correct—add
account.set_lamports(delegation_record.lamports);in the fetch path to align with the subscription path behavior.magicblock-chainlink/src/remote_account_provider/mod.rs (1)
784-854: Fix type mismatch injoin_all()result handling and remove unsafe unwrapsThe three critical issues are confirmed:
Type mismatch:
JoinSet::join_all().awaitreturnsVec<Result<FetchedRemoteAccounts, JoinError>>, but you're passing it directly toconsolidate_fetched_remote_accounts()which expectsVec<FetchedRemoteAccounts>. This will not compile. You must map/filter join errors before consolidation, or unwrap and handle them properly.Unsafe unwraps in async helpers:
fetch_from_rpc(line 961):.await.unwrap().unwrap()panics on task failure or inner error instead of returningRemoteAccountProviderError::FailedFetchingAccountsfetch_from_photon(line 1009):.await.unwrap()panics on any indexer error (already flagged with TODO comment)Both TODO comments acknowledge this. Replace with error propagation.
Consolidation robustness: If join errors occur before consolidation, they are lost. Once you handle join errors properly, pass only successful results to consolidation and log/early-return if both sources fail.
At minimum: handle join errors before consolidation, replace unwraps with proper error returns, and surface failures through
RemoteAccountProviderError.
♻️ Duplicate comments (1)
magicblock-committor-service/src/intent_executor/mod.rs (1)
811-819: Duplicate hardcoded PhotonIndexer initialization.This is the second instance of hardcoded
http://localhost:8784in the codebase (first was incommittor_processor.rslines 54-57). This duplication reinforces the need to centralize PhotonIndexer configuration.See the comment on
committor_processor.rslines 53-58 for the recommended centralization approach.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (26)
Cargo.lockis excluded by!**/*.lockcompressed-delegation-client/src/generated/accounts/compressed_delegation_record.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/accounts/mod.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/errors/mod.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/commit.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/delegate.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/delegate_compressed.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/finalize.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/mod.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/undelegate.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/instructions/undelegate_compressed.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/mod.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/programs.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/shared.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/commit_args.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/delegate_args.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/delegate_compressed_args.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/external_undelegate_args.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/finalize_args.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/mod.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/undelegate_args.rsis excluded by!**/generated/**compressed-delegation-client/src/generated/types/undelegate_compressed_args.rsis excluded by!**/generated/**magicblock-committor-program/bin/magicblock_committor_program.sois excluded by!**/*.sotest-integration/Cargo.lockis excluded by!**/*.locktest-integration/programs/compressed_delegation/compressed_delegation.sois excluded by!**/*.sotest-integration/schedulecommit/elfs/dlp.sois excluded by!**/*.so
📒 Files selected for processing (106)
.github/workflows/ci-test-integration.yml(1 hunks)Cargo.toml(6 hunks)compressed-delegation-client/Cargo.toml(1 hunks)compressed-delegation-client/src/lib.rs(1 hunks)compressed-delegation-client/src/utils.rs(1 hunks)configs/ephem-compression.toml(1 hunks)magicblock-account-cloner/src/bpf_loader_v1.rs(2 hunks)magicblock-account-cloner/src/lib.rs(1 hunks)magicblock-accounts/src/scheduled_commits_processor.rs(2 hunks)magicblock-aperture/README.md(0 hunks)magicblock-aperture/src/state/mod.rs(2 hunks)magicblock-api/Cargo.toml(1 hunks)magicblock-api/src/genesis_utils.rs(1 hunks)magicblock-api/src/magic_validator.rs(3 hunks)magicblock-api/src/tickers.rs(1 hunks)magicblock-chainlink/Cargo.toml(1 hunks)magicblock-chainlink/src/chainlink/fetch_cloner.rs(19 hunks)magicblock-chainlink/src/chainlink/mod.rs(8 hunks)magicblock-chainlink/src/remote_account_provider/errors.rs(2 hunks)magicblock-chainlink/src/remote_account_provider/mod.rs(21 hunks)magicblock-chainlink/src/remote_account_provider/photon_client.rs(1 hunks)magicblock-chainlink/src/remote_account_provider/remote_account.rs(4 hunks)magicblock-chainlink/src/testing/mod.rs(2 hunks)magicblock-chainlink/src/testing/photon_client_mock.rs(1 hunks)magicblock-chainlink/src/testing/utils.rs(1 hunks)magicblock-chainlink/tests/01_ensure-accounts.rs(3 hunks)magicblock-chainlink/tests/03_deleg_after_sub.rs(2 hunks)magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs(2 hunks)magicblock-chainlink/tests/05_redeleg_other_same_slot.rs(3 hunks)magicblock-chainlink/tests/06_redeleg_us_separate_slots.rs(3 hunks)magicblock-chainlink/tests/07_redeleg_us_same_slot.rs(3 hunks)magicblock-chainlink/tests/utils/accounts.rs(2 hunks)magicblock-chainlink/tests/utils/test_context.rs(4 hunks)magicblock-committor-program/src/instruction_builder/close_buffer.rs(2 hunks)magicblock-committor-program/src/instruction_builder/init_buffer.rs(2 hunks)magicblock-committor-program/src/instruction_builder/realloc_buffer.rs(2 hunks)magicblock-committor-program/src/instruction_builder/write_buffer.rs(2 hunks)magicblock-committor-program/src/processor.rs(3 hunks)magicblock-committor-program/src/state/changeset_chunks.rs(1 hunks)magicblock-committor-program/src/state/chunks.rs(2 hunks)magicblock-committor-service/Cargo.toml(1 hunks)magicblock-committor-service/src/committor_processor.rs(3 hunks)magicblock-committor-service/src/intent_execution_manager.rs(2 hunks)magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs(1 hunks)magicblock-committor-service/src/intent_executor/mod.rs(19 hunks)magicblock-committor-service/src/intent_executor/single_stage_executor.rs(4 hunks)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs(7 hunks)magicblock-committor-service/src/intent_executor/two_stage_executor.rs(4 hunks)magicblock-committor-service/src/tasks/args_task.rs(7 hunks)magicblock-committor-service/src/tasks/buffer_task.rs(3 hunks)magicblock-committor-service/src/tasks/mod.rs(7 hunks)magicblock-committor-service/src/tasks/task_builder.rs(7 hunks)magicblock-committor-service/src/tasks/task_strategist.rs(1 hunks)magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs(8 hunks)magicblock-committor-service/src/transaction_preparator/mod.rs(4 hunks)magicblock-core/Cargo.toml(2 hunks)magicblock-core/src/compression/mod.rs(1 hunks)magicblock-core/src/lib.rs(1 hunks)magicblock-magic-program-api/src/instruction.rs(4 hunks)magicblock-rpc-client/src/lib.rs(1 hunks)programs/magicblock/Cargo.toml(1 hunks)programs/magicblock/src/magic_context.rs(2 hunks)programs/magicblock/src/magic_scheduled_base_intent.rs(8 hunks)programs/magicblock/src/magicblock_processor.rs(2 hunks)programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs(3 hunks)programs/magicblock/src/schedule_transactions/process_schedule_base_intent.rs(1 hunks)programs/magicblock/src/schedule_transactions/process_schedule_commit.rs(2 hunks)programs/magicblock/src/utils/instruction_utils.rs(3 hunks)test-integration/Cargo.toml(3 hunks)test-integration/configs/chainlink-conf.devnet.toml(1 hunks)test-integration/configs/committor-conf.devnet.toml(1 hunks)test-integration/notes-babur.md(1 hunks)test-integration/programs/flexi-counter/Cargo.toml(1 hunks)test-integration/programs/flexi-counter/src/instruction.rs(17 hunks)test-integration/programs/flexi-counter/src/lib.rs(2 hunks)test-integration/programs/flexi-counter/src/processor.rs(5 hunks)test-integration/programs/flexi-counter/src/state.rs(1 hunks)test-integration/programs/schedulecommit/src/api.rs(4 hunks)test-integration/schedulecommit/test-security/Cargo.toml(1 hunks)test-integration/schedulecommit/test-security/tests/utils/mod.rs(3 hunks)test-integration/test-chainlink/Cargo.toml(2 hunks)test-integration/test-chainlink/scripts/miniv2-json-from-so.js(1 hunks)test-integration/test-chainlink/src/ixtest_context.rs(10 hunks)test-integration/test-chainlink/src/programs.rs(1 hunks)test-integration/test-chainlink/src/test_context.rs(4 hunks)test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs(1 hunks)test-integration/test-chainlink/tests/ix_03_deleg_after_sub.rs(1 hunks)test-integration/test-chainlink/tests/ix_06_redeleg_us_separate_slots.rs(1 hunks)test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs(1 hunks)test-integration/test-chainlink/tests/ix_remote_account_provider.rs(2 hunks)test-integration/test-committor-service/Cargo.toml(2 hunks)test-integration/test-committor-service/tests/common.rs(6 hunks)test-integration/test-committor-service/tests/test_delivery_preparator.rs(6 hunks)test-integration/test-committor-service/tests/test_intent_executor.rs(12 hunks)test-integration/test-committor-service/tests/test_ix_commit_local.rs(23 hunks)test-integration/test-committor-service/tests/test_transaction_preparator.rs(5 hunks)test-integration/test-committor-service/tests/utils/instructions.rs(3 hunks)test-integration/test-committor-service/tests/utils/mod.rs(2 hunks)test-integration/test-committor-service/tests/utils/transactions.rs(6 hunks)test-integration/test-runner/bin/run_tests.rs(15 hunks)test-integration/test-runner/src/cleanup.rs(1 hunks)test-integration/test-runner/src/signal.rs(2 hunks)test-integration/test-schedule-intent/tests/test_schedule_intents.rs(7 hunks)test-integration/test-tools/Cargo.toml(1 hunks)test-integration/test-tools/src/validator.rs(3 hunks)test-kit/src/lib.rs(1 hunks)
💤 Files with no reviewable changes (1)
- magicblock-aperture/README.md
🧰 Additional context used
🧠 Learnings (11)
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-chainlink/src/remote_account_provider/errors.rsmagicblock-aperture/src/state/mod.rsmagicblock-accounts/src/scheduled_commits_processor.rsmagicblock-chainlink/tests/03_deleg_after_sub.rsmagicblock-chainlink/tests/06_redeleg_us_separate_slots.rstest-integration/test-chainlink/tests/ix_03_deleg_after_sub.rsmagicblock-chainlink/src/remote_account_provider/photon_client.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-chainlink/src/testing/photon_client_mock.rsmagicblock-chainlink/tests/utils/test_context.rstest-integration/test-chainlink/tests/ix_remote_account_provider.rstest-integration/test-chainlink/src/test_context.rsmagicblock-chainlink/src/chainlink/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rsmagicblock-chainlink/src/remote_account_provider/remote_account.rstest-integration/test-chainlink/src/ixtest_context.rsmagicblock-api/src/magic_validator.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
test-integration/test-chainlink/tests/ix_01_ensure-accounts.rstest-integration/programs/flexi-counter/src/lib.rstest-integration/test-chainlink/tests/ix_06_redeleg_us_separate_slots.rstest-integration/programs/flexi-counter/src/state.rstest-integration/test-runner/bin/run_tests.rstest-integration/test-chainlink/tests/ix_03_deleg_after_sub.rsprograms/magicblock/src/schedule_transactions/process_schedule_base_intent.rstest-integration/programs/flexi-counter/src/instruction.rstest-integration/test-committor-service/tests/utils/mod.rstest-integration/test-committor-service/tests/utils/transactions.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rstest-integration/programs/flexi-counter/src/processor.rstest-integration/schedulecommit/test-security/tests/utils/mod.rstest-integration/test-chainlink/src/ixtest_context.rstest-integration/test-schedule-intent/tests/test_schedule_intents.rsmagicblock-chainlink/src/testing/mod.rstest-integration/test-chainlink/src/programs.rstest-integration/test-committor-service/tests/test_intent_executor.rstest-integration/programs/flexi-counter/Cargo.toml
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
test-integration/notes-babur.mdmagicblock-magic-program-api/src/instruction.rsmagicblock-api/src/tickers.rstest-integration/test-chainlink/tests/ix_06_redeleg_us_separate_slots.rsmagicblock-chainlink/tests/utils/accounts.rsmagicblock-chainlink/tests/03_deleg_after_sub.rsmagicblock-chainlink/tests/06_redeleg_us_separate_slots.rsprograms/magicblock/Cargo.tomlprograms/magicblock/src/mutate_accounts/process_mutate_accounts.rsmagicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_03_deleg_after_sub.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsprograms/magicblock/src/magic_context.rsmagicblock-committor-service/src/tasks/task_builder.rsmagicblock-committor-program/src/processor.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rstest-integration/test-committor-service/tests/utils/transactions.rstest-integration/test-committor-service/tests/utils/instructions.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsprograms/magicblock/src/utils/instruction_utils.rsprograms/magicblock/src/schedule_transactions/process_schedule_commit.rsmagicblock-committor-service/src/intent_executor/task_info_fetcher.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rsmagicblock-chainlink/src/remote_account_provider/remote_account.rstest-integration/test-chainlink/src/ixtest_context.rsmagicblock-account-cloner/src/lib.rsmagicblock-api/src/magic_validator.rsprograms/magicblock/src/magic_scheduled_base_intent.rstest-integration/test-committor-service/tests/test_ix_commit_local.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
test-integration/notes-babur.mdmagicblock-api/src/tickers.rsprograms/magicblock/src/magic_context.rsmagicblock-committor-program/src/processor.rsprograms/magicblock/src/magicblock_processor.rsmagicblock-committor-service/src/intent_executor/task_info_fetcher.rs
📚 Learning: 2025-10-21T13:06:38.900Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/http/get_program_accounts.rs:17-25
Timestamp: 2025-10-21T13:06:38.900Z
Learning: The magicblock validator does not support ledger forking, so commitment-based state queries (processed/confirmed/finalized) are not applicable. RPC methods can safely ignore commitment and minContextSlot parameters from Solana RPC config objects.
Applied to files:
magicblock-api/src/tickers.rs
📚 Learning: 2025-11-07T18:19:48.996Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 614
File: test-integration/programs/schedulecommit/Cargo.toml:19-20
Timestamp: 2025-11-07T18:19:48.996Z
Learning: In Solana programs (crates depending on solana-program), the empty feature flags `custom-heap = []` and `custom-panic = []` are standard declarations required to avoid compiler warnings. These should not be flagged as unused features in Cargo.toml files for Solana program crates.
Applied to files:
programs/magicblock/Cargo.tomlmagicblock-core/Cargo.tomltest-integration/test-chainlink/Cargo.tomltest-integration/programs/flexi-counter/Cargo.toml
📚 Learning: 2025-10-21T11:00:18.396Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/encoder.rs:176-187
Timestamp: 2025-10-21T11:00:18.396Z
Learning: In the magicblock validator, the current slot is always the root slot. The SlotEncoder in magicblock-aperture/src/encoder.rs correctly sets `root: slot` because there is no lag between current and root slots in this architecture.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rs
📚 Learning: 2025-11-04T10:53:50.922Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/locks.rs:110-122
Timestamp: 2025-11-04T10:53:50.922Z
Learning: In magicblock-processor, the TransactionScheduler runs in a single, dedicated thread and will always remain single-threaded. The `next_transaction_id()` function in scheduler/locks.rs uses `unsafe static mut` which is safe given this architectural guarantee.
Applied to files:
programs/magicblock/src/magicblock_processor.rsmagicblock-committor-service/src/committor_processor.rs
📚 Learning: 2025-11-07T13:09:52.253Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: test-kit/src/lib.rs:275-0
Timestamp: 2025-11-07T13:09:52.253Z
Learning: In test-kit, the transaction scheduler in ExecutionTestEnv is not expected to shut down during tests. Therefore, using `.unwrap()` in test helper methods like `schedule_transaction` is acceptable and will not cause issues in the test environment.
Applied to files:
programs/magicblock/src/magicblock_processor.rstest-integration/test-schedule-intent/tests/test_schedule_intents.rstest-integration/test-committor-service/tests/test_intent_executor.rs
📚 Learning: 2025-10-26T16:53:29.820Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 587
File: magicblock-chainlink/src/remote_account_provider/mod.rs:134-0
Timestamp: 2025-10-26T16:53:29.820Z
Learning: In magicblock-chainlink/src/remote_account_provider/mod.rs, the `Endpoint::separate_pubsub_url_and_api_key()` method uses `split_once("?api-key=")` because the api-key parameter is always the only query parameter right after `?`. No additional query parameter parsing is needed for this use case.
Applied to files:
test-integration/test-chainlink/tests/ix_remote_account_provider.rsmagicblock-api/src/magic_validator.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-10-26T08:49:31.543Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 585
File: magicblock-committor-service/src/tasks/buffer_task.rs:111-115
Timestamp: 2025-10-26T08:49:31.543Z
Learning: In the magicblock-committor-service, compute units returned by the `compute_units()` method in task implementations (such as `BufferTask`, `ArgsTask`, etc.) represent the compute budget for a single task. Transactions can comprise multiple tasks, and the total compute budget for a transaction is computed as the sum of the compute units of all tasks included in that transaction.
Applied to files:
magicblock-committor-service/src/tasks/args_task.rsmagicblock-committor-service/src/intent_executor/mod.rs
🧬 Code graph analysis (44)
test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs (3)
magicblock-chainlink/src/testing/utils.rs (1)
init_logger(51-59)test-integration/test-chainlink/src/ixtest_context.rs (2)
init(101-106)counter_pda(206-208)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
remote_slot(146-152)
magicblock-committor-service/src/intent_execution_manager.rs (1)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (3)
Self(298-306)Self(308-316)new(63-75)
magicblock-committor-service/src/tasks/buffer_task.rs (3)
magicblock-committor-service/src/tasks/mod.rs (4)
is_compressed(99-99)set_compressed_data(102-102)get_compressed_data(105-105)delegated_account(108-108)magicblock-committor-service/src/tasks/args_task.rs (4)
is_compressed(243-250)set_compressed_data(252-265)get_compressed_data(267-280)delegated_account(282-298)magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)
magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs (2)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (5)
pubkeys(498-501)pubkeys(509-512)pubkeys(513-516)pubkeys(1031-1034)pubkeys(1036-1039)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (2)
pubkeys(159-170)pubkeys(216-219)
test-integration/test-chainlink/tests/ix_06_redeleg_us_separate_slots.rs (4)
magicblock-chainlink/src/testing/utils.rs (2)
init_logger(51-59)sleep_ms(41-43)test-integration/test-chainlink/src/ixtest_context.rs (2)
init(101-106)counter_pda(206-208)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
remote_slot(146-152)compressed-delegation-client/src/lib.rs (1)
id(61-63)
magicblock-chainlink/tests/03_deleg_after_sub.rs (2)
magicblock-chainlink/src/testing/rpc_client_mock.rs (2)
accounts(95-101)slot(66-72)magicblock-chainlink/tests/utils/accounts.rs (2)
compressed_account_shared_with_owner_and_slot(22-47)account_shared_with_owner_and_slot(12-20)
magicblock-chainlink/tests/06_redeleg_us_separate_slots.rs (3)
magicblock-chainlink/src/testing/rpc_client_mock.rs (2)
accounts(95-101)slot(66-72)magicblock-chainlink/src/testing/accounts.rs (1)
account_shared_with_owner(4-10)magicblock-chainlink/tests/utils/accounts.rs (2)
compressed_account_shared_with_owner_and_slot(22-47)account_shared_with_owner_and_slot(12-20)
magicblock-core/src/compression/mod.rs (1)
test-integration/test-chainlink/src/ixtest_context.rs (1)
hashv_to_bn254_field_size_be_const_array(562-562)
test-integration/test-runner/bin/run_tests.rs (3)
test-integration/test-tools/src/validator.rs (4)
start_light_validator_with_config(130-221)start_magic_block_validator_with_config(23-75)start_test_validator_with_config(77-128)cleanup(359-363)test-integration/test-runner/src/cleanup.rs (3)
cleanup_devnet_only(21-24)cleanup_light_validator(26-38)cleanup_validators(3-10)test-integration/test-runner/src/signal.rs (1)
wait_for_ctrlc(9-33)
programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs (1)
magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
compressed(154-156)
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs (4)
magicblock-chainlink/src/testing/rpc_client_mock.rs (4)
accounts(95-101)slot(66-72)default(40-42)default(246-248)magicblock-chainlink/tests/utils/accounts.rs (3)
compressed_account_shared_with_owner_and_slot(22-47)default(57-63)account_shared_with_owner_and_slot(12-20)magicblock-chainlink/src/testing/photon_client_mock.rs (1)
get_account(49-68)magicblock-chainlink/src/testing/cloner_stub.rs (1)
get_account(41-45)
test-integration/test-chainlink/tests/ix_03_deleg_after_sub.rs (3)
magicblock-chainlink/src/testing/utils.rs (1)
init_logger(51-59)test-integration/test-chainlink/src/ixtest_context.rs (2)
init(101-106)counter_pda(206-208)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
remote_slot(146-152)
magicblock-chainlink/src/remote_account_provider/photon_client.rs (3)
magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)magicblock-chainlink/src/testing/photon_client_mock.rs (3)
new(26-30)get_account(49-68)get_multiple_accounts(70-87)magicblock-chainlink/src/remote_account_provider/mod.rs (10)
new(212-256)None(297-297)None(1214-1214)None(1259-1259)None(1337-1337)None(1537-1537)pubkeys(510-514)pubkeys(970-993)pubkeys(1175-1178)default(124-130)
magicblock-chainlink/tests/07_redeleg_us_same_slot.rs (4)
magicblock-chainlink/src/testing/rpc_client_mock.rs (4)
accounts(95-101)slot(66-72)default(40-42)default(246-248)magicblock-chainlink/src/testing/accounts.rs (1)
account_shared_with_owner(4-10)magicblock-chainlink/src/testing/deleg.rs (1)
add_delegation_record_for(22-46)magicblock-chainlink/tests/utils/accounts.rs (3)
compressed_account_shared_with_owner_and_slot(22-47)default(57-63)account_shared_with_owner_and_slot(12-20)
test-integration/test-committor-service/tests/test_delivery_preparator.rs (2)
test-integration/test-committor-service/tests/common.rs (3)
create_compressed_commit_task(185-212)generate_random_bytes(158-163)new(63-66)test-integration/test-committor-service/tests/utils/transactions.rs (1)
init_and_delegate_compressed_account_on_chain(233-322)
test-integration/test-runner/src/signal.rs (1)
test-integration/test-runner/src/cleanup.rs (2)
cleanup_light_validator(26-38)cleanup_validator(40-44)
magicblock-chainlink/src/testing/photon_client_mock.rs (2)
magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)magicblock-chainlink/src/remote_account_provider/photon_client.rs (5)
new(27-29)get_account(38-42)get_account(53-81)get_multiple_accounts(44-48)get_multiple_accounts(83-123)
magicblock-committor-service/src/tasks/task_builder.rs (4)
magicblock-committor-service/src/tasks/mod.rs (5)
instruction(68-68)instruction(304-310)get_compressed_data(105-105)task_type(93-93)is_compressed(99-99)magicblock-committor-service/src/tasks/args_task.rs (5)
instruction(51-152)new(42-47)get_compressed_data(267-280)task_type(216-228)is_compressed(243-250)magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)magicblock-committor-service/src/intent_execution_manager.rs (1)
new(36-71)
test-integration/programs/flexi-counter/src/instruction.rs (3)
compressed-delegation-client/src/generated/instructions/delegate_compressed.rs (8)
args(162-165)args(430-433)payer(123-126)payer(386-392)new(81-83)new(119-121)new(246-262)new(372-384)compressed-delegation-client/src/lib.rs (1)
id(61-63)test-integration/programs/flexi-counter/src/state.rs (2)
pda(43-46)new(25-31)
magicblock-chainlink/tests/utils/test_context.rs (2)
magicblock-chainlink/src/remote_account_provider/mod.rs (3)
rpc_client(1116-1118)rpc_client(1129-1131)new(212-256)magicblock-chainlink/src/testing/photon_client_mock.rs (1)
new(26-30)
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (5)
magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)magicblock-committor-service/src/tasks/mod.rs (4)
get_compressed_data(105-105)strategy(90-90)cleanup_task(289-294)delegated_account(108-108)magicblock-committor-service/src/tasks/args_task.rs (4)
get_compressed_data(267-280)strategy(212-214)delegated_account(282-298)from(36-38)magicblock-committor-service/src/tasks/buffer_task.rs (3)
get_compressed_data(157-161)strategy(121-123)delegated_account(163-169)magicblock-committor-service/src/utils.rs (1)
persist_status_update(6-22)
magicblock-chainlink/tests/05_redeleg_other_same_slot.rs (3)
magicblock-chainlink/src/testing/rpc_client_mock.rs (2)
accounts(95-101)slot(66-72)magicblock-chainlink/src/testing/accounts.rs (1)
account_shared_with_owner(4-10)magicblock-chainlink/tests/utils/accounts.rs (1)
compressed_account_shared_with_owner_and_slot(22-47)
magicblock-chainlink/tests/01_ensure-accounts.rs (1)
magicblock-chainlink/tests/utils/accounts.rs (1)
compressed_account_shared_with_owner_and_slot(22-47)
programs/magicblock/src/magicblock_processor.rs (1)
programs/magicblock/src/schedule_transactions/process_schedule_commit.rs (2)
process_schedule_commit(33-39)process_schedule_compressed_commit(41-47)
test-integration/test-committor-service/tests/utils/transactions.rs (2)
test-integration/test-committor-service/tests/utils/instructions.rs (2)
init_account_and_delegate_compressed_ixs(79-159)init_account_and_delegate_ixs(48-77)test-integration/test-committor-service/tests/utils/mod.rs (1)
sleep_millis(16-18)
test-integration/test-committor-service/tests/utils/instructions.rs (3)
magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)test-integration/programs/flexi-counter/src/instruction.rs (1)
create_delegate_compressed_ix(520-540)test-integration/programs/flexi-counter/src/state.rs (1)
pda(43-46)
test-integration/test-committor-service/tests/common.rs (3)
magicblock-committor-service/src/tasks/args_task.rs (2)
instruction(51-152)new(42-47)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (1)
new(63-75)magicblock-committor-service/src/intent_executor/mod.rs (1)
new(92-104)
magicblock-committor-service/src/committor_processor.rs (3)
magicblock-committor-service/src/intent_execution_manager.rs (1)
new(36-71)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (1)
new(63-75)magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(27-29)
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs (4)
magicblock-chainlink/src/testing/utils.rs (1)
init_logger(51-59)test-integration/test-chainlink/src/ixtest_context.rs (2)
init(101-106)counter_pda(206-208)test-integration/test-chainlink/src/test_context.rs (1)
init(61-129)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
remote_slot(146-152)
magicblock-committor-service/src/tasks/args_task.rs (3)
magicblock-committor-service/src/tasks/buffer_task.rs (7)
new(38-46)preparation_state(98-100)is_compressed(146-148)task_type(125-129)set_compressed_data(150-155)get_compressed_data(157-161)delegated_account(163-169)magicblock-committor-service/src/tasks/mod.rs (6)
preparation_state(77-77)is_compressed(99-99)task_type(93-93)set_compressed_data(102-102)get_compressed_data(105-105)delegated_account(108-108)magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)
test-integration/test-chainlink/src/test_context.rs (1)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (1)
new(139-163)
programs/magicblock/src/schedule_transactions/process_schedule_commit.rs (2)
programs/magicblock/src/utils/instruction_utils.rs (1)
schedule_commit(33-40)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
compressed(154-156)
test-integration/programs/flexi-counter/src/processor.rs (3)
compressed-delegation-client/src/generated/instructions/delegate_compressed.rs (12)
args(162-165)args(430-433)new(81-83)new(119-121)new(246-262)new(372-384)payer(123-126)payer(386-392)invoke(264-266)invoke(464-466)delegated_account(128-134)delegated_account(394-400)test-integration/programs/flexi-counter/src/state.rs (4)
seeds(32-34)seeds_with_bump(36-41)new(25-31)pda(43-46)test-integration/programs/flexi-counter/src/utils/asserts.rs (1)
assert_keys_equal(7-19)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (5)
magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)magicblock-chainlink/src/testing/photon_client_mock.rs (1)
new(26-30)magicblock-committor-service/src/intent_execution_manager.rs (1)
new(36-71)magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(27-29)compressed-delegation-client/src/generated/accounts/compressed_delegation_record.rs (1)
fetch_compressed_delegation_record(59-68)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (4)
magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(27-29)magicblock-chainlink/src/remote_account_provider/mod.rs (6)
new(212-256)None(297-297)None(1214-1214)None(1259-1259)None(1337-1337)None(1537-1537)magicblock-chainlink/src/remote_account_provider/remote_account.rs (3)
account(189-211)account_shared_data(122-128)slot(212-224)compressed-delegation-client/src/lib.rs (1)
id(61-63)
magicblock-chainlink/src/remote_account_provider/remote_account.rs (2)
compressed-delegation-client/src/generated/accounts/compressed_delegation_record.rs (1)
owner(158-160)compressed-delegation-client/src/lib.rs (1)
id(61-63)
test-integration/test-chainlink/src/ixtest_context.rs (4)
magicblock-core/src/compression/mod.rs (2)
hashv_to_bn254_field_size_be_const_array(14-14)derive_cda_from_pda(10-22)magicblock-committor-service/src/tasks/args_task.rs (2)
instruction(51-152)new(42-47)test-integration/programs/flexi-counter/src/state.rs (3)
new(25-31)seeds(32-34)pda(43-46)test-integration/programs/flexi-counter/src/instruction.rs (1)
create_delegate_compressed_ix(520-540)
test-integration/test-tools/src/validator.rs (2)
test-integration/test-runner/bin/run_tests.rs (1)
log_suffix(852-858)test-integration/test-tools/src/toml_to_args.rs (2)
rpc_port_from_config(122-125)config_to_args(59-120)
magicblock-committor-service/src/tasks/mod.rs (4)
magicblock-committor-service/src/tasks/args_task.rs (4)
is_compressed(243-250)set_compressed_data(252-265)get_compressed_data(267-280)delegated_account(282-298)magicblock-committor-service/src/tasks/buffer_task.rs (4)
is_compressed(146-148)set_compressed_data(150-155)get_compressed_data(157-161)delegated_account(163-169)magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)magicblock-committor-program/src/state/chunks.rs (7)
struct_size(60-67)chunks(219-219)chunks(234-234)count(126-128)std(64-64)std(66-66)chunk_size(130-132)
test-integration/test-committor-service/tests/test_intent_executor.rs (2)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (1)
new(63-75)magicblock-committor-service/src/tasks/task_builder.rs (2)
finalize_tasks(56-60)finalize_tasks(167-377)
programs/magicblock/src/magic_scheduled_base_intent.rs (1)
magicblock-committor-service/src/tasks/mod.rs (1)
is_compressed(99-99)
test-integration/test-committor-service/tests/test_ix_commit_local.rs (3)
magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)test-integration/test-committor-service/tests/utils/transactions.rs (1)
init_and_delegate_compressed_account_on_chain(233-322)compressed-delegation-client/src/lib.rs (1)
id(61-63)
magicblock-chainlink/src/remote_account_provider/mod.rs (3)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (3)
new(139-163)None(283-283)None(1603-1603)magicblock-chainlink/src/remote_account_provider/photon_client.rs (2)
new(27-29)new_from_url(30-33)magicblock-chainlink/src/remote_account_provider/remote_account.rs (6)
slot(212-224)data(72-78)from_fresh_account(172-187)source(225-232)fresh_lamports(248-250)fresh_data_len(252-254)
magicblock-committor-service/src/intent_executor/mod.rs (3)
magicblock-committor-service/src/tasks/task_builder.rs (7)
commit_tasks(48-53)commit_tasks(70-164)finalize_task(173-190)accounts(105-108)accounts(312-315)finalize_tasks(56-60)finalize_tasks(167-377)magicblock-committor-service/src/intent_execution_manager.rs (1)
new(36-71)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (5)
new(63-75)pubkeys(159-170)pubkeys(216-219)fetch_rent_reimbursements(39-42)fetch_rent_reimbursements(338-353)
🪛 LanguageTool
test-integration/notes-babur.md
[style] ~23-~23: In contexts where ‘if’ is followed by ‘or’, using ‘whether’ may be more appropriate (and formal).
Context: ....blocks.get(hash)`? - [x] we won't know if an account delegated to system program ...
(IF_WHETHER)
[grammar] ~25-~25: Ensure spelling is correct
Context: ... of those would fail (Gabriele) and the committor won't retry (Edwin) - [ ] LRU cache cap...
(QB_NEW_EN_ORTHOGRAPHY_ERROR_IDS_1)
[grammar] ~36-~36: Ensure spelling is correct
Context: ...undelegate -> turn on subscription (via committor service) 4. Get update for 2. -> turn o...
(QB_NEW_EN_ORTHOGRAPHY_ERROR_IDS_1)
[style] ~48-~48: This sentence contains multiple usages of the word “again”. Consider removing or replacing it.
Context: ...ates 5. We can now write to the account again and won't receive the undelegation upda...
(REPETITION_OF_AGAIN)
[style] ~83-~83: Who is ‘not sure’? Consider being more precise.
Context: ...pand --locked) ### Need Babur's Help Not sure why these fail (assume 0` return value...
(WHO_NOT_SURE)
[grammar] ~95-~95: Ensure spelling is correct
Context: ...add_ an Account to the test env which looses the delegated flag. Either we add th...
(QB_NEW_EN_ORTHOGRAPHY_ERROR_IDS_1)
🪛 markdownlint-cli2 (0.18.1)
test-integration/notes-babur.md
1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
22-22: Unordered list indentation
Expected: 2; Actual: 4
(MD007, ul-indent)
22-22: Unordered list indentation
Expected: 4; Actual: 6
(MD007, ul-indent)
128-128: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
test-integration/test-schedule-intent/tests/test_schedule_intents.rs
Outdated
Show resolved
Hide resolved
test-integration/test-committor-service/tests/test_transaction_preparator.rs
Show resolved
Hide resolved
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Improvements** * WebSocket communication now processes binary payloads in addition to text payloads, expanding the range of supported data types for transmission. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
Old `DeliveryPreparator::send_ixs_with_retry` wasn't parsing any errors and retrying blindly for specified `max_retries` times. This led to Intents taking more time retrying errors it couldn't recover from. Now we adapt error parsing from IntentExecutor. and not to duplicate quite cumbersome error dispatches of `MagicBlockRpcClientError` a generalized version that could be used from both `IntentExecutor` and `DeliveryPreparator` is introduced. See error explained in #590. We handle it by cleaning and then reainitializing buffer. Due structure of `magicblock-committer-program` reusing existing buffer is only possible if initialized buffer length is equal to current account state, which isn't guaranteed at all but could be a future optimization <!-- This is an auto-generated comment: release notes by coderabbit.ai --> * **New Features** * Added a public utils module with a pluggable async retry framework (send-with-retries), error mappers, and RPC error decision logic. * Delivery preparator and intent executor now surface richer send outcomes and mappers, exposing new public error types and APIs that can carry optional signatures. * **Bug Fixes** * Improved error propagation and messages to include RPC context and optional signatures across execution and retry flows. * Centralized retry/error-mapping for more consistent prepare/delivery behavior. * **Tests** * Added/updated integration tests for cleanup, re-prepare, already-initialized, and adjusted error assertions. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> * **Chores** * Optimized continuous integration pipeline with restructured test execution workflow. * Enhanced build system efficiency for integration testing processes. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Co-authored-by: Luca Cillario <luca@magicblock.gg> Co-authored-by: Thorsten Lorenz <thlorenz@gmx.de>
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Replaced workspace-wide dependencies with explicit version constraints for underlying libraries to improve build consistency. * Preserved existing serde derive feature while pinning its version. * No changes to public APIs or exported behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Refactor** * Enhanced internal debugging capabilities across encoder types and filter structures. * Improved subscription lifecycle management through centralized registration handling, replacing direct insertion patterns. * **Chores** * Updated type exports and visibility modifiers for subscription-related infrastructure. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added per-intent execution timing histogram and async CU usage reporting. * Failure counters now include intent and error labels for richer metrics. * **Chores** * Introduced labeled-value support for intent/outcome results so metrics carry context. * Metrics collection refactored to a dedicated flow invoked after execution. * **Bug Fixes** * Improved execution failure log formatting and adjusted logging level based on duration. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
…egated to another validator (#619)
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Enhanced RPC transaction configuration settings for improved transaction retrieval and metrics calculation across transaction processing paths. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Performance** * Compaction now preserves account-modification data during cleanup to avoid losing important account-related entries. * Faster metadata access via column-count caching, reducing overhead during storage operations. * **Monitoring** * Continuous metrics for ledger storage size, accounts storage size, and account counts. * New duration histogram for column-count measurements and renamed/updated execution-time histogram for improved observability. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Refactor** * Optimized ledger truncation process to execute asynchronously, reducing blocking operations and improving system responsiveness during maintenance operations. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
There was a problem hiding this comment.
Actionable comments posted: 30
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (12)
magicblock-api/src/tickers.rs (1)
105-105: Remove the stale#[allow(unused_variables)]attribute.All function parameters (
tick_duration,ledger,accountsdb,token) are now used in the implementation, so this attribute is no longer necessary.Apply this diff:
-#[allow(unused_variables)] pub fn init_system_metrics_ticker(magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (1)
507-517: Potential double-logging of cleanup errors.Line 510 maps errors to
InternalError, then line 511 inspects and logs them. If the caller also logs the returned error, this could result in duplicate error logs. Consider whether theinspectlogging is necessary here, or document that callers should not log these errors again.magicblock-committor-service/src/tasks/args_task.rs (1)
235-241: Apply the suggested diff to handle CompressedCommit in reset_commit_id.Both
CommitTaskandCompressedCommitTaskhave identicalcommit_id: u64fields (mod.rs:117-129). The current implementation only resetsCommitvariants, causingCompressedCommitreset calls to silently fail when invoked generically at the call site (intent_executor/mod.rs:393). This leaves invalid compressed commit tasks with stale IDs during error recovery.fn reset_commit_id(&mut self, commit_id: u64) { - let ArgsTaskType::Commit(commit_task) = &mut self.task_type else { - return; - }; - - commit_task.commit_id = commit_id; + match &mut self.task_type { + ArgsTaskType::Commit(commit_task) => { + commit_task.commit_id = commit_id; + } + ArgsTaskType::CompressedCommit(commit_task) => { + commit_task.commit_id = commit_id; + } + _ => {} + } }programs/magicblock/src/schedule_transactions/process_schedule_commit.rs (1)
141-173: Undelegation branch only enforces writability, not delegated state – potential invariant leakIn this block, when
opts.request_undelegationistrue, you only enforce that the account is writable viaget_writable_with_idx, and then you skip thedelegated()check that’s enforced for the non‑undelegate path:if opts.request_undelegation { // writable check... if !acc_writable { ... ReadonlyDataModified } } else if !acc.borrow().delegated() { ... IllegalOwner }That means a caller can pass an arbitrary writable but non‑delegated account and still get it committed and “undelegated”, which looks inconsistent with the intent of the surrounding checks and with the comment that “we rely on invariant
writable means delegated”. Since account metas are ultimately under user control, treating “writable ⇒ delegated” as a hard invariant here seems fragile.I’d strongly recommend also asserting
delegated()in the undelegation branch, so both flows enforce delegation, with undelegation adding the extra writability requirement. For example:{ - if opts.request_undelegation { + let is_delegated = acc.borrow().delegated(); + + if opts.request_undelegation { // Since we need to modify the account during undelegation, we expect it to be writable // We rely on invariant "writable means delegated" let acc_writable = get_writable_with_idx(transaction_context, idx as u16)?; if !acc_writable { ic_msg!( invoke_context, "ScheduleCommit ERR: account {} is required to be writable in order to be undelegated", acc_pubkey ); return Err(InstructionError::ReadonlyDataModified); } - } else if !acc.borrow().delegated() { + if !is_delegated { + ic_msg!( + invoke_context, + "ScheduleCommit ERR: account {} is required to be delegated to the current validator, in order to be undelegated", + acc_pubkey + ); + return Err(InstructionError::IllegalOwner); + } + } else if !is_delegated { ic_msg!( invoke_context, "ScheduleCommit ERR: account {} is required to be delegated to the current validator, in order to be committed", acc_pubkey ); return Err(InstructionError::IllegalOwner); };This keeps behavior for the existing commit-only path intact, tightens the undelegation path to the same delegation requirement, and aligns better with the comment expectation. You may also want to add a negative test for “undelegate with writable but non‑delegated account” to lock this in.
programs/magicblock/src/schedule_transactions/process_schedule_commit_tests.rs (2)
88-143: Three-committee helper parametrization is useful; minor duplication could be reducedExtending
prepare_transaction_with_three_committeeswithis_delegated: (bool, bool, bool)is a good way to control per-committee delegation state for tests, and settingdelegatedon eachAccountSharedDatahere centralizes the setup.There’s some repeated boilerplate when constructing the three committee accounts that could be made less error-prone. For example:
- { - let mut acc = AccountSharedData::new(0, 0, &program); - acc.set_delegated(is_delegated.0); - map.insert(committee_uno, acc); - } - { - let mut acc = AccountSharedData::new(0, 0, &program); - acc.set_delegated(is_delegated.1); - map.insert(committee_dos, acc); - } - { - let mut acc = AccountSharedData::new(0, 0, &program); - acc.set_delegated(is_delegated.2); - map.insert(committee_tres, acc); - } + for (committee, delegated) in [ + (committee_uno, is_delegated.0), + (committee_dos, is_delegated.1), + (committee_tres, is_delegated.2), + ] { + let mut acc = AccountSharedData::new(0, 0, &program); + acc.set_delegated(delegated); + map.insert(committee, acc); + }This keeps the three branches in lockstep and makes it harder to accidentally mismatch a committee/flag pairing in future refactors.
451-463: Callsites consistently passing(true, true, true); consider a named helper/constantAll updated callsites that rely on fully delegated committees now pass
(true, true, true)intoprepare_transaction_with_three_committees, which preserves previous semantics.To improve readability and avoid magic tuples scattered around, consider a small helper or constant, e.g.:
const ALL_DELEGATED: (bool, bool, bool) = (true, true, true); // ... prepare_transaction_with_three_committees(&payer, None, ALL_DELEGATED);or a dedicated helper like
prepare_fully_delegated_transaction_with_three_committees(...)that wraps this pattern. That would make the intent clearer and reduce duplication.Also applies to: 503-510, 563-575, 617-624, 707-715
magicblock-committor-service/src/intent_executor/mod.rs (2)
60-84: Metrics wiring (ExecutionOutput label + CU tracking) and result persistence look good.
ExecutionOutputnow implementsmetrics::LabelValuewith clear"single_stage_succeeded"/"two_stage_succeeded"labels, aligning with howmagicblock-metricswrapsResultvalues.intent_metricsfetches transaction status with the executor’s commitment level, extractscompute_units_consumed, and records a single CU value (sum across stages for two-stage). Overflow handling on the i64 cast is bounded.persist_resultcorrectly adapts to the newIntentExecutorErrorshapes (two-fieldCommitIDError/ActionsError/CpiLimitError) while preserving the previous status semantics.All of this is side-effect-only (logging/metrics) and doesn’t threaten the core execution paths.
If you want stronger guarantees around metrics visibility, you could consider logging a structured event when CU data is missing (currently swallowed as
_ => {}) to help debug gaps in CU reporting.Also applies to: 510-610, 713-770, 779-813
123-162: Compression guard intry_unite_tasksand test coverage are reasonable but could be expanded.The early return in
try_unite_taskswhen any commit/finalize task is compressed prevents accidental unification of compressed flows into a single transaction, which is a sensible conservative choice and keeps the existing single-stage optimization strictly for non-compressed cases. Thetest_try_unitetest still validates the happy path for uncompressed tasks.You may eventually want a dedicated test that builds a compressed intent and asserts that
try_unite_tasksreturnsOk(None)to lock in this guard.Also applies to: 221-239, 622-651, 861-899
magicblock-aperture/src/server/websocket/connection.rs (1)
98-157: Keep‑alive logic is sound; consider explicitly handling CLOSE framesThe new
PING_PERIOD/MAX_INACTIVE_INTERVALlogic with a pinnednext_pingfuture and rescheduling on inbound frames gives clear “ping every 30s, close after 60s of no inbound traffic” semantics, and updatinglast_activityplus resettingnext_pingon every received frame is consistent. The change tolet _ = self.report_failure(..).await;also avoids tearing down the connection when a parse‑error response cannot be delivered.Two minor points to consider:
- Frames with
OpCode::Closeare now treated as “non‑Text/Binary” and simply ignored (while still refreshinglast_activity). That means a client‑initiated close will keep the connection alive until the idle timeout or a later I/O failure. You may want to special‑caseOpCode::Closeto send a close frame and break the loop to honor the WebSocket close handshake promptly.- If
report_failurereturnsfalse(write failed), you currently continue the loop and rely on later I/O to fail. If you consider a failed error‑response write a terminal condition, you could break in that case to release resources earlier.magicblock-ledger/src/store/api.rs (3)
1784-1918: Tests generally match new per-slot indexing, but one assertion uses an oldresThe updated tests around
get_confirmed_signatures_for_addressnicely exercise the new per-slot index behavior and the addedwrite_blockcalls align with how indices are derived fromwrite_transaction_status.There is, however, a small correctness bug in
test_find_address_signatures_no_intra_slot_limits:// No before/until { let sigs = extract( store .get_confirmed_signatures_for_address( read_uno, slot_seis, None, None, 1000, ) .unwrap() .infos, ); assert!(!res.found_upper); assert_eq!(sigs, vec![/* ... */]); }Here
resrefers to the earlierlet res = store.get_confirmed_signatures_for_address(read_cuatro, ...), soassert!(!res.found_upper)is asserting against the wrong call and does not validate thefound_upperflag for theread_unoquery at all.You likely want:
- { - let sigs = extract( - store - .get_confirmed_signatures_for_address( - read_uno, slot_seis, None, None, 1000, - ) - .unwrap() - .infos, - ); - assert!(!res.found_upper); - assert_eq!( - sigs, - vec![ - (slot_seis, signature_seis), - (slot_tres, signature_tres), - (slot_dos, signature_dos), - (slot_uno, signature_uno), - ] - ); - } + { + let res = store + .get_confirmed_signatures_for_address( + read_uno, slot_seis, None, None, 1000, + ) + .unwrap(); + let sigs = extract(res.infos.clone()); + assert!(!res.found_upper); + assert_eq!( + sigs, + vec![ + (slot_seis, signature_seis), + (slot_tres, signature_tres), + (slot_dos, signature_dos), + (slot_uno, signature_uno), + ] + ); + }The rest of the test adjustments (extra
write_blockcalls, updated expected orderings and ranges) look consistent with the new indexing semantics.Also applies to: 1958-1965, 2000-2013, 2088-2095, 2101-2133, 2166-2168
480-523: Fix required for same-slot before+until edge case that produces duplicates and wrong resultsThe review identifies a real bug with major impact. When both
upper_limit_signatureandlower_limit_signaturerefer to transactions in the same slot, the current three-phase logic fails:Current behavior (lines 531-595 + 661-719):
- Upper phase adds all signatures with
tx_idx < upper_idx- Lower phase adds all signatures with
tx_idx > lower_idx- Middle phase skipped because
newest_slot = upper_slot - 1, makingnewest_slot > lower_slotfalseThis produces duplicates and includes out-of-range signatures. The test suite lacks coverage for this scenario—all dual-limit tests use different slots, leaving the same-slot case untested.
Required changes:
- Add special-case handling for
found_upper && found_lower && include_upper && include_lower && upper_slot == lower_slot- Fetch both
upper_idxandlower_idx, assertlower_idx < upper_idx- Execute a single pass collecting only entries where
lower_idx < tx_idx < upper_idx- Skip the existing upper- and lower-phase loops for this case
- Add a unit test exercising same-slot before+until to prevent regression
1225-1323: Clearblock_txn_indexescache during slot truncation indelete_slot_rangeThe issue is confirmed:
delete_slot_range(lines 1224–1324) properly removes all on-disk transaction data for the truncated slot range but leaves entries in the in-memoryblock_txn_indexes: HashCache<Slot, AtomicU32>cache uncleaned. This results in:
- Persistent in-memory cache entries until natural HashCache eviction
- If truncated slots are later rewritten, transaction indices resume from the old counter value rather than zero
The test
test_truncate_slotsverifies on-disk deletion but does not check in-memory cache state or test slot reuse scenarios. The suggested fix—iterating through the range and calling.remove()on the cache—is appropriate and safe:pub fn delete_slot_range( &self, from_slot: Slot, to_slot: Slot, ) -> LedgerResult<()> { self.set_lowest_cleanup_slot(to_slot); + // Best-effort cleanup of per-slot index counters for truncated slots. + for slot in from_slot..=to_slot { + let _ = self.block_txn_indexes.remove(&slot); + }
♻️ Duplicate comments (9)
programs/magicblock/src/utils/instruction_utils.rs (1)
61-95: Compressed commit helpers correctly mirror the existing patternThe implementation correctly mirrors the existing
schedule_commithelpers with the appropriateScheduleCompressedCommitinstruction variant and identical account metadata construction.test-integration/programs/schedulecommit/src/api.rs (1)
28-32: Serialization pattern and error handling concerns remain.The switch from
Instruction::new_with_borshto manualborsh::to_vec().unwrap()withInstruction::new_with_bytesintroduces the same concerns raised in previous reviews: potential panics from.unwrap()and repeated serialization patterns across multiple instruction builders.Also applies to: 86-90, 194-198, 218-227, 233-237
Cargo.toml (2)
96-100: Prefer pinning light-protocol deps byrevinstead of mutable branch.Lines 96–100 reference the light-protocol dependencies on the
dode/sdk-v2-2branch. Mutable branch references can cause non-reproducible builds over time if the branch is force-pushed or updated. Based on learnings, it's recommended to pin by a specific revision (rev) instead.Once the
dode/sdk-v2-2branch is stable, convert to revision-pinned dependencies:- light-client = { git = "https://github.com/magicblock-labs/light-protocol", branch = "dode/sdk-v2-2" } - light-compressed-account = { git = "https://github.com/magicblock-labs/light-protocol", branch = "dode/sdk-v2-2" } - light-hasher = { git = "https://github.com/magicblock-labs/light-protocol", branch = "dode/sdk-v2-2" } - light-sdk = { git = "https://github.com/magicblock-labs/light-protocol", branch = "dode/sdk-v2-2" } - light-sdk-types = { git = "https://github.com/magicblock-labs/light-protocol", branch = "dode/sdk-v2-2" } + light-client = { git = "https://github.com/magicblock-labs/light-protocol", rev = "<stable-commit-hash>" } + light-compressed-account = { git = "https://github.com/magicblock-labs/light-protocol", rev = "<stable-commit-hash>" } + light-hasher = { git = "https://github.com/magicblock-labs/light-protocol", rev = "<stable-commit-hash>" } + light-sdk = { git = "https://github.com/magicblock-labs/light-protocol", rev = "<stable-commit-hash>" } + light-sdk-types = { git = "https://github.com/magicblock-labs/light-protocol", rev = "<stable-commit-hash>" }
58-59: CRITICAL: Unresolved borsh version mismatch will cause build failures.Lines 58–59 have an incompatible major version mismatch:
borsh 0.10.4andborsh-derive 1.5.1. These major versions are not compatible; borsh-derive 1.5.x expects borsh 1.x, and using 0.10.x alongside 1.5.x commonly produces duplicate/trait/crate-not-found errors during macro expansion.Additionally,
magicblock-apidepends onborsh 1.5.3, fragmenting the workspace further.Required fix: Align borsh and borsh-derive to the same major version across the entire workspace. Choose one path:
- Option A (recommended for Solana 2.2): Use borsh 1.x series to align with modern ecosystem patterns.
- Option B: Use borsh 0.10.x series and downgrade borsh-derive to 0.10.x.
Apply this diff to start (choose Option A or B):
- borsh = { version = "0.10.4" } - borsh-derive = "1.5.1" + borsh = { version = "1.5.3" } + borsh-derive = "1.5.3"Then update
magicblock-api's Cargo.toml to match, and runcargo updateto ensure the workspace resolves to a single borsh version. Verify the build succeeds before merging..github/workflows/ci-test-integration.yml (1)
33-40: Use stable zk-compression CLI version instead of prerelease.Line 40 installs a prerelease version
@lightprotocol/zk-compression-cli@0.27.1-alpha.2. The latest stable version is0.27.0, which should be used in CI to avoid stability risks.Update the npm install command to use the stable release:
- npm i -g @lightprotocol/zk-compression-cli@0.27.1-alpha.2 + npm i -g @lightprotocol/zk-compression-cli@0.27.0magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs (1)
782-788: Consider simplifying the closure pattern.The addition of the
_compressedparameter aligns with the broader compressed delegation support. The mock implementation correctly ignores this parameter.As noted in a previous review, the closure pattern could be more concise:
- Ok(pubkeys.iter().map(|pubkey| (*pubkey, 1)).collect()) + Ok(pubkeys.iter().map(|&k| (k, 1)).collect())programs/magicblock/src/schedule_transactions/process_schedule_commit.rs (1)
34-55: Wrapper split and sharedschedule_commitcore look good; consider documentingcompressedRefactoring into
process_schedule_commit/process_schedule_compressed_commitwrappers with a sharedschedule_commit(..., compressed: bool)keeps the branching localized and is an appropriate abstraction for the new compressed flow.One follow-up: the bare
compressed: boolis not self-describing for future callers. A short doc comment (either onschedule_commitor both wrappers) explaining what “compressed” means in terms of expected callers and behavior (compressed-asset commit workflow vs standard commit) would help avoid misuse.magicblock-committor-service/src/intent_executor/mod.rs (1)
345-367: Mixed compressed/non-compressed tasks are still treated as a single boolean flag.
handle_commit_id_errorforwards a singlecompressedboolean tofetch_next_commit_idsbased onstrategy.optimized_tasks.iter().any(|task| task.is_compressed()), and there’s a TODO acknowledging mixed cases. This is the same concern previously raised: if a strategy ever contains both compressed and non-compressed tasks, this boolean conflates them and may fetch/assign incorrect commit IDs.Even if current strategies are homogeneous, it would be safer to either assert homogeneity up front (and fail fast if violated) or extend
fetch_next_commit_ids/strategy metadata to support per-task compression status.programs/magicblock/src/magic_scheduled_base_intent.rs (1)
109-115: Documentis_compressedonScheduledBaseIntentThis new public helper is part of the external API surface and encodes an important semantic (“uses compressed account delegation”). Adding a short doc comment will make its behavior and implications clearer to callers, especially around when compressed variants are set and what they mean for downstream processing (e.g., Photon indexer vs. normal account fetch).
For example:
- pub fn is_compressed(&self) -> bool { + /// Returns `true` if this base intent uses compressed account delegation. + /// + /// Compressed variants rely on the Photon indexer and compressed accounts + /// instead of standard Solana account fetching. + pub fn is_compressed(&self) -> bool { matches!( &self.base_intent, MagicBaseIntent::CompressedCommit(_) | MagicBaseIntent::CompressedCommitAndUndelegate(_) ) }
📜 Review details
Configuration used: CodeRabbit UI
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (2)
Cargo.lockis excluded by!**/*.locktest-integration/Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (70)
.github/workflows/ci-fmt.yml(1 hunks).github/workflows/ci-lint.yml(1 hunks).github/workflows/ci-test-integration.yml(2 hunks).github/workflows/ci-test-unit.yml(1 hunks).github/workflows/deploy-testnet-by-pr-comment.yml(1 hunks).github/workflows/publish-packages.yml(2 hunks).github/workflows/slack-notify.yml(1 hunks)Cargo.toml(6 hunks)Makefile(1 hunks)magicblock-accounts/src/scheduled_commits_processor.rs(3 hunks)magicblock-aperture/Cargo.toml(1 hunks)magicblock-aperture/src/encoder.rs(6 hunks)magicblock-aperture/src/requests/http/mod.rs(4 hunks)magicblock-aperture/src/requests/http/send_transaction.rs(3 hunks)magicblock-aperture/src/requests/http/simulate_transaction.rs(1 hunks)magicblock-aperture/src/requests/mod.rs(1 hunks)magicblock-aperture/src/requests/websocket/account_subscribe.rs(1 hunks)magicblock-aperture/src/requests/websocket/log_subscribe.rs(1 hunks)magicblock-aperture/src/requests/websocket/program_subscribe.rs(1 hunks)magicblock-aperture/src/requests/websocket/slot_subscribe.rs(1 hunks)magicblock-aperture/src/server/http/dispatch.rs(2 hunks)magicblock-aperture/src/server/websocket/connection.rs(7 hunks)magicblock-aperture/src/server/websocket/dispatch.rs(4 hunks)magicblock-aperture/src/state/subscriptions.rs(9 hunks)magicblock-aperture/src/utils.rs(1 hunks)magicblock-aperture/tests/transactions.rs(2 hunks)magicblock-api/src/tickers.rs(2 hunks)magicblock-chainlink/src/chainlink/mod.rs(10 hunks)magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs(4 hunks)magicblock-committor-service/src/intent_executor/error.rs(7 hunks)magicblock-committor-service/src/intent_executor/mod.rs(22 hunks)magicblock-committor-service/src/intent_executor/single_stage_executor.rs(7 hunks)magicblock-committor-service/src/intent_executor/two_stage_executor.rs(6 hunks)magicblock-committor-service/src/tasks/args_task.rs(7 hunks)magicblock-committor-service/src/tasks/buffer_task.rs(4 hunks)magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs(10 hunks)magicblock-committor-service/src/types.rs(2 hunks)magicblock-ledger/Cargo.toml(1 hunks)magicblock-ledger/src/database/columns.rs(2 hunks)magicblock-ledger/src/database/compaction_filter.rs(1 hunks)magicblock-ledger/src/database/ledger_column.rs(4 hunks)magicblock-ledger/src/database/rocks_db.rs(2 hunks)magicblock-ledger/src/ledger_truncator.rs(1 hunks)magicblock-ledger/src/store/api.rs(22 hunks)magicblock-ledger/src/store/utils.rs(1 hunks)magicblock-ledger/tests/common.rs(1 hunks)magicblock-ledger/tests/get_block.rs(1 hunks)magicblock-ledger/tests/test_ledger_truncator.rs(4 hunks)magicblock-magic-program-api/Cargo.toml(1 hunks)magicblock-metrics/src/metrics/mod.rs(7 hunks)magicblock-metrics/src/metrics/types.rs(1 hunks)magicblock-processor/Cargo.toml(1 hunks)magicblock-processor/src/executor/mod.rs(1 hunks)magicblock-processor/src/executor/processing.rs(2 hunks)magicblock-processor/src/scheduler.rs(1 hunks)magicblock-rpc-client/src/lib.rs(3 hunks)magicblock-rpc-client/src/utils.rs(1 hunks)programs/magicblock/src/magic_scheduled_base_intent.rs(12 hunks)programs/magicblock/src/mutate_accounts/account_mod_data.rs(1 hunks)programs/magicblock/src/schedule_transactions/process_schedule_commit.rs(5 hunks)programs/magicblock/src/schedule_transactions/process_schedule_commit_tests.rs(10 hunks)programs/magicblock/src/utils/instruction_utils.rs(3 hunks)test-integration/programs/schedulecommit/src/api.rs(6 hunks)test-integration/programs/schedulecommit/src/lib.rs(2 hunks)test-integration/schedulecommit/client/src/schedule_commit_context.rs(1 hunks)test-integration/schedulecommit/test-scenarios/tests/01_commits.rs(2 hunks)test-integration/test-chainlink/tests/ix_remote_account_provider.rs(4 hunks)test-integration/test-cloning/tests/01_program-deploy.rs(1 hunks)test-integration/test-committor-service/tests/test_delivery_preparator.rs(6 hunks)test-integration/test-committor-service/tests/test_intent_executor.rs(15 hunks)
🧰 Additional context used
🧠 Learnings (15)
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-ledger/Cargo.tomlprograms/magicblock/src/mutate_accounts/account_mod_data.rstest-integration/programs/schedulecommit/src/api.rsmagicblock-ledger/tests/common.rsmagicblock-aperture/src/requests/websocket/account_subscribe.rsmagicblock-aperture/src/requests/websocket/log_subscribe.rsmagicblock-aperture/src/encoder.rsmagicblock-aperture/src/requests/http/mod.rsmagicblock-aperture/src/requests/http/send_transaction.rsmagicblock-aperture/src/requests/websocket/slot_subscribe.rsprograms/magicblock/src/schedule_transactions/process_schedule_commit_tests.rsprograms/magicblock/src/schedule_transactions/process_schedule_commit.rsmagicblock-processor/src/executor/processing.rsmagicblock-ledger/src/store/api.rsmagicblock-aperture/src/state/subscriptions.rstest-integration/test-chainlink/tests/ix_remote_account_provider.rsmagicblock-aperture/tests/transactions.rsCargo.tomlprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-aperture/src/server/websocket/dispatch.rsprograms/magicblock/src/utils/instruction_utils.rsmagicblock-committor-service/src/transaction_preparator/delivery_preparator.rsmagicblock-api/src/tickers.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
programs/magicblock/src/mutate_accounts/account_mod_data.rsmagicblock-ledger/src/database/columns.rsmagicblock-ledger/src/ledger_truncator.rsmagicblock-processor/src/executor/mod.rsmagicblock-aperture/src/requests/http/mod.rsprograms/magicblock/src/schedule_transactions/process_schedule_commit_tests.rsprograms/magicblock/src/schedule_transactions/process_schedule_commit.rsmagicblock-processor/src/executor/processing.rsmagicblock-ledger/src/store/api.rsmagicblock-processor/src/scheduler.rsmagicblock-api/src/tickers.rs
📚 Learning: 2025-11-12T09:46:27.553Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 614
File: magicblock-task-scheduler/src/db.rs:26-0
Timestamp: 2025-11-12T09:46:27.553Z
Learning: In magicblock-task-scheduler, task parameter validation (including ensuring iterations > 0 and enforcing minimum execution intervals) is performed in the Magic program (on-chain) before ScheduleTaskRequest instances reach the scheduler service. The From<&ScheduleTaskRequest> conversion in db.rs does not need additional validation because inputs are already validated at the program level.
Applied to files:
magicblock-committor-service/src/types.rs
📚 Learning: 2025-11-07T13:09:52.253Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: test-kit/src/lib.rs:275-0
Timestamp: 2025-11-07T13:09:52.253Z
Learning: In test-kit, the transaction scheduler in ExecutionTestEnv is not expected to shut down during tests. Therefore, using `.unwrap()` in test helper methods like `schedule_transaction` is acceptable and will not cause issues in the test environment.
Applied to files:
test-integration/schedulecommit/test-scenarios/tests/01_commits.rsmagicblock-processor/src/executor/mod.rsmagicblock-aperture/src/requests/http/send_transaction.rsprograms/magicblock/src/schedule_transactions/process_schedule_commit_tests.rsmagicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rsmagicblock-ledger/tests/get_block.rsmagicblock-processor/src/scheduler.rsmagicblock-aperture/tests/transactions.rsmagicblock-committor-service/src/intent_executor/single_stage_executor.rstest-integration/test-committor-service/tests/test_intent_executor.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
test-integration/schedulecommit/test-scenarios/tests/01_commits.rsmagicblock-ledger/tests/get_block.rsmagicblock-aperture/tests/transactions.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-accounts/src/scheduled_commits_processor.rsmagicblock-aperture/src/requests/websocket/account_subscribe.rsmagicblock-aperture/src/requests/websocket/log_subscribe.rsmagicblock-aperture/src/requests/websocket/slot_subscribe.rsmagicblock-processor/src/executor/processing.rsmagicblock-aperture/src/state/subscriptions.rstest-integration/test-chainlink/tests/ix_remote_account_provider.rsmagicblock-aperture/src/requests/websocket/program_subscribe.rsmagicblock-aperture/src/server/websocket/dispatch.rsmagicblock-chainlink/src/chainlink/mod.rs
📚 Learning: 2025-10-28T13:15:42.706Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 596
File: magicblock-processor/src/scheduler.rs:1-1
Timestamp: 2025-10-28T13:15:42.706Z
Learning: In magicblock-processor, transaction indexes were always set to 0 even before the changes in PR #596. The proper transaction indexing within slots will be addressed during the planned ledger rewrite.
Applied to files:
magicblock-ledger/tests/common.rsmagicblock-ledger/tests/test_ledger_truncator.rsmagicblock-processor/src/executor/mod.rsmagicblock-processor/src/executor/processing.rsmagicblock-ledger/tests/get_block.rsmagicblock-ledger/src/store/api.rsmagicblock-processor/src/scheduler.rs
📚 Learning: 2025-10-21T11:00:18.396Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/encoder.rs:176-187
Timestamp: 2025-10-21T11:00:18.396Z
Learning: In the magicblock validator, the current slot is always the root slot. The SlotEncoder in magicblock-aperture/src/encoder.rs correctly sets `root: slot` because there is no lag between current and root slots in this architecture.
Applied to files:
magicblock-ledger/tests/test_ledger_truncator.rsmagicblock-aperture/src/encoder.rsmagicblock-ledger/src/store/api.rsmagicblock-processor/src/scheduler.rs
📚 Learning: 2025-11-04T10:53:50.922Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/locks.rs:110-122
Timestamp: 2025-11-04T10:53:50.922Z
Learning: In magicblock-processor, the TransactionScheduler runs in a single, dedicated thread and will always remain single-threaded. The `next_transaction_id()` function in scheduler/locks.rs uses `unsafe static mut` which is safe given this architectural guarantee.
Applied to files:
magicblock-aperture/src/server/http/dispatch.rsmagicblock-processor/src/executor/mod.rsmagicblock-processor/src/scheduler.rs
📚 Learning: 2025-11-13T09:38:43.804Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/locks.rs:64-102
Timestamp: 2025-11-13T09:38:43.804Z
Learning: In magicblock-processor's TransactionScheduler (scheduler/mod.rs line 59), the executor count is clamped to MAX_SVM_EXECUTORS (63) at initialization time, and executor IDs are assigned sequentially from 0 to count-1. This architectural guarantee ensures that executor IDs used in the bitmask-based AccountLock (scheduler/locks.rs) will always be within valid bounds for bit shifting operations, making runtime bounds checks unnecessary.
Applied to files:
magicblock-aperture/src/server/http/dispatch.rsmagicblock-processor/src/executor/mod.rsmagicblock-processor/src/executor/processing.rsmagicblock-processor/src/scheduler.rs
📚 Learning: 2025-10-21T10:34:59.140Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-accounts-db/src/lib.rs:63-72
Timestamp: 2025-10-21T10:34:59.140Z
Learning: In magicblock-validator, the AccountsDb "stop-the-world" synchronizer is managed at the processor/executor level, not at the AccountsDb API level. Transaction executors in magicblock-processor hold a read lock (sync.read()) for the duration of each slot and release it only at slot boundaries, ensuring all account writes happen under the read lock. Snapshot operations acquire a write lock, blocking until all executors release their read locks. This pattern ensures mutual exclusion between writes and snapshots without requiring read guards in AccountsDb write APIs.
Applied to files:
magicblock-processor/src/executor/mod.rsmagicblock-processor/src/executor/processing.rsmagicblock-processor/src/scheduler.rs
📚 Learning: 2025-11-04T10:48:00.070Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/mod.rs:217-219
Timestamp: 2025-11-04T10:48:00.070Z
Learning: In magicblock-validator, the codebase uses a pattern where types containing non-Send/non-Sync fields (like Rc<RefCell<...>>) are marked with unsafe impl Send when they are guaranteed to be confined to a single thread through careful API design and thread spawning patterns.
Applied to files:
magicblock-processor/src/executor/mod.rsmagicblock-processor/src/executor/processing.rsmagicblock-processor/src/scheduler.rs
📚 Learning: 2025-10-21T13:06:38.900Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/http/get_program_accounts.rs:17-25
Timestamp: 2025-10-21T13:06:38.900Z
Learning: The magicblock validator does not support ledger forking, so commitment-based state queries (processed/confirmed/finalized) are not applicable. RPC methods can safely ignore commitment and minContextSlot parameters from Solana RPC config objects.
Applied to files:
magicblock-aperture/Cargo.tomlmagicblock-api/src/tickers.rs
📚 Learning: 2025-11-07T18:19:48.996Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 614
File: test-integration/programs/schedulecommit/Cargo.toml:19-20
Timestamp: 2025-11-07T18:19:48.996Z
Learning: In Solana programs (crates depending on solana-program), the empty feature flags `custom-heap = []` and `custom-panic = []` are standard declarations required to avoid compiler warnings. These should not be flagged as unused features in Cargo.toml files for Solana program crates.
Applied to files:
magicblock-magic-program-api/Cargo.toml
📚 Learning: 2025-10-26T08:49:31.543Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 585
File: magicblock-committor-service/src/tasks/buffer_task.rs:111-115
Timestamp: 2025-10-26T08:49:31.543Z
Learning: In the magicblock-committor-service, compute units returned by the `compute_units()` method in task implementations (such as `BufferTask`, `ArgsTask`, etc.) represent the compute budget for a single task. Transactions can comprise multiple tasks, and the total compute budget for a transaction is computed as the sum of the compute units of all tasks included in that transaction.
Applied to files:
magicblock-committor-service/src/tasks/args_task.rsmagicblock-committor-service/src/tasks/buffer_task.rsmagicblock-committor-service/src/intent_executor/mod.rs
🧬 Code graph analysis (29)
magicblock-committor-service/src/types.rs (3)
magicblock-committor-service/src/intent_executor/mod.rs (1)
value(75-83)magicblock-committor-service/src/intent_executor/error.rs (1)
value(105-112)magicblock-metrics/src/metrics/types.rs (2)
value(73-73)value(81-86)
test-integration/schedulecommit/test-scenarios/tests/01_commits.rs (3)
test-integration/programs/schedulecommit/src/api.rs (4)
delegate_account_cpi_instruction(58-91)init_account_instruction(15-33)pda_and_bump(263-267)schedule_commit_cpi_instruction(104-122)test-integration/test-tools/src/integration_test_context.rs (3)
signature(63-65)try_new(113-115)airdrop_chain(483-494)test-integration/schedulecommit/test-scenarios/tests/utils/mod.rs (1)
extract_transaction_error(255-264)
magicblock-metrics/src/metrics/types.rs (3)
magicblock-committor-service/src/intent_executor/mod.rs (1)
value(75-83)magicblock-committor-service/src/intent_executor/error.rs (1)
value(105-112)magicblock-committor-service/src/types.rs (1)
value(23-29)
magicblock-ledger/src/ledger_truncator.rs (1)
magicblock-api/src/magic_validator.rs (1)
ledger(741-743)
magicblock-ledger/tests/test_ledger_truncator.rs (1)
magicblock-ledger/tests/common.rs (1)
write_dummy_transaction(25-47)
test-integration/test-committor-service/tests/test_delivery_preparator.rs (2)
test-integration/test-committor-service/tests/common.rs (4)
create_commit_task(166-182)create_compressed_commit_task(185-212)generate_random_bytes(158-163)new(63-66)test-integration/test-committor-service/tests/utils/transactions.rs (1)
init_and_delegate_compressed_account_on_chain(233-322)
magicblock-aperture/src/requests/http/mod.rs (1)
magicblock-aperture/src/requests/http/get_multiple_accounts.rs (1)
pubkeys(31-39)
magicblock-aperture/src/requests/http/send_transaction.rs (1)
magicblock-aperture/src/requests/mod.rs (1)
params(22-26)
magicblock-ledger/src/database/ledger_column.rs (1)
magicblock-ledger/src/database/columns.rs (9)
columns(105-117)slot(131-131)slot(181-183)slot(249-251)slot(329-331)slot(404-406)slot(503-505)slot(577-579)slot(649-651)
magicblock-ledger/src/database/compaction_filter.rs (1)
magicblock-ledger/src/database/columns.rs (10)
key(125-125)key(170-174)key(234-243)key(318-323)key(393-398)key(495-497)key(566-571)key(641-643)keep_all_on_compaction(132-134)keep_all_on_compaction(659-661)
programs/magicblock/src/schedule_transactions/process_schedule_commit_tests.rs (3)
test-integration/programs/schedulecommit/src/lib.rs (1)
process_instruction(110-151)programs/magicblock/src/test_utils/mod.rs (1)
process_instruction(44-61)programs/magicblock/src/utils/instruction_utils.rs (1)
schedule_commit_instruction(43-59)
programs/magicblock/src/schedule_transactions/process_schedule_commit.rs (2)
programs/magicblock/src/utils/accounts.rs (1)
get_writable_with_idx(104-111)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
compressed(154-156)
magicblock-committor-service/src/tasks/args_task.rs (3)
magicblock-committor-service/src/tasks/buffer_task.rs (7)
new(38-46)preparation_state(98-100)is_compressed(146-148)task_type(125-129)set_compressed_data(150-155)get_compressed_data(157-161)delegated_account(163-169)magicblock-committor-service/src/tasks/mod.rs (6)
preparation_state(77-77)is_compressed(99-99)task_type(93-93)set_compressed_data(102-102)get_compressed_data(105-105)delegated_account(108-108)magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)
magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs (2)
magicblock-metrics/src/metrics/mod.rs (2)
observe_committor_intent_execution_time_histogram(383-391)inc_committor_failed_intents_count(370-377)programs/magicblock/src/magic_scheduled_base_intent.rs (2)
get_committed_pubkeys(97-99)get_committed_pubkeys(198-202)
magicblock-ledger/tests/get_block.rs (1)
magicblock-ledger/tests/common.rs (1)
write_dummy_transaction(25-47)
magicblock-aperture/src/state/subscriptions.rs (3)
magicblock-aperture/src/server/websocket/dispatch.rs (1)
new(51-63)magicblock-aperture/src/server/websocket/connection.rs (1)
new(66-84)magicblock-aperture/src/state/mod.rs (1)
new(77-95)
magicblock-committor-service/src/tasks/buffer_task.rs (3)
magicblock-committor-service/src/tasks/args_task.rs (4)
is_compressed(243-250)set_compressed_data(252-265)get_compressed_data(267-280)delegated_account(282-298)magicblock-committor-service/src/tasks/mod.rs (4)
is_compressed(99-99)set_compressed_data(102-102)get_compressed_data(105-105)delegated_account(108-108)magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)
test-integration/test-chainlink/tests/ix_remote_account_provider.rs (3)
magicblock-chainlink/src/testing/utils.rs (2)
airdrop(24-27)sleep_ms(41-43)test-integration/test-chainlink/src/lib.rs (1)
sleep_ms(8-11)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
fresh_lamports(248-250)
programs/magicblock/src/magic_scheduled_base_intent.rs (2)
programs/magicblock/src/utils/accounts.rs (2)
get_writable_with_idx(104-111)get_instruction_pubkey_with_idx(94-102)magicblock-committor-service/src/tasks/mod.rs (1)
is_compressed(99-99)
magicblock-aperture/src/server/websocket/dispatch.rs (1)
magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs (1)
subscriptions(146-151)
programs/magicblock/src/utils/instruction_utils.rs (1)
programs/magicblock/src/schedule_task/process_schedule_task.rs (1)
pdas(203-214)
magicblock-rpc-client/src/utils.rs (4)
magicblock-rpc-client/src/lib.rs (3)
error(193-195)signature(82-90)from(228-230)magicblock-committor-service/src/intent_executor/error.rs (4)
signature(29-34)from(129-131)from(215-220)try_map(203-211)magicblock-committor-service/src/intent_executor/mod.rs (4)
map(669-681)decide_flow(683-692)None(874-874)None(894-894)magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (4)
map(361-371)decide_flow(373-382)from(542-546)try_map(333-349)
magicblock-chainlink/src/chainlink/mod.rs (2)
magicblock-chainlink/src/remote_account_provider/program_account.rs (1)
try_new(231-259)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
account(189-211)
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (4)
magicblock-rpc-client/src/utils.rs (6)
decide_rpc_error_flow(176-201)map_magicblock_client_error(93-141)send_transaction_with_retries(34-71)try_map(84-88)map(13-13)decide_flow(14-16)magicblock-committor-service/src/tasks/mod.rs (6)
instruction(68-68)instruction(304-310)get_compressed_data(105-105)chunks_pda(271-278)chunks_pda(322-329)write_instructions(243-260)magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(402-452)magicblock-committor-service/src/utils.rs (1)
persist_status_update(6-22)
test-integration/test-committor-service/tests/test_intent_executor.rs (2)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (1)
new(63-75)magicblock-committor-service/src/tasks/task_builder.rs (2)
finalize_tasks(56-60)finalize_tasks(167-377)
magicblock-api/src/tickers.rs (1)
magicblock-metrics/src/metrics/mod.rs (4)
set_ledger_size(284-286)set_accounts_size(339-341)set_accounts_count(343-345)observe_columns_count_duration(332-337)
magicblock-metrics/src/metrics/mod.rs (3)
magicblock-aperture/src/state/subscriptions.rs (2)
new(342-355)new(410-413)magicblock-metrics/src/metrics/types.rs (2)
value(73-73)value(81-86)magicblock-api/src/tickers.rs (1)
set_accounts_count(152-159)
magicblock-committor-service/src/intent_executor/mod.rs (7)
magicblock-rpc-client/src/lib.rs (4)
error(193-195)signature(82-90)new(235-237)from(228-230)magicblock-rpc-client/src/utils.rs (5)
decide_rpc_error_flow(176-201)map_magicblock_client_error(93-141)send_transaction_with_retries(34-71)map(13-13)decide_flow(14-16)magicblock-committor-service/src/intent_executor/error.rs (4)
signature(29-34)value(105-112)from(129-131)from(215-220)magicblock-metrics/src/metrics/types.rs (2)
value(73-73)value(81-86)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (7)
Self(298-306)Self(308-316)new(63-75)pubkeys(159-170)pubkeys(216-219)fetch_rent_reimbursements(39-42)fetch_rent_reimbursements(338-353)magicblock-committor-service/src/tasks/task_builder.rs (7)
commit_tasks(48-53)commit_tasks(70-164)finalize_task(173-190)accounts(105-108)accounts(312-315)finalize_tasks(56-60)finalize_tasks(167-377)magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs (4)
None(391-391)new(78-95)new(671-677)fetch_rent_reimbursements(790-795)
magicblock-committor-service/src/intent_executor/error.rs (6)
magicblock-rpc-client/src/lib.rs (3)
error(193-195)signature(82-90)from(228-230)magicblock-committor-service/src/intent_executor/mod.rs (1)
value(75-83)magicblock-committor-service/src/types.rs (1)
value(23-29)magicblock-metrics/src/metrics/types.rs (2)
value(73-73)value(81-86)magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (2)
from(542-546)try_map(333-349)magicblock-rpc-client/src/utils.rs (1)
try_map(84-88)
🪛 actionlint (1.7.8)
.github/workflows/ci-test-integration.yml
16-16: label "extra-large" is unknown. available labels are "windows-latest", "windows-latest-8-cores", "windows-2025", "windows-2022", "windows-11-arm", "ubuntu-latest", "ubuntu-latest-4-cores", "ubuntu-latest-8-cores", "ubuntu-latest-16-cores", "ubuntu-24.04", "ubuntu-24.04-arm", "ubuntu-22.04", "ubuntu-22.04-arm", "macos-latest", "macos-latest-xl", "macos-latest-xlarge", "macos-latest-large", "macos-26-xlarge", "macos-26", "macos-15-intel", "macos-15-xlarge", "macos-15-large", "macos-15", "macos-14-xl", "macos-14-xlarge", "macos-14-large", "macos-14", "macos-13-xl", "macos-13-xlarge", "macos-13-large", "macos-13", "self-hosted", "x64", "arm", "arm64", "linux", "macos", "windows". if it is a custom label for self-hosted runner, set list of labels in actionlint.yaml config file
(runner-label)
52-52: label "extra-large" is unknown. available labels are "windows-latest", "windows-latest-8-cores", "windows-2025", "windows-2022", "windows-11-arm", "ubuntu-latest", "ubuntu-latest-4-cores", "ubuntu-latest-8-cores", "ubuntu-latest-16-cores", "ubuntu-24.04", "ubuntu-24.04-arm", "ubuntu-22.04", "ubuntu-22.04-arm", "macos-latest", "macos-latest-xl", "macos-latest-xlarge", "macos-latest-large", "macos-26-xlarge", "macos-26", "macos-15-intel", "macos-15-xlarge", "macos-15-large", "macos-15", "macos-14-xl", "macos-14-xlarge", "macos-14-large", "macos-14", "macos-13-xl", "macos-13-xlarge", "macos-13-large", "macos-13", "self-hosted", "x64", "arm", "arm64", "linux", "macos", "windows". if it is a custom label for self-hosted runner, set list of labels in actionlint.yaml config file
(runner-label)
test-integration/test-committor-service/tests/test_delivery_preparator.rs
Show resolved
Hide resolved
test-integration/test-committor-service/tests/test_delivery_preparator.rs
Show resolved
Hide resolved
test-integration/test-committor-service/tests/test_delivery_preparator.rs
Show resolved
Hide resolved
There was a problem hiding this comment.
Actionable comments posted: 14
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (5)
magicblock-api/Cargo.toml (1)
13-13: Critical: Borsh version fragmentation in workspace.Line 13 directly pins
borsh = "1.5.3", but the workspace root (Cargo.toml line 58) specifiesborsh = "0.10.4". This creates two incompatible borsh versions in the same workspace, breaking crate interoperability. Additionally, the workspace borsh-derive is 1.5.1, which doesn't match either version properly.Required fix: Either align magicblock-api to use workspace borsh 0.10.4 (by changing line 13 to
borsh = { workspace = true }), or align the entire workspace to borsh 1.5.x and update borsh-derive accordingly. Choose one major version series consistently across all files.Also applies to: 18-18, 42-42
magicblock-committor-service/src/tasks/task_builder.rs (3)
71-105: Compressed commit task construction movesaccountsand can panic on fetch failureIn
commit_tasks, the compressed branch has two problems:
- Ownership bug –
for account in accounts { ... }movesaccounts(aVec<_>), so the subsequentaccounts.iter().zip(compressed_results)will not compile.- Panic on compressed data failure –
compressed_resultsis aVec<Result<CompressedData, TaskBuilderError>>, but you later docompressed_data.expect("Compressed commit task must be provided with compressed data"), which will panic instead of returning aTaskBuilderErrorwhenget_compressed_datafails.Both should be handled by iterating over
&accountsand propagating errors instead of panicking. For example:- let tasks = if compressed { - // For compressed accounts, prepare compression data - let photon_client = photon_client - .as_ref() - .ok_or(TaskBuilderError::PhotonClientNotFound)?; - let mut compressed_results = vec![]; - for account in accounts { - compressed_results.push( - get_compressed_data(&account.pubkey, photon_client).await, - ); - } - - accounts.iter().zip(compressed_results).map(|(account, compressed_data)| { - let commit_id = *commit_ids.get(&account.pubkey).expect("CommitIdFetcher provide commit ids for all listed pubkeys, or errors!"); - let compressed_data = compressed_data.expect("Compressed commit task must be provided with compressed data"); - let task = ArgsTaskType::CompressedCommit(CompressedCommitTask { - commit_id, - allow_undelegation, - committed_account: account.clone(), - compressed_data - }); - Box::new(ArgsTask::new(task)) as Box<dyn BaseTask> - }) - .collect() - } else { + let tasks = if compressed { + // For compressed accounts, prepare compression data + let photon_client = photon_client + .as_ref() + .ok_or(TaskBuilderError::PhotonClientNotFound)?; + + let mut tasks: Vec<Box<dyn BaseTask>> = + Vec::with_capacity(accounts.len()); + for account in &accounts { + let commit_id = *commit_ids.get(&account.pubkey).expect( + "CommitIdFetcher provide commit ids for all listed pubkeys, or errors!", + ); + let compressed_data = + get_compressed_data(&account.pubkey, photon_client).await?; + let task = ArgsTaskType::CompressedCommit(CompressedCommitTask { + commit_id, + allow_undelegation, + committed_account: account.clone(), + compressed_data, + }); + tasks.push(Box::new(ArgsTask::new(task)) as Box<dyn BaseTask>); + } + tasks + } else { accounts .iter() .map(|account| { let commit_id = *commit_ids.get(&account.pubkey).expect("CommitIdFetcher provide commit ids for all listed pubkeys, or errors!"); let task = ArgsTaskType::Commit(CommitTask { commit_id, allow_undelegation, committed_account: account.clone(), }); Box::new(ArgsTask::new(task)) as Box<dyn BaseTask> }) .collect() };This keeps
accountsborrowed, avoids the move, and uses?so compressed data fetch failures surface asTaskBuilderErrorinstead of panicking.Also applies to: 124-162
168-291: Finalize path silently downgrades compressed intents to non‑compressed finalize on fetch failureIn
finalize_tasks:
process_commit’s compressed branches invokeget_compressed_data(...).await.ok(), turningResult<CompressedData, TaskBuilderError>intoOption<CompressedData>.finalize_taskthen receivescompressed_data: Option<CompressedData>and falls back to the non‑compressedFinalizeTaskwhenNone.This means a compressed commit intent will be finalized via the non‑compressed DLP instruction if compressed data fetch fails (or Photon is temporarily unavailable), which is likely inconsistent with how compressed delegations are supposed to be closed out, and it hides the root cause.
Consider one of the following:
- Prefer strict behavior: propagate
TaskBuilderErroronget_compressed_datafailure so compressed finalize fails loudly, aligning with the commit path, e.g.:- let mut compressed_data = vec![]; - let photon_client = photon_client - .as_ref() - .ok_or(TaskBuilderError::PhotonClientNotFound)?; - for account in committed_accounts { - compressed_data.push( - get_compressed_data(&account.pubkey, photon_client) - .await - .ok(), - ); - } - - Ok(committed_accounts - .iter() - .zip(compressed_data) - .map(|(account, compressed_data)| { - finalize_task(account, compressed_data) - }) - .collect()) + let photon_client = photon_client + .as_ref() + .ok_or(TaskBuilderError::PhotonClientNotFound)?; + let mut tasks = Vec::with_capacity(committed_accounts.len()); + for account in committed_accounts { + let compressed_data = + get_compressed_data(&account.pubkey, photon_client) + .await?; + tasks.push(finalize_task( + account, + Some(compressed_data), + )); + } + Ok(tasks)…and similarly for the
WithBaseActionsbranch.
- Or, if downgrade is intentional: at least log/metric the failure when
get_compressed_datareturnsErr, so operators can see when compressed paths are failing and being converted to non‑compressed finalize.Right now the error is completely swallowed, which makes debugging and monitoring compressed flows hard and could mask real consistency issues.
381-455:get_compressed_dataassumes non‑emptypacked_tree_infosand can panic
get_compressed_datadoes good work surfacing most failure modes asTaskBuilderError, but:let packed_tree_accounts = proof_result .pack_tree_infos(&mut remaining_accounts) .state_trees .ok_or(TaskBuilderError::MissingStateTrees)?; let account_meta = CompressedAccountMeta { tree_info: packed_tree_accounts.packed_tree_infos[0], // ... };This will panic with an out‑of‑bounds access if
packed_tree_infosis empty even whenstate_treesisSome(_). To keep failure modes consistent with the rest of the function, consider:
- Checking
packed_tree_accounts.packed_tree_infos.first()and returning a dedicated error (e.g., reusingMissingStateTreesor adding a more specific variant) instead of indexing blindly, or- At least adding a debug assertion and comment if the light‑client API guarantees at least one entry here.
That will avoid a hard panic in the face of unexpected or malformed proofs.
magicblock-committor-service/src/intent_executor/mod.rs (1)
349-405: Usestrategy.compressedinstead of recomputing and clean up stale TODOIn
handle_commit_id_error(and related helpers) you now trackcompressedonTransactionStrategyand propagate it:let commit_ids = self .task_info_fetcher .fetch_next_commit_ids( committed_pubkeys, // TODO(dode): Handle cases where some tasks are compressed and some are not strategy .optimized_tasks .iter() .any(|task| task.is_compressed()), ) .await?; ... TransactionStrategy { optimized_tasks: to_cleanup, lookup_tables_keys: old_alts, compressed: strategy.compressed, }Given the new
TaskStrategistError::InconsistentTaskCompressionand the fact thatTransactionStrategyalready carries acompressedflag, mixed compressed/non‑compressed task sets should now be rejected earlier. Recomputing compression with.any(|task.is_compressed())plus a TODO is both redundant and potentially misleading.I’d suggest:
- Replace the
.any(|task.is_compressed())withstrategy.compressed, and- Remove or reword the TODO to reflect that inconsistent compression is now handled by
TaskStrategist.- let commit_ids = self - .task_info_fetcher - .fetch_next_commit_ids( - committed_pubkeys, - // TODO(dode): Handle cases where some tasks are compressed and some are not - strategy - .optimized_tasks - .iter() - .any(|task| task.is_compressed()), - ) + let commit_ids = self + .task_info_fetcher + .fetch_next_commit_ids(committed_pubkeys, strategy.compressed)Same comment applies to other places where you construct new
TransactionStrategyvalues: keepingcompressed: strategy.compressedis good; relying on the field everywhere will make the invariant clearer.Also applies to: 423-427, 457-482
♻️ Duplicate comments (2)
magicblock-api/src/magic_validator.rs (1)
402-414: Hard‑coded compression endpoint now diverges from configured Photon URL
init_chainlinkstill pushes anEndpoint::Compressionwith a baked‑in"http://127.0.0.1:8784"URL, whileinit_committor_servicenow usescompression_config.photon_urlandcompression_config.api_keyto construct thePhotonIndexer. This creates two problems:
- The Chainlink/remote‑account side may talk to a different Photon instance than the committor if
compression.photon_urlis nothttp://127.0.0.1:8784.- The compression endpoint is still not configurable at all from this code path, and configs that set a remote Photon URL won’t affect Chainlink here.
Recommend:
- Drive the compression endpoint URL (and auth, if needed) from the same
CompressionConfigused for the committor, or from a shared constant derived from config, and- Ensure you don’t end up with divergent Photon endpoints between committor and Chainlink.
This also keeps the earlier concern about the hard‑coded URL applicable.
- // TODO: @@@ HACK, make this configurable - endpoints.push(Endpoint::Compression { - url: "http://127.0.0.1:8784".to_string(), - }); + // TODO: @@@ HACK, make this configurable + endpoints.push(Endpoint::Compression { + url: config.compression.photon_url.clone(), + });test-integration/test-committor-service/tests/common.rs (1)
185-213:create_compressed_commit_taskbuilds placeholder compressed data – ensure usage stays preparator-onlyThis helper intentionally sets
compressed_delegation_record_bytestovec![]and uses defaultCompressedAccountMeta/ValidityProof, relying on later preparation (e.g., via Photon-backedprepare_task) to populate real compressed delegation data. That’s fine for the current tests, but it would be invalid if any path tried to submit these tasks on-chain without running through the preparator first; consider documenting that expectation at the call sites or in the helper’s comment to prevent future misuse.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (2)
Cargo.lockis excluded by!**/*.locktest-integration/Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (35)
Cargo.toml(6 hunks)magicblock-api/Cargo.toml(2 hunks)magicblock-api/src/magic_validator.rs(8 hunks)magicblock-chainlink/src/testing/mod.rs(1 hunks)magicblock-chainlink/tests/07_redeleg_us_same_slot.rs(2 hunks)magicblock-committor-program/src/instruction_builder/close_buffer.rs(2 hunks)magicblock-committor-program/src/instruction_builder/init_buffer.rs(2 hunks)magicblock-committor-program/src/instruction_builder/realloc_buffer.rs(2 hunks)magicblock-committor-program/src/instruction_builder/write_buffer.rs(2 hunks)magicblock-committor-service/Cargo.toml(1 hunks)magicblock-committor-service/src/committor_processor.rs(3 hunks)magicblock-committor-service/src/intent_execution_manager.rs(2 hunks)magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs(4 hunks)magicblock-committor-service/src/intent_executor/error.rs(7 hunks)magicblock-committor-service/src/intent_executor/intent_executor_factory.rs(3 hunks)magicblock-committor-service/src/intent_executor/mod.rs(28 hunks)magicblock-committor-service/src/service.rs(5 hunks)magicblock-committor-service/src/tasks/task_builder.rs(7 hunks)magicblock-committor-service/src/tasks/task_strategist.rs(6 hunks)magicblock-committor-service/src/transaction_preparator/error.rs(2 hunks)magicblock-committor-service/src/types.rs(2 hunks)magicblock-config/src/compression.rs(1 hunks)magicblock-config/src/lib.rs(8 hunks)magicblock-config/tests/fixtures/11_everything-defined.toml(1 hunks)magicblock-config/tests/parse_config.rs(3 hunks)magicblock-config/tests/read_config.rs(4 hunks)programs/magicblock/src/magicblock_processor.rs(4 hunks)programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs(4 hunks)programs/magicblock/src/utils/instruction_utils.rs(4 hunks)test-integration/Cargo.toml(3 hunks)test-integration/schedulecommit/test-scenarios/Cargo.toml(1 hunks)test-integration/schedulecommit/test-scenarios/tests/01_commits.rs(2 hunks)test-integration/test-committor-service/tests/common.rs(7 hunks)test-integration/test-committor-service/tests/test_delivery_preparator.rs(8 hunks)test-integration/test-committor-service/tests/test_intent_executor.rs(16 hunks)
🧰 Additional context used
🧠 Learnings (20)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.681Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
📚 Learning: 2025-11-19T09:34:37.890Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.890Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rstest-integration/test-committor-service/tests/common.rsmagicblock-api/src/magic_validator.rstest-integration/schedulecommit/test-scenarios/tests/01_commits.rstest-integration/test-committor-service/tests/test_delivery_preparator.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsmagicblock-api/src/magic_validator.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-service/src/types.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsmagicblock-committor-service/src/tasks/task_builder.rstest-integration/test-committor-service/tests/common.rsCargo.tomlprograms/magicblock/src/mutate_accounts/process_mutate_accounts.rsmagicblock-api/src/magic_validator.rstest-integration/schedulecommit/test-scenarios/tests/01_commits.rsmagicblock-api/Cargo.tomltest-integration/test-committor-service/tests/test_delivery_preparator.rsmagicblock-config/tests/parse_config.rsmagicblock-chainlink/src/testing/mod.rsprograms/magicblock/src/utils/instruction_utils.rs
📚 Learning: 2025-11-07T13:09:52.253Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: test-kit/src/lib.rs:275-0
Timestamp: 2025-11-07T13:09:52.253Z
Learning: In test-kit, the transaction scheduler in ExecutionTestEnv is not expected to shut down during tests. Therefore, using `.unwrap()` in test helper methods like `schedule_transaction` is acceptable and will not cause issues in the test environment.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rstest-integration/schedulecommit/test-scenarios/tests/01_commits.rstest-integration/test-committor-service/tests/test_intent_executor.rstest-integration/test-committor-service/tests/test_delivery_preparator.rsprograms/magicblock/src/magicblock_processor.rs
📚 Learning: 2025-10-28T13:15:42.706Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 596
File: magicblock-processor/src/scheduler.rs:1-1
Timestamp: 2025-10-28T13:15:42.706Z
Learning: In magicblock-processor, transaction indexes were always set to 0 even before the changes in PR #596. The proper transaction indexing within slots will be addressed during the planned ledger rewrite.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsCargo.tomlmagicblock-chainlink/src/testing/mod.rs
📚 Learning: 2025-10-21T13:06:38.900Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/http/get_program_accounts.rs:17-25
Timestamp: 2025-10-21T13:06:38.900Z
Learning: The magicblock validator does not support ledger forking, so commitment-based state queries (processed/confirmed/finalized) are not applicable. RPC methods can safely ignore commitment and minContextSlot parameters from Solana RPC config objects.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsCargo.tomlmagicblock-config/tests/parse_config.rsmagicblock-chainlink/src/testing/mod.rs
📚 Learning: 2025-10-26T16:54:39.084Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 587
File: test-manual/Cargo.toml:0-0
Timestamp: 2025-10-26T16:54:39.084Z
Learning: In the magicblock-validator repository, use git branch references (not commit hashes or tags) for the helius-laserstream dependency to allow automatic updates when the branch is pushed to.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsCargo.tomlmagicblock-chainlink/src/testing/mod.rs
📚 Learning: 2025-11-04T13:22:38.811Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-aperture/src/requests/http/get_fee_for_message.rs:25-31
Timestamp: 2025-11-04T13:22:38.811Z
Learning: In magicblock-aperture, request size validation (including limits on message size) is enforced at the request ingestion layer, before handlers like get_fee_for_message are invoked. Therefore, handlers do not need additional size guards on decoded data.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsCargo.tomlmagicblock-chainlink/src/testing/mod.rs
📚 Learning: 2025-10-21T11:00:18.396Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/encoder.rs:176-187
Timestamp: 2025-10-21T11:00:18.396Z
Learning: In the magicblock validator, the current slot is always the root slot. The SlotEncoder in magicblock-aperture/src/encoder.rs correctly sets `root: slot` because there is no lag between current and root slots in this architecture.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rsmagicblock-chainlink/src/testing/mod.rs
📚 Learning: 2025-11-04T10:48:00.070Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/mod.rs:217-219
Timestamp: 2025-11-04T10:48:00.070Z
Learning: In magicblock-validator, the codebase uses a pattern where types containing non-Send/non-Sync fields (like Rc<RefCell<...>>) are marked with unsafe impl Send when they are guaranteed to be confined to a single thread through careful API design and thread spawning patterns.
Applied to files:
magicblock-committor-program/src/instruction_builder/realloc_buffer.rsmagicblock-committor-program/src/instruction_builder/write_buffer.rsmagicblock-committor-program/src/instruction_builder/close_buffer.rsmagicblock-committor-program/src/instruction_builder/init_buffer.rs
📚 Learning: 2025-11-12T09:46:27.553Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 614
File: magicblock-task-scheduler/src/db.rs:26-0
Timestamp: 2025-11-12T09:46:27.553Z
Learning: In magicblock-task-scheduler, task parameter validation (including ensuring iterations > 0 and enforcing minimum execution intervals) is performed in the Magic program (on-chain) before ScheduleTaskRequest instances reach the scheduler service. The From<&ScheduleTaskRequest> conversion in db.rs does not need additional validation because inputs are already validated at the program level.
Applied to files:
magicblock-committor-service/src/types.rs
📚 Learning: 2025-11-18T08:47:39.681Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.681Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-chainlink/tests/07_redeleg_us_same_slot.rstest-integration/test-committor-service/tests/common.rsprograms/magicblock/src/mutate_accounts/process_mutate_accounts.rsmagicblock-api/src/magic_validator.rstest-integration/test-committor-service/tests/test_delivery_preparator.rsprograms/magicblock/src/utils/instruction_utils.rs
📚 Learning: 2025-10-21T10:34:59.140Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-accounts-db/src/lib.rs:63-72
Timestamp: 2025-10-21T10:34:59.140Z
Learning: In magicblock-validator, the AccountsDb "stop-the-world" synchronizer is managed at the processor/executor level, not at the AccountsDb API level. Transaction executors in magicblock-processor hold a read lock (sync.read()) for the duration of each slot and release it only at slot boundaries, ensuring all account writes happen under the read lock. Snapshot operations acquire a write lock, blocking until all executors release their read locks. This pattern ensures mutual exclusion between writes and snapshots without requiring read guards in AccountsDb write APIs.
Applied to files:
magicblock-chainlink/tests/07_redeleg_us_same_slot.rsmagicblock-chainlink/src/testing/mod.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
test-integration/test-committor-service/tests/common.rstest-integration/schedulecommit/test-scenarios/tests/01_commits.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs
📚 Learning: 2025-10-26T16:53:29.820Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 587
File: magicblock-chainlink/src/remote_account_provider/mod.rs:134-0
Timestamp: 2025-10-26T16:53:29.820Z
Learning: In magicblock-chainlink/src/remote_account_provider/mod.rs, the `Endpoint::separate_pubsub_url_and_api_key()` method uses `split_once("?api-key=")` because the api-key parameter is always the only query parameter right after `?`. No additional query parameter parsing is needed for this use case.
Applied to files:
magicblock-api/src/magic_validator.rs
📚 Learning: 2025-11-04T10:53:50.922Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/locks.rs:110-122
Timestamp: 2025-11-04T10:53:50.922Z
Learning: In magicblock-processor, the TransactionScheduler runs in a single, dedicated thread and will always remain single-threaded. The `next_transaction_id()` function in scheduler/locks.rs uses `unsafe static mut` which is safe given this architectural guarantee.
Applied to files:
magicblock-committor-service/src/committor_processor.rs
📚 Learning: 2025-11-13T09:38:43.804Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/locks.rs:64-102
Timestamp: 2025-11-13T09:38:43.804Z
Learning: In magicblock-processor's TransactionScheduler (scheduler/mod.rs line 59), the executor count is clamped to MAX_SVM_EXECUTORS (63) at initialization time, and executor IDs are assigned sequentially from 0 to count-1. This architectural guarantee ensures that executor IDs used in the bitmask-based AccountLock (scheduler/locks.rs) will always be within valid bounds for bit shifting operations, making runtime bounds checks unnecessary.
Applied to files:
magicblock-committor-service/src/committor_processor.rs
📚 Learning: 2025-10-26T08:49:31.543Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 585
File: magicblock-committor-service/src/tasks/buffer_task.rs:111-115
Timestamp: 2025-10-26T08:49:31.543Z
Learning: In the magicblock-committor-service, compute units returned by the `compute_units()` method in task implementations (such as `BufferTask`, `ArgsTask`, etc.) represent the compute budget for a single task. Transactions can comprise multiple tasks, and the total compute budget for a transaction is computed as the sum of the compute units of all tasks included in that transaction.
Applied to files:
magicblock-committor-service/src/intent_executor/mod.rs
🧬 Code graph analysis (17)
magicblock-committor-service/src/intent_execution_manager.rs (2)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (3)
Self(298-306)Self(308-316)new(63-75)magicblock-committor-service/src/intent_executor/mod.rs (1)
new(111-125)
magicblock-committor-service/src/types.rs (3)
magicblock-committor-service/src/intent_executor/error.rs (1)
value(107-114)magicblock-committor-service/src/intent_executor/mod.rs (1)
value(76-84)magicblock-metrics/src/metrics/types.rs (2)
value(73-73)value(81-86)
magicblock-config/src/lib.rs (1)
magicblock-config/src/compression.rs (1)
default(24-29)
magicblock-chainlink/tests/07_redeleg_us_same_slot.rs (2)
magicblock-chainlink/src/testing/rpc_client_mock.rs (2)
accounts(95-101)slot(66-72)magicblock-chainlink/tests/utils/accounts.rs (2)
compressed_account_shared_with_owner_and_slot(22-47)account_shared_with_owner_and_slot(12-20)
magicblock-committor-service/src/tasks/task_builder.rs (6)
magicblock-committor-service/src/tasks/buffer_task.rs (5)
instruction(68-89)new(38-46)get_compressed_data(157-161)task_type(125-129)is_compressed(146-148)magicblock-committor-service/src/tasks/args_task.rs (6)
instruction(51-152)new(42-47)get_compressed_data(267-280)task_type(216-228)is_compressed(243-250)from(36-38)magicblock-committor-service/src/tasks/mod.rs (5)
instruction(68-68)instruction(304-310)get_compressed_data(105-105)task_type(93-93)is_compressed(99-99)magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(10-22)magicblock-committor-service/src/intent_executor/mod.rs (1)
new(111-125)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (1)
new(63-75)
magicblock-committor-service/src/intent_executor/error.rs (5)
magicblock-rpc-client/src/lib.rs (3)
error(193-195)from(228-230)signature(82-90)magicblock-committor-service/src/transaction_preparator/error.rs (1)
from(21-29)magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (2)
from(542-546)try_map(333-349)magicblock-metrics/src/metrics/types.rs (2)
value(73-73)value(81-86)magicblock-rpc-client/src/utils.rs (1)
try_map(84-88)
test-integration/test-committor-service/tests/common.rs (5)
magicblock-committor-service/src/intent_executor/mod.rs (1)
new(111-125)magicblock-committor-service/src/intent_execution_manager.rs (1)
new(36-72)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (1)
new(63-75)test-integration/test-committor-service/tests/test_delivery_preparator.rs (15)
None(53-53)None(54-54)None(126-126)None(127-127)None(210-210)None(211-211)None(251-251)None(252-252)None(292-292)None(293-293)None(361-361)None(362-362)None(462-462)None(463-463)None(548-548)test-integration/test-committor-service/tests/test_intent_executor.rs (1)
None(148-148)
programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs (1)
magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
compressed(154-156)
magicblock-committor-service/src/intent_execution_manager/intent_execution_engine.rs (2)
magicblock-metrics/src/metrics/mod.rs (2)
observe_committor_intent_execution_time_histogram(383-391)inc_committor_failed_intents_count(370-377)programs/magicblock/src/magic_scheduled_base_intent.rs (2)
get_committed_pubkeys(97-99)get_committed_pubkeys(198-202)
magicblock-api/src/magic_validator.rs (2)
magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(27-29)magicblock-chainlink/src/remote_account_provider/mod.rs (1)
new(212-256)
magicblock-committor-service/src/intent_executor/mod.rs (3)
magicblock-rpc-client/src/utils.rs (5)
decide_rpc_error_flow(176-201)map_magicblock_client_error(93-141)send_transaction_with_retries(34-71)map(13-13)decide_flow(14-16)magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (7)
Self(298-306)Self(308-316)new(63-75)pubkeys(159-170)pubkeys(216-219)fetch_rent_reimbursements(39-42)fetch_rent_reimbursements(338-353)magicblock-committor-service/src/tasks/task_builder.rs (7)
commit_tasks(49-54)commit_tasks(71-165)finalize_task(174-191)accounts(106-109)accounts(313-316)finalize_tasks(57-61)finalize_tasks(168-378)
test-integration/schedulecommit/test-scenarios/tests/01_commits.rs (3)
test-integration/programs/schedulecommit/src/api.rs (4)
delegate_account_cpi_instruction(58-91)init_account_instruction(15-33)pda_and_bump(263-267)schedule_commit_cpi_instruction(104-122)test-integration/test-tools/src/integration_test_context.rs (3)
signature(63-65)try_new(113-115)airdrop_chain(483-494)test-integration/schedulecommit/test-scenarios/tests/utils/mod.rs (1)
extract_transaction_error(255-264)
magicblock-config/src/compression.rs (2)
magicblock-config-macro/src/lib.rs (2)
clap_from_serde(153-163)clap_prefix(44-58)magicblock-config/src/lib.rs (3)
test_merge_with_default(203-291)test_merge_default_with_non_default(294-381)test_merge_non_default(384-552)
magicblock-committor-service/src/tasks/task_strategist.rs (2)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (2)
Self(298-306)Self(308-316)magicblock-chainlink/src/remote_account_provider/remote_account.rs (1)
compressed(154-156)
test-integration/test-committor-service/tests/test_intent_executor.rs (3)
magicblock-committor-service/src/intent_executor/mod.rs (1)
new(111-125)magicblock-committor-service/src/intent_execution_manager.rs (1)
new(36-72)magicblock-committor-service/src/tasks/task_builder.rs (2)
finalize_tasks(57-61)finalize_tasks(168-378)
test-integration/test-committor-service/tests/test_delivery_preparator.rs (3)
test-integration/test-committor-service/tests/common.rs (4)
create_commit_task(167-183)create_compressed_commit_task(186-213)generate_random_bytes(159-164)new(63-66)test-integration/test-committor-service/tests/utils/transactions.rs (1)
init_and_delegate_compressed_account_on_chain(233-322)magicblock-committor-service/src/tasks/mod.rs (5)
cleanup_task(289-294)buffer_pda(280-287)buffer_pda(331-338)chunks_pda(271-278)chunks_pda(322-329)
programs/magicblock/src/magicblock_processor.rs (1)
programs/magicblock/src/schedule_transactions/process_schedule_commit.rs (2)
process_schedule_commit(34-40)process_schedule_compressed_commit(42-48)
| let compressed_account = compressed_account_shared_with_owner_and_slot( | ||
| pubkey, | ||
| ctx.validator_pubkey, | ||
| program_pubkey, | ||
| slot, | ||
| ); | ||
| photon_client.add_account( | ||
| pubkey, | ||
| compressed_account.clone().into(), | ||
| slot, | ||
| ); |
There was a problem hiding this comment.
Also set remote_slot on compressed accounts before adding to photon_client
In both the seeding and redelegation steps, the compressed delegation record’s delegation_slot is set via compressed_account_shared_with_owner_and_slot(...), but the AccountSharedData.remote_slot is left at its default. Per prior compressed-account tests, both fields should advance together so slot-based logic sees the update correctly. Based on learnings
- let compressed_account = compressed_account_shared_with_owner_and_slot(
+ let mut compressed_account = compressed_account_shared_with_owner_and_slot(
pubkey,
ctx.validator_pubkey,
program_pubkey,
slot,
);
- photon_client.add_account(
+ compressed_account.set_remote_slot(slot);
+
+ photon_client.add_account(
pubkey,
compressed_account.clone().into(),
slot,
);
@@
- let compressed_account = compressed_account_shared_with_owner_and_slot(
+ let mut compressed_account = compressed_account_shared_with_owner_and_slot(
pubkey,
ctx.validator_pubkey,
program_pubkey,
slot,
);
- photon_client.add_account(
+ compressed_account.set_remote_slot(slot);
+
+ photon_client.add_account(
pubkey,
compressed_account.clone().into(),
slot,
);Also applies to: 178-188
🤖 Prompt for AI Agents
In magicblock-chainlink/tests/07_redeleg_us_same_slot.rs around lines 141-151
(and also apply the same change at 178-188), the compressed account is created
with the delegation_slot set but AccountSharedData.remote_slot is left at its
default; before calling photon_client.add_account, set the compressed account's
remote_slot to the same slot (i.e., assign remote_slot = slot or call the
account setter that updates remote_slot) so the remote_slot advances in tandem
with delegation_slot; do this in both the initial seeding block and the
redelegation block so slot-based logic observes the update correctly.
| @@ -1,3 +1,4 @@ | |||
| use borsh::BorshSerialize; | |||
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Write instruction serialization is correct; consider a small helper to dedupe pattern
The move to Instruction::new_with_bytes using ix.try_to_vec().expect(...) is consistent with the other builders and keeps semantics equivalent to the prior new_with_borsh path.
Given create_write_ix, create_close_ix, create_realloc_buffer_ix, and create_init_ix now all repeat the same new_with_bytes + try_to_vec().expect(...) sequence, you might consider a tiny helper like:
fn build_ix(
program_id: Pubkey,
ix: &impl BorshSerialize,
accounts: Vec<AccountMeta>,
) -> Instruction {
Instruction::new_with_bytes(
program_id,
&ix.try_to_vec()
.expect("Serialization of instruction should never fail"),
accounts,
)
}to reduce duplication and keep the panic message / serialization behavior centralized.
Also applies to: 51-56
🤖 Prompt for AI Agents
In magicblock-committor-program/src/instruction_builder/write_buffer.rs around
lines 1 and 51-56, multiple functions repeat the sequence
Instruction::new_with_bytes(..., &ix.try_to_vec().expect("Serialization of
instruction should never fail"), ...); extract that pattern into a small helper
(e.g., build_ix) that takes program_id, a reference to a BorshSerialize
instruction, and accounts, calls try_to_vec().expect(...) once, and returns
Instruction::new_with_bytes(...). Replace the repeated sequences in
create_write_ix, create_close_ix, create_realloc_buffer_ix, and create_init_ix
with calls to this helper so serialization behavior and panic message are
centralized.
| executable: false, | ||
| data, | ||
| rent_epoch: u64::MAX, | ||
| rent_epoch: 0, |
There was a problem hiding this comment.
🧩 Analysis chain
Verify authority account rent_epoch inconsistency.
The authority account assertion expects rent_epoch: 0, but other tests in this file expect rent_epoch: u64::MAX for the authority account (lines 440 and 551). Since the authority account is not being modified, its rent_epoch should remain at the value set by ensure_started_validator.
Run the following script to check the initialization behavior:
🏁 Script executed:
#!/bin/bash
# Description: Check ensure_started_validator and how it initializes authority rent_epoch
# Search for ensure_started_validator implementation
ast-grep --pattern 'fn ensure_started_validator($$$) {
$$$
}'
# Search for authority account setup patterns
rg -n -A5 'ensure_started_validator'Length of output: 8922
🏁 Script executed:
#!/bin/bash
# Check AccountSharedData::new implementation and default rent_epoch
rg -n -B2 -A10 'impl.*AccountSharedData' --type=rs
# Look at the specific assertions in the test file
rg -n -B5 -A5 'rent_epoch' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rsLength of output: 4779
🏁 Script executed:
#!/bin/bash
# Find test function definitions and their scope
rg -n 'fn test_' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs
# Look at the full test function containing line 360
sed -n '300,385p' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs
# Look at the full test function containing line 440
sed -n '385,475p' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs
# Look for any authority rent_epoch modifications
rg -n 'validator_authority_id\|authority.*rent_epoch' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rsLength of output: 5635
🏁 Script executed:
#!/bin/bash
# Get the full third test
sed -n '478,620p' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs
# Check if there's any code that sets authority rent_epoch to u64::MAX
rg -n 'set_rent_epoch.*MAX\|rent_epoch.*=.*MAX' programs/magicblock/src/mutate_accounts/Length of output: 4581
🏁 Script executed:
#!/bin/bash
# Search for where rent_epoch might be modified to u64::MAX in the processor
rg -n 'u64::MAX' programs/magicblock/src/mutate_accounts/ -A2 -B2
# Look for process_instruction implementation
rg -n 'fn process_instruction' programs/magicblock/src/test_utils/mod.rs -A50 | head -100Length of output: 5620
🏁 Script executed:
#!/bin/bash
# Search for the actual instruction processor implementation
find programs/magicblock/src -name "*.rs" -type f | xargs grep -l "process_mutate_accounts" | grep -v test
# Look for the mutate accounts processor code (not tests)
rg -n 'pub fn process_mutate_accounts' programs/magicblock/src/ -A 100 | head -150Length of output: 161
🏁 Script executed:
#!/bin/bash
# Look at mod.rs for processor implementation
cat programs/magicblock/src/mutate_accounts/mod.rs
# Also check the file structure - get line counts
wc -l programs/magicblock/src/mutate_accounts/process_mutate_accounts.rsLength of output: 282
🏁 Script executed:
#!/bin/bash
# View the processor implementation (first 300 lines)
sed -n '1,300p' programs/magicblock/src/mutate_accounts/process_mutate_accounts.rsLength of output: 11114
Fix incorrect authority account rent_epoch assertions in tests.
The authority account is created identically in all three tests via ensure_started_validator() and is never modified by the processor. Therefore, all tests should expect the same rent_epoch value. Test test_mod_all_fields_of_one_account correctly asserts rent_epoch: 0, but test_mod_lamports_of_two_accounts (line 440) and test_mod_different_properties_of_four_accounts (line 551) incorrectly assert rent_epoch: u64::MAX. These should be changed to rent_epoch: 0 to match the actual authority account value set by ensure_started_validator().
🤖 Prompt for AI Agents
In programs/magicblock/src/mutate_accounts/process_mutate_accounts.rs around
lines 360, 440 and 551, the tests incorrectly expect different rent_epoch values
for the authority account; the authority is created by
ensure_started_validator() and never modified, so all tests should assert
rent_epoch: 0. Update the assertions in test_mod_lamports_of_two_accounts
(around line 440) and test_mod_different_properties_of_four_accounts (around
line 551) to use rent_epoch: 0 instead of rent_epoch: u64::MAX so they match
test_mod_all_fields_of_one_account and the actual account state.
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
magicblock-chainlink/src/remote_account_provider/mod.rs (2)
1235-1246: Use== 0for unsigned integer zero-check in retry macro.The condition
remaining_retries <= 0at line 1239 for au64is misleading since unsigned integers can never be negative. After the decrement, this effectively checks== 0, but the<= 0reads as if negative values are possible.🔎 Suggested fix
macro_rules! retry { ($msg:expr) => {{ trace!($msg); remaining_retries -= 1; - if remaining_retries <= 0 { + if remaining_retries == 0 { let err_msg = format!("Max retries {MAX_RETRIES} reached, giving up on fetching accounts: {pubkeys:?}"); notify_error(&err_msg); return Err(RemoteAccountProviderError::FailedFetchingAccounts(err_msg)); } tokio::time::sleep(Duration::from_millis(400)).await; continue; }}; }
1200-1204: Consider adding explicit invariant comment for Mutex unwrap.Line 1202 uses
.unwrap()onfetching_accounts.lock(). Per coding guidelines, this should have proper error handling or explicit justification.🔎 Suggested documentation
let notify_error = |error_msg: &str| { - let mut fetching = fetching_accounts.lock().unwrap(); + // INVARIANT: Mutex poisoning indicates a thread panicked while holding the lock, + // which is a critical bug. Panicking here is appropriate. + let mut fetching = fetching_accounts.lock().unwrap(); error!("{error_msg}");
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (2)
Cargo.lockis excluded by!**/*.locktest-integration/Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (2)
magicblock-chainlink/src/remote_account_provider/mod.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs
🧰 Additional context used
📓 Path-based instructions (2)
{test-*,tools}/**
⚙️ CodeRabbit configuration file
Usage of
.unwrap()or.expect()in test code is acceptable and may be treated as trivial.
Files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-chainlink/src/remote_account_provider/mod.rs
🧠 Learnings (17)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-12-17T12:46:36.207Z
Learnt from: GabrielePicco
Repo: magicblock-labs/magicblock-validator PR: 738
File: magicblock-chainlink/src/chainlink/fetch_cloner.rs:1109-1196
Timestamp: 2025-12-17T12:46:36.207Z
Learning: In magicblock-chainlink/src/chainlink/fetch_cloner.rs, eATA account subscriptions are intentionally NOT cancelled even when the eATA doesn't exist or delegation conversion fails. The subscriptions are kept active to watch for future creation or delegation of eATA accounts (per maintainer GabrielePicco).
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-10-26T16:53:29.820Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 587
File: magicblock-chainlink/src/remote_account_provider/mod.rs:134-0
Timestamp: 2025-10-26T16:53:29.820Z
Learning: In magicblock-chainlink/src/remote_account_provider/mod.rs, the `Endpoint::separate_pubsub_url_and_api_key()` method uses `split_once("?api-key=")` because the api-key parameter is always the only query parameter right after `?`. No additional query parameter parsing is needed for this use case.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-21T11:03:26.756Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 664
File: magicblock-chainlink/src/testing/mod.rs:342-370
Timestamp: 2025-11-21T11:03:26.756Z
Learning: In the magicblock-validator codebase, avoid leaving review comments that merely acknowledge code is correct or well-structured when there is no actionable suggestion, improvement, or issue to flag. Only comment when there is something specific to recommend, fix, or clarify.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-26T13:27:46.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-21T13:56:03.885Z
Learnt from: snawaz
Repo: magicblock-labs/magicblock-validator PR: 575
File: test-integration/programs/schedulecommit/src/utils/mod.rs:71-71
Timestamp: 2025-11-21T13:56:03.885Z
Learning: In Solana programs (BPF runtime), defensive checked conversions to `usize` (e.g., `try_into()`) are unnecessary when casting from `u64`. Solana exclusively runs in 64-bit environments where `usize` is 64 bits, so truncation is not a concern and simple `as usize` casts are appropriate.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
🔇 Additional comments (4)
magicblock-chainlink/src/remote_account_provider/mod.rs (3)
982-1054: LGTM! Well-structured compressed account fetch with retry logic.The async closure cleanly handles the compressed account fetch path:
- Correctly checks for RPC accounts owned by compressed delegation program
- Early returns
Ok(None)when no compressed accounts need fetching- Properly validates photon_client availability
- Implements retry loop with reasonable backoff (400ms)
The retry logic waiting for Photon to index compressed accounts aligns with the project's approach where PhotonIndexer has built-in retry logic. Based on learnings, this pattern is intentional.
1450-1530: LGTM! Consolidation logic handles all edge cases correctly.The consolidation function properly:
- Returns early for single-result cases
- Prefers compressed accounts when both RPC and Photon find an account (with appropriate warning)
- Falls back to RPC accounts on length mismatch
- Returns empty vec for unexpected >2 results (acceptable per maintainer feedback)
The
debug_assert!statements at lines 1488-1491 provide development-time validation without impacting release builds. Based on learnings, the error logging + empty vec return for >2 results is the preferred approach during development.
2245-2404: LGTM! Comprehensive test coverage for compressed account scenarios.The new tests properly validate:
- Multiple photon-only accounts (
test_multiple_photon_accounts)- Mixed RPC + compressed accounts (
test_multiple_mixed_accounts)- Missing accounts in both paths (
test_multiple_mixed_accounts_some_missing)The assertion macros (
assert_compressed_account!,assert_regular_account!) make the tests readable and maintainable. The tests correctly verifyRemoteAccountUpdateSource::CompressedvsFetchtagging.test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs (1)
115-115: [No changes needed.AccountFetchOrigin::GetMultipleAccountsis the correct and consistent pattern across all compressed delegation tests in the codebase.]Likely an incorrect or invalid review comment.
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (3)
magicblock-chainlink/tests/05_redeleg_other_same_slot.rs (1)
94-94: Typo in comment.- // Then immediateljky delegate to other authority (simulating same slot operation) + // Then immediately delegate to other authority (simulating same slot operation)magicblock-chainlink/src/remote_account_provider/mod.rs (1)
948-1199: Compressed metrics count all fetches when Photon is available, even if no compressed data was fetchedThe fetch pipeline logic is correct, but the compressed metrics are misleading. When no accounts are owned by the compressed delegation program (
compressed_accounts_count == 0), the closure returnsOk(None)and skips thefetch_from_photoncall entirely. However, the metrics block still executes because it gates only on(*photon_client).is_some(), not on whether an actual compressed fetch occurred:if (*photon_client).is_some() { inc_compressed_account_fetches_success(pubkeys.len() as u64); // Always increments, even if no fetch happened inc_compressed_account_fetches_found(fetch_origin, compressed_found_count); inc_compressed_account_fetches_not_found(fetch_origin, compressed_not_found_count); }This means every fetch in a Photon-enabled setup counts toward compressed metrics, regardless of whether a Photon lookup was actually needed or performed. To reflect real Photon usage, gate the metrics block on
compressed_found_count + compressed_not_found_count > 0or on the presence of aFetchedRemoteAccounts::Compressedentry instead.magicblock-chainlink/src/chainlink/fetch_cloner.rs (1)
1004-1045: Silently dropped compressed delegation records lack error visibility to callersThe bulk path silently omits accounts with unparseable compressed delegation records via
filter_map, logging only an error:let Ok(delegation_record) = CompressedDelegationRecord::try_from_slice(account.data()) … else { return None; };This differs from the non-compressed path (lines 1067–1091), where invalid records trigger
ChainlinkError::InvalidDelegationRecordand abort the batch. The compressed path provides no way for callers to distinguish between accounts legitimately not found on-chain versus accounts found but unparseable—neither the account nor the error is recorded inFetchAndCloneResult. While subscriptions are correctly cancelled (since dropped accounts don't enteraccounts_to_clone), the silent skip hides data integrity issues.Consider either treating parse failures as hard errors or adding an
invalid_recordsfield toFetchAndCloneResultso callers can detect and react to malformed compressed delegation records.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (11)
magicblock-chainlink/src/chainlink/fetch_cloner.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/tests/04_redeleg_other_separate_slots.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-magic-program-api/src/instruction.rstest-integration/configs/committor-conf.devnet.tomltest-integration/programs/schedulecommit/src/lib.rstest-integration/programs/schedulecommit/src/magicblock_program.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rstest-integration/test-tools/src/integration_test_context.rs
💤 Files with no reviewable changes (2)
- test-integration/programs/schedulecommit/src/lib.rs
- test-integration/programs/schedulecommit/src/magicblock_program.rs
🧰 Additional context used
📓 Path-based instructions (2)
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rsmagicblock-magic-program-api/src/instruction.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
{test-*,tools}/**
⚙️ CodeRabbit configuration file
Usage of
.unwrap()or.expect()in test code is acceptable and may be treated as trivial.
Files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rstest-integration/test-tools/src/integration_test_context.rstest-integration/configs/committor-conf.devnet.toml
🧠 Learnings (19)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-magic-program-api/src/instruction.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-magic-program-api/src/instruction.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-10-21T11:00:18.396Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/encoder.rs:176-187
Timestamp: 2025-10-21T11:00:18.396Z
Learning: In the magicblock validator, the current slot is always the root slot. The SlotEncoder in magicblock-aperture/src/encoder.rs correctly sets `root: slot` because there is no lag between current and root slots in this architecture.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rsmagicblock-chainlink/tests/01_ensure-accounts.rs
📚 Learning: 2025-11-26T13:27:46.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-12-17T12:46:36.207Z
Learnt from: GabrielePicco
Repo: magicblock-labs/magicblock-validator PR: 738
File: magicblock-chainlink/src/chainlink/fetch_cloner.rs:1109-1196
Timestamp: 2025-12-17T12:46:36.207Z
Learning: In magicblock-chainlink/src/chainlink/fetch_cloner.rs, eATA account subscriptions are intentionally NOT cancelled even when the eATA doesn't exist or delegation conversion fails. The subscriptions are kept active to watch for future creation or delegation of eATA accounts (per maintainer GabrielePicco).
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rstest-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/tests/05_redeleg_other_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rsmagicblock-chainlink/src/remote_account_provider/mod.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs
📚 Learning: 2025-11-07T13:09:52.253Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: test-kit/src/lib.rs:275-0
Timestamp: 2025-11-07T13:09:52.253Z
Learning: In test-kit, the transaction scheduler in ExecutionTestEnv is not expected to shut down during tests. Therefore, using `.unwrap()` in test helper methods like `schedule_transaction` is acceptable and will not cause issues in the test environment.
Applied to files:
test-integration/test-tools/src/integration_test_context.rsmagicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-10-26T16:53:29.820Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 587
File: magicblock-chainlink/src/remote_account_provider/mod.rs:134-0
Timestamp: 2025-10-26T16:53:29.820Z
Learning: In magicblock-chainlink/src/remote_account_provider/mod.rs, the `Endpoint::separate_pubsub_url_and_api_key()` method uses `split_once("?api-key=")` because the api-key parameter is always the only query parameter right after `?`. No additional query parameter parsing is needed for this use case.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-21T13:56:03.885Z
Learnt from: snawaz
Repo: magicblock-labs/magicblock-validator PR: 575
File: test-integration/programs/schedulecommit/src/utils/mod.rs:71-71
Timestamp: 2025-11-21T13:56:03.885Z
Learning: In Solana programs (BPF runtime), defensive checked conversions to `usize` (e.g., `try_into()`) are unnecessary when casting from `u64`. Solana exclusively runs in 64-bit environments where `usize` is 64 bits, so truncation is not a concern and simple `as usize` casts are appropriate.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
📚 Learning: 2025-11-04T10:48:00.070Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/mod.rs:217-219
Timestamp: 2025-11-04T10:48:00.070Z
Learning: In magicblock-validator, the codebase uses a pattern where types containing non-Send/non-Sync fields (like Rc<RefCell<...>>) are marked with unsafe impl Send when they are guaranteed to be confined to a single thread through careful API design and thread spawning patterns.
Applied to files:
magicblock-chainlink/src/remote_account_provider/mod.rs
🧬 Code graph analysis (3)
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs (3)
test-integration/test-chainlink/src/ixtest_context.rs (1)
counter_pda(203-205)test-integration/test-chainlink/src/lib.rs (1)
sleep_ms(8-11)compressed-delegation-client/src/lib.rs (1)
id(61-63)
magicblock-chainlink/tests/05_redeleg_other_same_slot.rs (4)
magicblock-chainlink/src/testing/accounts.rs (1)
account_shared_with_owner(4-10)magicblock-chainlink/src/testing/deleg.rs (1)
add_delegation_record_for(22-46)magicblock-chainlink/src/testing/utils.rs (1)
init_logger(58-66)magicblock-chainlink/tests/utils/accounts.rs (3)
account_shared_with_owner_and_slot(11-19)compressed_account_shared_with_owner_and_slot(21-48)default(58-64)
magicblock-chainlink/tests/01_ensure-accounts.rs (1)
magicblock-chainlink/tests/utils/accounts.rs (1)
compressed_account_shared_with_owner_and_slot(21-48)
🔇 Additional comments (19)
test-integration/test-tools/src/integration_test_context.rs (1)
729-738: LGTM! Improved test diagnostics.The addition of a custom failure message to the
assert!macro improves debuggability by displaying the actual result when the assertion fails. The underlying logic remains correct and unchanged.magicblock-chainlink/tests/05_redeleg_other_same_slot.rs (3)
11-25: LGTM!Import style is now consistent, and the new compressed account utilities are properly imported alongside the existing helpers.
126-156: LGTM!The compressed account initialization with the explanatory comment and the assertion pattern (only checking pubkey, not delegation record) correctly align with the compressed delegation design where delegation records are never subscribed. Based on learnings.
158-205: LGTM!The compressed redelegation flow correctly:
- Forces undelegation and triggers subscription
- Updates Photon with new delegation to
other_authority- Simulates the subscription update via
send_and_receive_account_update- Verifies the account is marked as undelegated (since delegated to another authority)
The assertion with timeout (line 198) properly guards against flaky test failures.
test-integration/test-chainlink/tests/ix_07_redeleg_us_same_slot.rs (3)
91-106: LGTM! Test setup correctly mirrors non-compressed flow.The test setup properly initializes the context and delegates the counter account using the compressed path. The omission of
deleg_record_pubkey(compared to the non-compressed test at line 29) is correct, as compressed delegation records are not subscribed to by design.
107-133: LGTM! First verification block correctly implements compressed flow.The sleep after
ensure_accounts(line 120) addresses the previous review concern. The subscription check correctly validates only thecounter_pdawithout checking a delegation record, which aligns with the compressed delegation design where delegation records are never subscribed. Based on learnings, compressed delegation records are not tracked via subscription by design.
135-157: LGTM! Second verification block correctly validates same-slot redelegate.The undelegate/redelegate flow properly uses the compressed-specific methods and maintains consistent assertions with the first verification block. The sleep after undelegation allows for pubsub updates to propagate before re-verification.
test-integration/configs/committor-conf.devnet.toml (1)
32-34: Configuration is correct. The compressed delegation program binary exists at the specified path (../programs/compressed_delegation/compressed_delegation.sorelative to the config file), and the program IDDEL2rPzhFaS5qzo8XY9ZNxSzuunWueySq3p2dxJfwPbTis consistently referenced throughout the codebase in compressed-delegation-client.magicblock-chainlink/tests/01_ensure-accounts.rs (2)
20-20: LGTM!The import of
compressed_account_shared_with_owner_and_slotis necessary for the new compressed delegation tests and follows the existing import pattern.
465-494: LGTM!The test correctly validates that an undelegated compressed account is cloned as undelegated and subscribed, following the expected behavior for accounts without delegation.
magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs (4)
21-21: LGTM!The import addition is necessary for the compressed account test variant and follows the existing import patterns.
133-182: LGTM! Correct compressed delegation setup.The test correctly implements the compressed delegation flow:
- Line 165's
set_remote_slot(slot)call is necessary (setsAccountSharedDataslot, distinct from the delegation record slot set by the helper)- Line 181 correctly verifies only the account pubkey is not subscribed, consistent with the design that compressed delegation records are never subscribed to
- The test properly uses
photon_clientfor compressed account storage and verifies the expected cloning behaviorBased on learnings, the implementation correctly distinguishes between the two slot fields and follows compressed delegation subscription patterns.
184-210: LGTM! Undelegation flow correctly implemented.The compressed undelegation sequence correctly mirrors the non-compressed test logic and properly verifies:
- Forced undelegation prevents writes while undelegating
- Commit and undelegation properly updates account state
- Final state correctly shows account cloned as undelegated and subscribed (without delegation record, per compressed flow design)
212-252: LGTM! Redelegation flow correctly adapted for compressed accounts.The compressed redelegation sequence is well-implemented:
- Correctly creates compressed delegation record for
other_authorityviacompressed_account_shared_with_owner_and_slotand stores in photon client- Uses
send_and_receive_account_updateto simulate the delegation update notification (appropriate for compressed flow vs the programmaticdelegate_existing_account_to()used in non-compressed tests)- Line 237's descriptive assertion message is good practice
- Final state verification correctly shows account cloned as undelegated with subscription active (no delegation record subscription per compressed design)
magicblock-magic-program-api/src/instruction.rs (2)
110-140: Compressed schedule instruction variants mirror existing API; watch ABI consumersThese variants mirror
ScheduleCommit/ScheduleCommitAndUndelegatesemantics with compressed-specific docs and account layout, which looks correct. SinceMagicBlockInstructionis serialized (bincode/serde) and shared between on-chain program and off-chain clients, ensure all consumers are upgraded in lockstep so older nodes don’t fail to decode when these new variants appear.
149-172:compressedflags extend account modification structs consistentlyAdding
compressed: Option<bool>alongsidedelegated/confinedkeeps the tri-state “no change vs true vs false” pattern and preserves struct layout. From a type/serde perspective this is fine; just verify producers set it only when compressed delegation semantics actually apply.magicblock-chainlink/src/chainlink/fetch_cloner.rs (2)
417-587: Compressed delegation subscription handling matches bulk semanticsThe new
owned_by_compressed_delegation_programbranch correctly derives state fromCompressedDelegationRecord: it marks accounts as compressed, rewrites lamports/owner/data, and sets the delegated flag based on authority, while returning a synthesizedDelegationRecordsodelegated_to_otherand undelegation checks keep working. The fallback totry_getwhen direct deserialization fails is a reasonable recovery path and keeps subscription and bulk behavior aligned.
1476-1531: Compressed undelegating accounts correctly skip RPC-based refreshThe new
in_bank.compressed()branch inshould_refresh_undelegating_in_bank_accountshort-circuits refresh for compressed undelegations and avoids fetching delegation records for them, which matches the compressed delegation design where the delegation program itself handles ownership reassignment during undelegate, and owner programs only perform local cleanup. This behavior looks correct. Based on learnings, compressed undelegations should not go through the standard refresh path.magicblock-chainlink/src/remote_account_provider/mod.rs (1)
100-141: PhotonClient generic integration in RemoteAccountProvider is coherentThreading
P: PhotonClientthroughRemoteAccountProvider, storingphoton_client: Option<P>, and updatingtry_from_urls_and_config/try_from_clients_and_mode/new/try_new_from_endpointsis consistent. The default impls usePhotonClientImpl, while tests correctly instantiateNone::<PhotonClientImpl>orSome(PhotonClientMock), so type signatures and lifecycle wiring look sound.Also applies to: 162-205, 207-235, 320-377, 379-461
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (1)
programs/elfs/guinea.sois excluded by!**/*.so
📒 Files selected for processing (5)
magicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-committor-service/src/pubkeys_provider.rsmagicblock-committor-service/src/tasks/task_builder.rsprograms/magicblock/src/magic_scheduled_base_intent.rstest-integration/test-committor-service/tests/test_intent_executor.rs
🧰 Additional context used
📓 Path-based instructions (2)
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-committor-service/src/pubkeys_provider.rsmagicblock-chainlink/tests/01_ensure-accounts.rsprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-committor-service/src/tasks/task_builder.rs
{test-*,tools}/**
⚙️ CodeRabbit configuration file
Usage of
.unwrap()or.expect()in test code is acceptable and may be treated as trivial.
Files:
test-integration/test-committor-service/tests/test_intent_executor.rs
🧠 Learnings (17)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-committor-service/src/pubkeys_provider.rsmagicblock-chainlink/tests/01_ensure-accounts.rstest-integration/test-committor-service/tests/test_intent_executor.rsprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
magicblock-committor-service/src/pubkeys_provider.rsmagicblock-chainlink/tests/01_ensure-accounts.rsprograms/magicblock/src/magic_scheduled_base_intent.rs
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rsprograms/magicblock/src/magic_scheduled_base_intent.rs
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rstest-integration/test-committor-service/tests/test_intent_executor.rsprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rsprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rsprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rs
📚 Learning: 2025-12-17T12:46:36.207Z
Learnt from: GabrielePicco
Repo: magicblock-labs/magicblock-validator PR: 738
File: magicblock-chainlink/src/chainlink/fetch_cloner.rs:1109-1196
Timestamp: 2025-12-17T12:46:36.207Z
Learning: In magicblock-chainlink/src/chainlink/fetch_cloner.rs, eATA account subscriptions are intentionally NOT cancelled even when the eATA doesn't exist or delegation conversion fails. The subscriptions are kept active to watch for future creation or delegation of eATA accounts (per maintainer GabrielePicco).
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rs
📚 Learning: 2025-10-21T11:00:18.396Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/encoder.rs:176-187
Timestamp: 2025-10-21T11:00:18.396Z
Learning: In the magicblock validator, the current slot is always the root slot. The SlotEncoder in magicblock-aperture/src/encoder.rs correctly sets `root: slot` because there is no lag between current and root slots in this architecture.
Applied to files:
magicblock-chainlink/tests/01_ensure-accounts.rs
📚 Learning: 2025-11-21T10:22:07.520Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 661
File: magicblock-committor-service/src/intent_executor/single_stage_executor.rs:20-28
Timestamp: 2025-11-21T10:22:07.520Z
Learning: In magicblock-committor-service's SingleStageExecutor and TwoStageExecutor (single_stage_executor.rs and two_stage_executor.rs), the fields transaction_strategy, junk, and patched_errors are intentionally public because these executors are designed to be used independently outside of the IntentExecutor scope, and callers need access to these execution reports for cleanup and error handling.
Applied to files:
test-integration/test-committor-service/tests/test_intent_executor.rsprograms/magicblock/src/magic_scheduled_base_intent.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-07T13:09:52.253Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: test-kit/src/lib.rs:275-0
Timestamp: 2025-11-07T13:09:52.253Z
Learning: In test-kit, the transaction scheduler in ExecutionTestEnv is not expected to shut down during tests. Therefore, using `.unwrap()` in test helper methods like `schedule_transaction` is acceptable and will not cause issues in the test environment.
Applied to files:
test-integration/test-committor-service/tests/test_intent_executor.rs
📚 Learning: 2025-11-21T11:03:26.756Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 664
File: magicblock-chainlink/src/testing/mod.rs:342-370
Timestamp: 2025-11-21T11:03:26.756Z
Learning: In the magicblock-validator codebase, avoid leaving review comments that merely acknowledge code is correct or well-structured when there is no actionable suggestion, improvement, or issue to flag. Only comment when there is something specific to recommend, fix, or clarify.
Applied to files:
programs/magicblock/src/magic_scheduled_base_intent.rs
📚 Learning: 2025-11-20T08:57:07.217Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 650
File: magicblock-chainlink/src/submux/subscription_task.rs:13-99
Timestamp: 2025-11-20T08:57:07.217Z
Learning: In the magicblock-validator repository, avoid posting review comments that merely confirm code is correct or matches intended behavior without providing actionable feedback, suggestions for improvement, or identifying potential issues. Such confirmatory comments are considered unhelpful noise by the maintainers.
Applied to files:
programs/magicblock/src/magic_scheduled_base_intent.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rs
🧬 Code graph analysis (1)
test-integration/test-committor-service/tests/test_intent_executor.rs (3)
magicblock-committor-service/src/intent_executor/mod.rs (1)
new(126-142)test-integration/test-committor-service/tests/test_transaction_preparator.rs (1)
None(62-62)test-integration/test-committor-service/tests/test_delivery_preparator.rs (15)
None(54-54)None(55-55)None(125-125)None(126-126)None(207-207)None(208-208)None(249-249)None(250-250)None(291-291)None(292-292)None(359-359)None(360-360)None(461-461)None(462-462)None(552-552)
🔇 Additional comments (10)
magicblock-chainlink/tests/01_ensure-accounts.rs (2)
367-410: LGTM! Compressed delegation test correctly validates the "delegated to us" scenario.The test properly sets up a compressed delegation record delegated to the validator, verifies it's cloned as delegated with the correct owner, and confirms it's NOT subscribed—aligning with the design that compressed delegation records are never subscribed. The previous review concerns about missing parameters and owner verification have been addressed.
416-494: LGTM! Compressed delegation tests correctly validate "delegated to other" and "undelegated" scenarios.Both tests follow the correct patterns:
test_compressed_delegation_record_delegated_to_otherproperly verifies that an account delegated to another authority is cloned as undelegated and subscribed (with owner verification).test_compressed_account_undelegatedcorrectly tests an empty/undelegated compressed record, expecting undelegated clone and subscription.The previous review concerns about missing parameters and owner verification have been addressed in these tests as well.
magicblock-committor-service/src/pubkeys_provider.rs (3)
54-68: LGTM: Logging maintains observability.The trace logging correctly includes the compressed delegation program alongside other program IDs, maintaining consistency and visibility into which programs are being used.
73-73: LGTM: Correct inclusion in common pubkeys.The compressed delegation program is properly added to the common pubkeys set, ensuring it's available for all process/finalize transactions alongside other program IDs.
48-48: Compressed-delegation-client integration is correctly implemented.The crate is properly configured as a workspace member with valid workspace dependency declarations in
magicblock-committor-service/Cargo.toml. Thecompressed_delegation_client::id()function usage follows the established pattern used for other program IDs (e.g.,dlp::id(),magicblock_committor_program::id()), and the program ID is correctly registered in the common pubkeys set.test-integration/test-committor-service/tests/test_intent_executor.rs (5)
15-15: LGTM! PhotonIndexer properly integrated into test infrastructure.The PhotonIndexer is correctly threaded through the test environment setup:
- Imported from light-client
- Passed to CacheTaskInfoFetcher constructor
- Passed to IntentExecutorImpl constructor
This enables the test infrastructure to support both RPC and Photon-based execution paths.
Also applies to: 88-91, 102-102
138-155: LGTM! Appropriate use ofNonefor photon_client in non-compressed tests.These error recovery tests don't exercise compressed account flows, so passing
Nonefor the photon_client parameter is correct. The tests focus on:
- Commit ID error parsing and recovery
- Undelegation error parsing and recovery
- Action error parsing and recovery
- CPI limit error parsing and recovery
All of which operate on standard delegated accounts, not compressed accounts.
Also applies to: 199-204, 258-264, 320-327
275-280: LGTM! Improved assertion message for better debugging.The enhanced assertion now displays both the actual and expected error messages when the test fails, making it easier to diagnose issues:
assert!( execution_err.to_string().contains(EXPECTED_ERR_MSG), "{} != {}", execution_err.to_string(), EXPECTED_ERR_MSG );This is a useful improvement for test maintainability.
363-367: LGTM! Correct usage offetch_next_commit_idsfor test setup.These calls invalidate the commit nonce cache to force CommitIDError scenarios in the error recovery tests. This is a valid test setup pattern to exercise the error recovery logic.
Also applies to: 551-555, 755-758
979-994: LGTM! TaskBuilderImpl calls correctly updated with photon_client parameter.The helper function properly passes the photon_client parameter to both
commit_tasksandfinalize_tasks:
&None::<Arc<PhotonIndexer>>is passed since these tests don't use compressed accounts- Both async calls are properly awaited and unwrapped
- Aligns with the updated TaskBuilderImpl API
The changes maintain the existing test behavior while supporting the new PhotonIndexer infrastructure.
There was a problem hiding this comment.
Actionable comments posted: 2
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (2)
magicblock-committor-service/src/tasks/task_builder.rstest-integration/test-chainlink/src/ixtest_context.rs
🧰 Additional context used
📓 Path-based instructions (2)
{test-*,tools}/**
⚙️ CodeRabbit configuration file
Usage of
.unwrap()or.expect()in test code is acceptable and may be treated as trivial.
Files:
test-integration/test-chainlink/src/ixtest_context.rs
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-committor-service/src/tasks/task_builder.rs
🧠 Learnings (17)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rsmagicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-10-28T13:15:42.706Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 596
File: magicblock-processor/src/scheduler.rs:1-1
Timestamp: 2025-10-28T13:15:42.706Z
Learning: In magicblock-processor, transaction indexes were always set to 0 even before the changes in PR #596. The proper transaction indexing within slots will be addressed during the planned ledger rewrite.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-11-19T12:55:48.931Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-tools/src/validator.rs:193-197
Timestamp: 2025-11-19T12:55:48.931Z
Learning: In the magicblock-validator codebase, when constructing arguments for the light CLI test validator in start_light_validator_with_config (test-integration/test-tools/src/validator.rs), the shlex::split approach is necessary because the light CLI does not handle whitespaces properly in validator arguments. The string concatenation + shlex parsing pattern should not be refactored to direct argument construction in this specific case.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-11-07T13:09:52.253Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: test-kit/src/lib.rs:275-0
Timestamp: 2025-11-07T13:09:52.253Z
Learning: In test-kit, the transaction scheduler in ExecutionTestEnv is not expected to shut down during tests. Therefore, using `.unwrap()` in test helper methods like `schedule_transaction` is acceptable and will not cause issues in the test environment.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-11-24T14:21:00.996Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: Cargo.toml:58-58
Timestamp: 2025-11-24T14:21:00.996Z
Learning: In the magicblock-validator codebase, magicblock-api/Cargo.toml intentionally uses borsh = "1.5.3" (instead of the workspace version 0.10.4) because it needs to deserialize types from the magic-domain-program external dependency, which requires borsh 1.5.x compatibility. This is an intentional exception for interoperability with the magic domain program.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
test-integration/test-chainlink/src/ixtest_context.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-21T10:22:07.520Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 661
File: magicblock-committor-service/src/intent_executor/single_stage_executor.rs:20-28
Timestamp: 2025-11-21T10:22:07.520Z
Learning: In magicblock-committor-service's SingleStageExecutor and TwoStageExecutor (single_stage_executor.rs and two_stage_executor.rs), the fields transaction_strategy, junk, and patched_errors are intentionally public because these executors are designed to be used independently outside of the IntentExecutor scope, and callers need access to these execution reports for cleanup and error handling.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rs
🧬 Code graph analysis (2)
test-integration/test-chainlink/src/ixtest_context.rs (5)
compressed-delegation-client/src/generated/instructions/undelegate_compressed.rs (2)
instruction(23-28)instruction(135-150)compressed-delegation-client/src/generated/instructions/delegate_compressed.rs (2)
instruction(29-34)instruction(185-209)magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(14-26)magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs (3)
url(63-65)new(53-61)new(391-404)test-integration/programs/flexi-counter/src/state.rs (1)
pda(45-48)
magicblock-committor-service/src/tasks/task_builder.rs (2)
magicblock-committor-service/src/tasks/args_task.rs (3)
instruction(72-192)get_compressed_data(322-335)from(44-46)magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(14-26)
🔇 Additional comments (12)
test-integration/test-chainlink/src/ixtest_context.rs (5)
3-60: LGTM!The imports and type alias changes correctly integrate PhotonIndexer and compressed-delegation types. The
IxtestChainlinktype alias properly includesPhotonClientImplas the new generic parameter.
62-190: LGTM!The
IxtestContextstruct correctly adds thephoton_indexerfield and properly wires it through initialization. TheEndpoint::Compressionconfiguration andPhotonIndexerinstantiation are correctly integrated into all code paths.
773-782: LGTM!The
get_packed_accountshelper correctly encapsulates the common pattern of creatingPackedAccountswith system accounts, reducing duplication across the compressed delegation methods.
784-801: LGTM!The helper correctly abstracts the fetch-account-and-proof pattern. Per coding guidelines,
.unwrap()is acceptable in test code. Based on learnings, PhotonIndexer already has built-in retry logic (~10 attempts), so no additional retry handling is needed.
454-458: Address queue and tree indices are intentionally identical in test setup.Both
address_queue_pubkey_indexandaddress_merkle_tree_pubkey_indexare set to the same value across test utilities, withstate_queue_pubkey_indexcomputed separately. This pattern is consistent in both test-chainlink and test-committor-service test files. If this intentionally reflects your Light Protocol test configuration (where address tree and queue are the same account), it's acceptable—just verify this matches your deployment expectations.magicblock-committor-service/src/tasks/task_builder.rs (7)
4-16: LGTM: Clean imports for compressed account support.The new imports for
light_client,light_sdk, and related types are well-organized and appropriate for the compressed data workflow. The logging import supports proper observability.
40-47: LGTM: CompressedData struct is well-defined.The struct correctly encapsulates all necessary data for compressed operations: hash, serialized record bytes, remaining accounts, metadata, and validity proof. The derives are appropriate for the use case.
49-65: LGTM: Trait signatures correctly extended for compressed support.The addition of optional
PhotonIndexerparameter to bothcommit_tasksandfinalize_tasksenables the compressed workflow while maintaining backward compatibility for non-compressed paths.
191-253: LGTM: Compressed commit path with proper error handling and parallelization.The implementation correctly:
- Validates
photon_clientpresence with structured error (PhotonClientNotFound)- Uses
ok_or(TaskBuilderError::MissingCommitId(...))instead of.expect()for commit_id lookup (previous review addressed)- Parallelizes compressed data fetching using
FuturesUnorderedfor performance- Propagates all errors properly via
?
309-337: LGTM: Efficient parallel compressed data fetching in finalize flow.The
get_compressed_data_for_accountshelper correctly:
- Validates
photon_clientpresence before attempting fetches- Uses
FuturesUnorderedfor concurrent fetching (addressing previous performance suggestion)- Returns
vec![None; ...]for non-compressed path, avoiding unnecessary allocations- Propagates errors properly via
try_collect()
482-521: LGTM: Comprehensive error handling for compressed workflow.The error enum additions properly cover all failure modes in the compressed path:
CompressedDataFetchErrorfor indexer issuesLightSdkErrorfor SDK-level failuresMissingStateTrees,MissingAddress,MissingCompressedDatafor validation failuresPhotonClientNotFoundfor configuration issuesMissingCommitId(Pubkey)for lookup failures with contextThe
signature()method correctly returnsNonefor all new variants since these aren't transaction-related errors.
525-581: LGTM: Robust compressed data fetching with proper error handling.The function correctly addresses all previous review concerns:
- Lines 557-561: Uses
.first().copied().ok_or()to safely extracttree_info, preventing panic on emptypacked_tree_infosvector- Line 577: Removed unnecessary
.clone()onto_account_metas().0since it's already owned- All error paths use
map_errorok_orto produce structuredTaskBuilderErrorvariants- No
.unwrap()or.expect()callsThe flow correctly derives the CDA, fetches the compressed account and validity proof, packs the tree info, and constructs the
CompressedDatafor downstream task consumption.
There was a problem hiding this comment.
Actionable comments posted: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
magicblock-committor-service/src/transaction_preparator/mod.rs (1)
103-124: Replace.expect()with proper error propagation.Line 120 uses
.expect()which violates the project's coding guidelines for production Rust code inmagicblock-*paths. Although the comment states the possibility was "checked above," the prior check at lines 91-99 only validates with dummy lookup tables—the actual lookup tables returned fromprepare_for_deliverymay differ, making this expectation potentially invalid.As per coding guidelines,
.expect()in production code is a MAJOR issue.🔎 Proposed fix
- let message = TransactionUtils::assemble_tasks_tx( - authority, - &tx_strategy.optimized_tasks, - self.compute_budget_config.compute_unit_price, - &lookup_tables, - ) - .expect("Possibility to assemble checked above") - .message; + let message = TransactionUtils::assemble_tasks_tx( + authority, + &tx_strategy.optimized_tasks, + self.compute_budget_config.compute_unit_price, + &lookup_tables, + )? + .message;This requires updating the return type or error conversion, but provides proper error handling instead of panicking.
♻️ Duplicate comments (1)
magicblock-committor-service/src/intent_executor/two_stage_executor.rs (1)
192-216: Log errors when fetching commit transaction slot.The closure at lines 197-216 silently discards RPC errors via
.ok()but only logs on success (line 211). Any RPC failure collapses toNonewithout visibility. Additionally, line 211 logs the entire transaction object which is verbose—consider logging just the slot.🔎 Proposed fix
let commit_slot_fn: CommitSlotFn = Arc::new(move || { let rpc_client = rpc_client.clone(); Box::pin(async move { - rpc_client + match rpc_client .get_transaction( &sig, Some(RpcTransactionConfig { commitment: Some(CommitmentConfig::confirmed()), max_supported_transaction_version: Some(0), ..Default::default() }), ) .await - .map(|tx| { - info!("Commit slot for tx {:?}", tx); - tx.slot - }) - .ok() + { + Ok(tx) => { + info!("Commit slot: {}", tx.slot); + Some(tx.slot) + } + Err(err) => { + warn!("Failed to fetch commit transaction slot for {}: {}", sig, err); + None + } + } }) });
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (5)
magicblock-committor-service/src/intent_executor/mod.rsmagicblock-committor-service/src/intent_executor/two_stage_executor.rsmagicblock-committor-service/src/transaction_preparator/delivery_preparator.rsmagicblock-committor-service/src/transaction_preparator/mod.rstest-integration/test-committor-service/tests/test_ix_commit_local.rs
🧰 Additional context used
📓 Path-based instructions (2)
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-committor-service/src/transaction_preparator/mod.rsmagicblock-committor-service/src/intent_executor/two_stage_executor.rsmagicblock-committor-service/src/transaction_preparator/delivery_preparator.rsmagicblock-committor-service/src/intent_executor/mod.rs
{test-*,tools}/**
⚙️ CodeRabbit configuration file
Usage of
.unwrap()or.expect()in test code is acceptable and may be treated as trivial.
Files:
test-integration/test-committor-service/tests/test_ix_commit_local.rs
🧠 Learnings (20)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
magicblock-committor-service/src/transaction_preparator/mod.rsmagicblock-committor-service/src/intent_executor/two_stage_executor.rsmagicblock-committor-service/src/transaction_preparator/delivery_preparator.rstest-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-21T10:22:07.520Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 661
File: magicblock-committor-service/src/intent_executor/single_stage_executor.rs:20-28
Timestamp: 2025-11-21T10:22:07.520Z
Learning: In magicblock-committor-service's SingleStageExecutor and TwoStageExecutor (single_stage_executor.rs and two_stage_executor.rs), the fields transaction_strategy, junk, and patched_errors are intentionally public because these executors are designed to be used independently outside of the IntentExecutor scope, and callers need access to these execution reports for cleanup and error handling.
Applied to files:
magicblock-committor-service/src/transaction_preparator/mod.rsmagicblock-committor-service/src/intent_executor/two_stage_executor.rsmagicblock-committor-service/src/transaction_preparator/delivery_preparator.rstest-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-committor-service/src/intent_executor/two_stage_executor.rsmagicblock-committor-service/src/transaction_preparator/delivery_preparator.rstest-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-committor-service/src/intent_executor/two_stage_executor.rstest-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-10-28T13:15:42.706Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 596
File: magicblock-processor/src/scheduler.rs:1-1
Timestamp: 2025-10-28T13:15:42.706Z
Learning: In magicblock-processor, transaction indexes were always set to 0 even before the changes in PR #596. The proper transaction indexing within slots will be addressed during the planned ledger rewrite.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs
📚 Learning: 2025-11-26T13:27:46.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rstest-integration/test-committor-service/tests/test_ix_commit_local.rs
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rstest-integration/test-committor-service/tests/test_ix_commit_local.rs
📚 Learning: 2025-10-21T13:06:38.900Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/http/get_program_accounts.rs:17-25
Timestamp: 2025-10-21T13:06:38.900Z
Learning: The magicblock validator does not support ledger forking, so commitment-based state queries (processed/confirmed/finalized) are not applicable. RPC methods can safely ignore commitment and minContextSlot parameters from Solana RPC config objects.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs
📚 Learning: 2025-10-21T11:00:18.396Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/encoder.rs:176-187
Timestamp: 2025-10-21T11:00:18.396Z
Learning: In the magicblock validator, the current slot is always the root slot. The SlotEncoder in magicblock-aperture/src/encoder.rs correctly sets `root: slot` because there is no lag between current and root slots in this architecture.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs
📚 Learning: 2025-11-20T08:57:07.217Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 650
File: magicblock-chainlink/src/submux/subscription_task.rs:13-99
Timestamp: 2025-11-20T08:57:07.217Z
Learning: In the magicblock-validator repository, avoid posting review comments that merely confirm code is correct or matches intended behavior without providing actionable feedback, suggestions for improvement, or identifying potential issues. Such confirmatory comments are considered unhelpful noise by the maintainers.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-21T11:03:26.756Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 664
File: magicblock-chainlink/src/testing/mod.rs:342-370
Timestamp: 2025-11-21T11:03:26.756Z
Learning: In the magicblock-validator codebase, avoid leaving review comments that merely acknowledge code is correct or well-structured when there is no actionable suggestion, improvement, or issue to flag. Only comment when there is something specific to recommend, fix, or clarify.
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rstest-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
test-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
test-integration/test-committor-service/tests/test_ix_commit_local.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
test-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
test-integration/test-committor-service/tests/test_ix_commit_local.rsmagicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-24T14:21:00.996Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: Cargo.toml:58-58
Timestamp: 2025-11-24T14:21:00.996Z
Learning: In the magicblock-validator codebase, magicblock-api/Cargo.toml intentionally uses borsh = "1.5.3" (instead of the workspace version 0.10.4) because it needs to deserialize types from the magic-domain-program external dependency, which requires borsh 1.5.x compatibility. This is an intentional exception for interoperability with the magic domain program.
Applied to files:
test-integration/test-committor-service/tests/test_ix_commit_local.rs
📚 Learning: 2025-10-14T09:56:14.047Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 564
File: test-integration/programs/flexi-counter/src/processor/call_handler.rs:122-125
Timestamp: 2025-10-14T09:56:14.047Z
Learning: The file test-integration/programs/flexi-counter/src/processor/call_handler.rs contains a test smart contract used for integration testing, not production code.
Applied to files:
magicblock-committor-service/src/intent_executor/mod.rs
📚 Learning: 2025-11-13T09:38:43.804Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/locks.rs:64-102
Timestamp: 2025-11-13T09:38:43.804Z
Learning: In magicblock-processor's TransactionScheduler (scheduler/mod.rs line 59), the executor count is clamped to MAX_SVM_EXECUTORS (63) at initialization time, and executor IDs are assigned sequentially from 0 to count-1. This architectural guarantee ensures that executor IDs used in the bitmask-based AccountLock (scheduler/locks.rs) will always be within valid bounds for bit shifting operations, making runtime bounds checks unnecessary.
Applied to files:
magicblock-committor-service/src/intent_executor/mod.rs
🧬 Code graph analysis (3)
magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (3)
magicblock-committor-service/src/tasks/task_builder.rs (1)
get_compressed_data(525-581)magicblock-committor-service/src/tasks/buffer_task.rs (4)
get_compressed_data(171-175)strategy(135-137)delegated_account(177-183)from(72-78)magicblock-committor-service/src/intent_executor/two_stage_executor.rs (1)
None(86-86)
test-integration/test-committor-service/tests/test_ix_commit_local.rs (5)
magicblock-core/src/compression/mod.rs (1)
derive_cda_from_pda(14-26)magicblock-committor-service/src/service.rs (1)
try_start(265-294)magicblock-committor-service/src/tasks/buffer_task.rs (1)
is_compressed(160-162)magicblock-committor-service/src/tasks/mod.rs (1)
is_compressed(103-103)compressed-delegation-client/src/lib.rs (1)
id(61-63)
magicblock-committor-service/src/intent_executor/mod.rs (2)
magicblock-committor-service/src/intent_executor/task_info_fetcher.rs (2)
Self(314-322)Self(324-332)magicblock-committor-service/src/tasks/task_builder.rs (2)
finalize_tasks(60-64)finalize_tasks(259-479)
🔇 Additional comments (26)
test-integration/test-committor-service/tests/test_ix_commit_local.rs (8)
8-11: LGTM!The new imports for compressed account support are well-organized and correctly bring in the necessary types from the light-client indexer and compressed delegation client.
58-63: LGTM!The
CommitAccountModeenum cleanly captures the four test mode combinations and follows Rust naming conventions.
184-200: LGTM!The explicit match arms for
CommitAccountModevariants are now used, addressing the previous review feedback about replacing the wildcard pattern. This provides compile-time safety for future enum additions.
619-628: LGTM!The expected strategy count is now correctly set to 5 commits for the 5 compressed accounts, addressing the previous review feedback.
697-738: LGTM!The
create_bundlesfunction is well-structured with clean branching logic for compressed vs. regular account initialization. The async spawning pattern is appropriate for parallel account creation.
767-771: LGTM!Good use of the
matches!macro for checking compressed variants - this is idiomatic Rust and maintains clarity.
928-964: LGTM!The compressed account validation logic correctly:
- Validates the on-chain "shell" account has owner
compressed_delegation_client::IDwith empty data- Derives the CDA address and fetches the compressed account from the Photon indexer
- Validates the compressed account data against expectations
Based on learnings, the PhotonIndexer has built-in retry logic (~10 attempts), so the absence of explicit retry loops here is intentional and correct.
1115-1163: LGTM!The
validate_compressed_accountfunction correctly:
- Parses the
CompressedDelegationRecordfrom the compressed account data- Validates lamports, data, and owner against expectations
- Logs the actual compared values (
data.lamports,data.owner) in trace messagesThis addresses the previous review feedback about inconsistent logging.
magicblock-committor-service/src/transaction_preparator/mod.rs (2)
1-24: LGTM - Import updates for PhotonIndexer integration.The new imports properly support the PhotonIndexer and CommitSlotFn threading through the transaction preparation flow.
29-49: LGTM - Trait signature extended for compressed account support.The
prepare_for_strategysignature now acceptsphoton_clientandcommit_slot_fnparameters to enable compressed data fetching during task preparation. The lifetime annotation'ais correctly applied to support the async closure.magicblock-committor-service/src/intent_executor/two_stage_executor.rs (3)
1-22: LGTM - Import updates for commit slot functionality.The imports properly support the new
CommitSlotFnmechanism and RPC transaction configuration.
71-134: LGTM - Commit phase passes None for commit_slot_fn.During the commit phase, there's no prior commit to reference, so passing
None::<CommitSlotFn>is semantically correct.
218-279: LGTM - Finalize loop properly threads commit_slot_fn.The finalize execution loop correctly passes the cloned
commit_slot_fnto enable slot-aware data fetching for compressed accounts.magicblock-committor-service/src/intent_executor/mod.rs (7)
63-65: LGTM - Type aliases for async commit slot function.The
BoxFutandCommitSlotFntype aliases properly encapsulate the async slot-fetching capability with correct lifetime bounds and thread-safety constraints (Send + Sync).
114-148: LGTM - PhotonIndexer field added to executor.The
photon_client: Option<Arc<PhotonIndexer>>field is properly integrated intoIntentExecutorImpland initialized in the constructor, addressing the earlier concern about hard-coded localhost instances.
170-196: LGTM - Single-stage flow threads photon_client.The
commit_taskscall now properly receives&self.photon_clientto support compressed account handling in standalone actions.
198-246: LGTM - Two-stage task building threads photon_client.Both
commit_tasksandfinalize_tasksfutures correctly receive the photon client reference for compressed data support.
334-378: Acknowledge existing TODO for mixed compression handling.The TODO at line 338 correctly flags the gap where mixed compressed/non-compressed tasks in a single strategy aren't properly handled. The current
.any()approach treats the entire strategy as compressed if any task is compressed, which may cause incorrect commit ID fetching for non-compressed tasks.This was flagged in a prior review and marked as addressed—verify the intended handling is documented or tracked.
563-575: LGTM - Error handling extended for compression-related failures.The new
InconsistentTaskCompressionerror variant is properly handled inpersist_result, mapping toCommitStatus::Failedwhich is appropriate for this configuration error.
635-665: LGTM - prepare_and_execute_strategy properly threads new parameters.The function now accepts
commit_slot_fnand passes both it andphoton_clienttoprepare_for_strategy, enabling the full compressed-data flow.magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs (6)
1-42: LGTM - Import updates for compressed data support.The imports properly support
PhotonIndexer,IndexerRpcConfig,CommitSlotFn, and the new task types (BufferPreparationTask,PreparationTask).
63-107: LGTM - prepare_for_delivery extended for photon context.The signature now accepts
photon_clientandcommit_slot_fn, which are correctly cloned and threaded into per-task preparation via the async closure.
109-205: LGTM - Task preparation handles Buffer and Compressed variants.The new match structure properly separates:
- Buffer path (lines 125-166): Maintains existing buffer initialization, chunk writing, and status persistence with proper state transitions.
- Compressed path (lines 167-201): Fetches fresh compressed data from the indexer with slot-aware configuration, includes proper error handling with contextual logging.
The error handling at lines 193-198 addresses the prior review feedback about preserving error context. Based on learnings, Photon handles slot validation internally, so the
IndexerRpcConfigusage is correct.
207-266: LGTM - Error recovery properly threads photon context.The
prepare_task_handling_errorsfunction correctly:
- Passes
commit_slot_fn.clone()to handle the lifetime properly- Handles
AccountAlreadyInitializedErrorwith cleanup and retry- Threads photon context through the retry path
268-335: LGTM - Buffer functions updated to accept BufferPreparationTask.The parameter type change from generic
PreparationTasktoBufferPreparationTaskimproves type safety and clarifies the expected input for buffer-specific operations.
620-648: LGTM - Error variants properly distinguish failure modes.New variants added:
CompressedDataNotFound- for missing compressed dataDelegatedAccountNotFound- when task lacks delegated accountPhotonClientNotFound- when photon client is required but absentTaskBuilderError- wraps upstream builder errorsTransactionCreationError/TransactionSigningError- for transaction construction failuresThese provide clear error categorization for debugging compressed flows.
Co-authored-by: Gabriele Picco <piccogabriele@gmail.com>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
There was a problem hiding this comment.
Actionable comments posted: 2
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (2)
Cargo.lockis excluded by!**/*.locktest-integration/Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (3)
config.example.tomlmagicblock-chainlink/src/chainlink/fetch_cloner.rsmagicblock-committor-service/src/tasks/task_builder.rs
🧰 Additional context used
📓 Path-based instructions (1)
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-committor-service/src/tasks/task_builder.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
🧠 Learnings (13)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-07T13:20:13.793Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 589
File: magicblock-processor/src/scheduler/coordinator.rs:227-238
Timestamp: 2025-11-07T13:20:13.793Z
Learning: In magicblock-processor's ExecutionCoordinator (scheduler/coordinator.rs), the `account_contention` HashMap intentionally does not call `shrink_to_fit()`. Maintaining slack capacity is beneficial for performance by avoiding frequent reallocations during high transaction throughput. As long as empty entries are removed from the map (which `clear_account_contention` does), the capacity overhead is acceptable.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rsmagicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-21T10:22:07.520Z
Learnt from: taco-paco
Repo: magicblock-labs/magicblock-validator PR: 661
File: magicblock-committor-service/src/intent_executor/single_stage_executor.rs:20-28
Timestamp: 2025-11-21T10:22:07.520Z
Learning: In magicblock-committor-service's SingleStageExecutor and TwoStageExecutor (single_stage_executor.rs and two_stage_executor.rs), the fields transaction_strategy, junk, and patched_errors are intentionally public because these executors are designed to be used independently outside of the IntentExecutor scope, and callers need access to these execution reports for cleanup and error handling.
Applied to files:
magicblock-committor-service/src/tasks/task_builder.rs
📚 Learning: 2025-12-17T12:46:36.207Z
Learnt from: GabrielePicco
Repo: magicblock-labs/magicblock-validator PR: 738
File: magicblock-chainlink/src/chainlink/fetch_cloner.rs:1109-1196
Timestamp: 2025-12-17T12:46:36.207Z
Learning: In magicblock-chainlink/src/chainlink/fetch_cloner.rs, eATA account subscriptions are intentionally NOT cancelled even when the eATA doesn't exist or delegation conversion fails. The subscriptions are kept active to watch for future creation or delegation of eATA accounts (per maintainer GabrielePicco).
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
🧬 Code graph analysis (1)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (2)
magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(29-31)compressed-delegation-client/src/lib.rs (1)
id(61-63)
🔇 Additional comments (10)
magicblock-committor-service/src/tasks/task_builder.rs (8)
1-38: Imports look well-organized.The imports are properly structured with external dependencies followed by local module imports. The duplicate import issue from previous reviews has been addressed.
40-47: Data structure is well-defined.The
CompressedDatastruct appropriately encapsulates all necessary compressed account information for task processing.
49-65: Trait API cleanly extends for compression support.The optional
PhotonIndexerparameter enables compressed paths without breaking existing non-compressed flows.
67-109: Commit task creation logic is correct.The threshold-based decision between
CommitDiffandCommittasks is well-implemented, with appropriate fallback when base account is unavailable.
111-256: Commit tasks implementation is well-structured.Good improvements from previous reviews:
- Proper error propagation with
ok_or(TaskBuilderError::MissingCommitId)instead of.expect()- Concurrent compressed data fetching with
FuturesUnorderedandtry_collect- Clean separation between compressed and non-compressed paths
258-479: Finalize tasks implementation addresses previous feedback well.Key improvements:
get_compressed_data_for_accountsnow usesFuturesUnorderedfor concurrent fetching- Consistent error handling via
?propagation throughout- TODO at lines 451-460 properly documents the compressed undelegate limitation with issue reference
The separation of helpers (
finalize_task,undelegate_task,get_compressed_data_for_accounts,process_commit) improves readability and maintainability.
482-521: Error handling is comprehensive.The
TaskBuilderErrorenum properly covers all compression-related failure modes including:
MissingCommitId(replaces previous.expect()usage)- Distinct errors for state trees, address, and data availability
- Proper source attribution with
#[source]for error chaining
525-581: Compressed data fetching is robust.All previous review concerns have been addressed:
- Line 557-561: Safe extraction with
.first().copied().ok_or()prevents panic on emptypacked_tree_infos- Line 577: Removed unnecessary
.clone()on ownedVec- Consistent error propagation via
map_errand?magicblock-chainlink/src/chainlink/fetch_cloner.rs (2)
1105-1142: Compressed delegation handling in bulk fetch path looks correct.The logic correctly deserializes the
CompressedDelegationRecord, sets all necessary fields (compressed,lamports,owner,data,delegated), and properly calculatesdelegated_to_otherby checking for bothvalidator_pubkeyandPubkey::default()(confined accounts).This addresses the previously flagged issues about missing lamports/compressed flag and delegated_to_other field.
574-582: Track commit frequency limitation for compressed delegation.The TODO comment indicates that
commit_frequency_msis not supported for compressed delegation records. This limitation means that compressed delegated accounts always use a commit frequency of 0, which might not match the intended behavior.Would you like me to open an issue to track implementing commit frequency support for compressed delegation records, or help implement this feature?
⛔ Skipped due to learnings
Learnt from: Dodecahedr0x Repo: magicblock-labs/magicblock-validator PR: 639 File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114 Timestamp: 2025-11-19T11:31:24.218Z Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.Learnt from: Dodecahedr0x Repo: magicblock-labs/magicblock-validator PR: 639 File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881 Timestamp: 2025-12-03T09:33:48.707Z Learning: Repo: magicblock-labs/magicblock-validator PR: 639 Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local) Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.Learnt from: bmuddha Repo: magicblock-labs/magicblock-validator PR: 578 File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27 Timestamp: 2025-10-21T14:00:54.642Z Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579Learnt from: Dodecahedr0x Repo: magicblock-labs/magicblock-validator PR: 639 File: test-integration/programs/flexi-counter/src/processor.rs:643-680 Timestamp: 2025-11-25T11:07:20.001Z Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (3)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (3)
1879-1879: MAJOR: Avoid.unwrap()in production code paths.The
.unwrap()here assumes the vector always has exactly 2 elements, but if this invariant is violated, the entire validator could panic. Even with the safety comment, this violates the coding guideline requiring proper error handling in production code.🔎 Proposed fix
.await - // SAFETY: we always get two results here - .map(|mut accs| { - let acc_last = accs.pop().unwrap(); - let acc_first = accs.pop().unwrap(); + .and_then(|mut accs| { + if accs.len() != 2 { + return Err(ChainlinkError::ProgramAccountResolutionsFailed( + format!("Expected 2 accounts, got {}", accs.len()) + ).into()); + } + let acc_last = accs.pop().expect("checked len"); + let acc_first = accs.pop().expect("checked len"); (acc_first, acc_last) }) - .map_err(ChainlinkError::from) - .and_then(|(acc, deleg)| { + .and_then(|(acc, deleg)| {As per coding guidelines.
2728-2728: MAJOR: Replace.expect()on lock poisoning with proper error handling.The
.expect()here will panic the entire validator if the lock is poisoned. Lock poisoning should be handled gracefully by returning an error result rather than crashing the process.🔎 Proposed fix
{ - let mut pending = self - .pending_requests - .lock() - .expect("pending_requests lock poisoned"); + let mut pending = self.pending_requests.lock().map_err(|_| { + ChainlinkError::Other( + "Failed to acquire pending_requests lock (poisoned)".into(), + ) + })?;As per coding guidelines. Note that this requires propagating the error from this code block. Similar fix needed at line 2783.
2783-2783: MAJOR: Replace.expect()on lock poisoning with proper error handling.Same issue as line 2728 - panicking on lock poisoning will crash the validator. This should return an error instead.
🔎 Proposed fix
{ - let mut pending = self - .pending_requests - .lock() - .expect("pending_requests lock poisoned"); + let mut pending = self.pending_requests.lock().map_err(|_| { + ChainlinkError::Other( + "Failed to acquire pending_requests lock (poisoned)".into(), + ) + })?;As per coding guidelines.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (1)
Cargo.lockis excluded by!**/*.lock
📒 Files selected for processing (2)
config.example.tomlmagicblock-chainlink/src/chainlink/fetch_cloner.rs
🧰 Additional context used
📓 Path-based instructions (1)
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
🧠 Learnings (12)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-17T12:46:36.207Z
Learnt from: GabrielePicco
Repo: magicblock-labs/magicblock-validator PR: 738
File: magicblock-chainlink/src/chainlink/fetch_cloner.rs:1109-1196
Timestamp: 2025-12-17T12:46:36.207Z
Learning: In magicblock-chainlink/src/chainlink/fetch_cloner.rs, eATA account subscriptions are intentionally NOT cancelled even when the eATA doesn't exist or delegation conversion fails. The subscriptions are kept active to watch for future creation or delegation of eATA accounts (per maintainer GabrielePicco).
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-15T20:32:57.435Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 746
File: magicblock-config/src/tests.rs:680-703
Timestamp: 2025-12-15T20:32:57.435Z
Learning: In the magicblock-validator repository, gRPC remotes do not support URL aliases (mainnet, devnet, local) like RPC and WebSocket remotes do. The resolve_url function passes gRPC URLs through as-is without alias resolution.
Applied to files:
config.example.toml
🧬 Code graph analysis (1)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (2)
magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(29-31)compressed-delegation-client/src/lib.rs (1)
id(61-63)
🔇 Additional comments (4)
config.example.toml (2)
61-65: LGTM: Formatting improvement for remotes array.The multi-line format with trailing commas improves readability and makes version control diffs cleaner when adding or removing remote endpoints.
233-246: LGTM: Compression configuration section is well-structured.The new
[compression]section appropriately introduces optional Photon indexer configuration fields. The fields are correctly commented out as examples, environment variable naming follows the established pattern, and the documentation is consistent with other sections in the config file. This aligns well with the PR's objective of adding compressed-account support.magicblock-chainlink/src/chainlink/fetch_cloner.rs (2)
574-575: Track the TODO for compressed delegation commit frequency support.The TODO comment indicates that commit frequency is not yet supported for compressed delegation records. This converts the record with
commit_frequency_ms: 0, which may not reflect the actual desired behavior.Consider opening an issue to track this limitation and determine if commit frequency support should be added for compressed delegation records, or if the current behavior (always using 0) is intentional and acceptable.
1587-1590: LGTM: Compressed account handling in undelegation refresh.Correctly short-circuits the delegation record fetch for compressed accounts, since compressed delegation records are stored inline with the account data rather than in a separate PDA.
Based on learnings.
There was a problem hiding this comment.
Actionable comments posted: 2
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
📒 Files selected for processing (1)
magicblock-chainlink/src/chainlink/fetch_cloner.rs
🧰 Additional context used
📓 Path-based instructions (1)
{magicblock-*,programs,storage-proto}/**
⚙️ CodeRabbit configuration file
{magicblock-*,programs,storage-proto}/**: Treat any usage of.unwrap()or.expect()in production Rust code as a MAJOR issue.
These should not be categorized as trivial or nit-level concerns.
Request proper error handling or explicit justification with invariants.
Files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
🧠 Learnings (11)
📓 Common learnings
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-committor-service/src/transaction_preparator/delivery_preparator.rs:172-193
Timestamp: 2025-11-26T13:27:46.702Z
Learning: The Photon indexer (PhotonIndexer from light-client) handles slot validation internally when an IndexerRpcConfig with a slot constraint is provided. Additional slot validation at the call site is not needed when fetching compressed data via get_compressed_data.
📚 Learning: 2025-12-03T09:33:48.707Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-committor-service/tests/test_ix_commit_local.rs:867-881
Timestamp: 2025-12-03T09:33:48.707Z
Learning: Repo: magicblock-labs/magicblock-validator PR: 639
Context: test-integration/test-committor-service/tests/test_ix_commit_local.rs (ix_commit_local)
Learning: The PhotonIndexer used for compressed account fetches (get_compressed_account) has built‑in retry logic (defaults to ~10 attempts), so tests should not add separate retry loops around compressed fetches unless there’s a specific need.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-17T12:46:36.207Z
Learnt from: GabrielePicco
Repo: magicblock-labs/magicblock-validator PR: 738
File: magicblock-chainlink/src/chainlink/fetch_cloner.rs:1109-1196
Timestamp: 2025-12-17T12:46:36.207Z
Learning: In magicblock-chainlink/src/chainlink/fetch_cloner.rs, eATA account subscriptions are intentionally NOT cancelled even when the eATA doesn't exist or delegation conversion fails. The subscriptions are kept active to watch for future creation or delegation of eATA accounts (per maintainer GabrielePicco).
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-03T09:36:01.527Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/src/remote_account_provider/mod.rs:1350-1353
Timestamp: 2025-12-03T09:36:01.527Z
Learning: Repo: magicblock-labs/magicblock-validator
File: magicblock-chainlink/src/remote_account_provider/mod.rs
Context: consolidate_fetched_remote_accounts
Learning: For unexpected result counts (>2), the project prefers logging an error and returning an empty Vec over panicking; acceptable during development per maintainer (Dodecahedr0x).
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T09:34:37.917Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: test-integration/test-chainlink/tests/ix_remote_account_provider.rs:62-63
Timestamp: 2025-11-19T09:34:37.917Z
Learning: In test-integration/test-chainlink/tests/ix_remote_account_provider.rs and similar test files, the `_fwd_rx` receiver returned by `init_remote_account_provider()` is intentionally kept alive (but unused) to prevent "receiver dropped" errors on the sender side. The pattern `let (remote_account_provider, _fwd_rx) = init_remote_account_provider().await;` should NOT be changed to `let (remote_account_provider, _) = ...` because dropping the receiver would cause send() operations to fail.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-10-21T14:00:54.642Z
Learnt from: bmuddha
Repo: magicblock-labs/magicblock-validator PR: 578
File: magicblock-aperture/src/requests/websocket/account_subscribe.rs:18-27
Timestamp: 2025-10-21T14:00:54.642Z
Learning: In magicblock-aperture account_subscribe handler (src/requests/websocket/account_subscribe.rs), the RpcAccountInfoConfig fields data_slice, commitment, and min_context_slot are currently ignored—only encoding is applied. This is tracked as technical debt in issue #579: https://github.com/magicblock-labs/magicblock-validator/issues/579
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-07T14:20:31.457Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 621
File: magicblock-chainlink/src/remote_account_provider/chain_pubsub_actor.rs:457-495
Timestamp: 2025-11-07T14:20:31.457Z
Learning: In magicblock-chainlink/src/remote_account_provider/chain_pubsub_client.rs, the unsubscribe closure returned by PubSubConnection::account_subscribe(...) resolves to () (unit), not a Result. Downstream code should not attempt to inspect an unsubscribe result and can optionally wrap it in a timeout to guard against hangs.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-19T11:31:24.218Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/test-chainlink/tests/ix_01_ensure-accounts.rs:85-114
Timestamp: 2025-11-19T11:31:24.218Z
Learning: In the magicblock-validator codebase, compressed delegation records are never subscribed to by design. Tests for compressed delegation flows should focus on verifying that the delegated account (e.g., counter PDA) is cloned as delegated and not subscribed, but do not need to check for subscription to delegation-record-like PDAs in the compressed path because subscriptions to compressed delegation records never occur.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-18T08:47:39.702Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: magicblock-chainlink/tests/04_redeleg_other_separate_slots.rs:158-165
Timestamp: 2025-11-18T08:47:39.702Z
Learning: In magicblock-chainlink tests involving compressed accounts, `set_remote_slot()` sets the slot of the `AccountSharedData`, while `compressed_account_shared_with_owner_and_slot()` sets the slot of the delegation record. These are two different fields and both calls are necessary.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-12-01T16:02:05.367Z
Learnt from: thlorenz
Repo: magicblock-labs/magicblock-validator PR: 703
File: magicblock-chainlink/src/submux/mod.rs:652-654
Timestamp: 2025-12-01T16:02:05.367Z
Learning: In magicblock-chainlink/src/submux/mod.rs, the subscribe_program method intentionally adds program_id to program_subs before attempting the subscription. This ensures that even if the initial subscription fails or only partially succeeds across clients, the reconnection logic will retry the subscription. This is a deliberate design pattern for resilience in the multi-client architecture and should not be "fixed" to remove entries on failure.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
📚 Learning: 2025-11-25T11:07:20.001Z
Learnt from: Dodecahedr0x
Repo: magicblock-labs/magicblock-validator PR: 639
File: test-integration/programs/flexi-counter/src/processor.rs:643-680
Timestamp: 2025-11-25T11:07:20.001Z
Learning: In the compressed delegation architecture, ownership reassignment during undelegation is handled by the delegation program itself, not by the owner program's external undelegate handler. The external undelegate handler (e.g., `process_external_undelegate_compressed` in flexi-counter) is a callback for program-specific cleanup (restoring lamports and data) after the delegation program has already reassigned ownership back to the original owner stored in the CompressedDelegationRecord.
Applied to files:
magicblock-chainlink/src/chainlink/fetch_cloner.rs
🧬 Code graph analysis (1)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (4)
magicblock-chainlink/src/remote_account_provider/photon_client.rs (1)
new(29-31)magicblock-chainlink/src/remote_account_provider/mod.rs (1)
new(323-377)compressed-delegation-client/src/lib.rs (1)
id(61-63)magicblock-chainlink/src/chainlink/account_still_undelegating_on_chain.rs (1)
account_still_undelegating_on_chain(22-89)
🔇 Additional comments (6)
magicblock-chainlink/src/chainlink/fetch_cloner.rs (6)
11-12: LGTM: Generic PhotonClient parameter properly threaded.The addition of the
P: PhotonClientgeneric parameter and related imports is correctly implemented. The type signatures properly propagate the constraint throughFetchCloner,RemoteAccountProvider, andimplblocks.Also applies to: 25-25, 42-42, 56-83, 152-158
574-584: Documented limitation: commit_frequency_ms not supported for compressed delegation.The TODO comment notes that
commit_frequency_msis set to 0 for compressed delegation records. This is acceptable as documented technical debt, but verify that downstream code handles 0 commit frequency correctly or has appropriate defaults.
1105-1142: LGTM: Compressed delegation bulk fetch correctly mirrors subscription path.The bulk fetch logic for compressed delegation accounts properly:
- Deserializes
CompressedDelegationRecordfrom account data- Sets
compressed(true),lamports,owner,datafrom the record- Determines delegation status based on
authoritycomparison- Computes
delegated_to_otherfor accounts delegated to other validators- Handles errors appropriately by filtering out unparseable records
This matches the subscription path behavior (lines 565-573) and addresses previous review feedback about propagating
delegated_to_other.
1587-1623: LGTM: Compressed account undelegation refresh correctly skips delegation record fetch.For compressed accounts, the code correctly returns
RefreshDecision::Yesimmediately without fetching a delegation record. This is appropriate because compressed delegation records are embedded in the account data itself, not stored separately.For non-compressed delegated accounts, the existing logic is preserved: fetch the delegation record and check if undelegation is complete on-chain.
2132-2139: LGTM: Helper function signature correctly updated for PhotonClient generic.The
cancel_subshelper function signature is properly updated to include theP: PhotonClientconstraint and acceptArc<RemoteAccountProvider<T, U, P>>. The implementation logic remains unchanged.
2225-2405: LGTM: Test infrastructure correctly updated for PhotonClient generic.The test infrastructure changes properly thread the
PhotonClientMocktype through:
- Type aliases (
TestFetchClonerResult)- Test context structures (
FetcherTestCtx)- Helper functions (
setup,init_fetch_cloner)RemoteAccountProvider::new()calls withNone::<PhotonClientMock>These changes enable testing both compressed and non-compressed delegation flows with appropriate mock implementations.
| } else if owned_by_compressed_delegation_program { | ||
| // If the account is compressed, the delegation record is in the account itself | ||
| let delegation_record = | ||
| match CompressedDelegationRecord::try_from_slice( | ||
| account.data(), | ||
| ) { | ||
| Ok(delegation_record) => Some(delegation_record), | ||
| Err(_err) => { | ||
| debug!("The account's data did not contain a valid compressed delegation record, fetching..."); | ||
| if let Some(acc) = self | ||
| .remote_account_provider | ||
| .try_get(pubkey, AccountFetchOrigin::GetAccount) | ||
| .await | ||
| .map(|acc| acc.fresh_account().cloned()) | ||
| .ok() | ||
| .flatten() | ||
| { | ||
| CompressedDelegationRecord::try_from_slice( | ||
| acc.data(), | ||
| ) | ||
| .ok() | ||
| } else { | ||
| error!("Failed to parse compressed delegation record for {pubkey} directly from the data."); | ||
| None | ||
| } | ||
| } | ||
| }; |
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Minor: Improve error context in compressed delegation record fallback path.
When direct parsing of the compressed delegation record fails (line 543), the error is discarded without logging. Later, if the fallback fetch succeeds but parsing still fails, the error message at line 558 says "Failed to parse" without indicating whether the issue was the fetch or the parse.
Consider:
- Log the initial parse error at debug level for troubleshooting
- Distinguish between fetch failures and parse failures in the error message
🔎 Suggested refinement
let delegation_record =
match CompressedDelegationRecord::try_from_slice(
account.data(),
) {
Ok(delegation_record) => Some(delegation_record),
- Err(_err) => {
- debug!("The account's data did not contain a valid compressed delegation record, fetching...");
+ Err(err) => {
+ debug!("Failed to parse compressed delegation record directly: {err}. Fetching fresh account...");
if let Some(acc) = self
.remote_account_provider
.try_get(pubkey, AccountFetchOrigin::GetAccount)
.await
.map(|acc| acc.fresh_account().cloned())
.ok()
.flatten()
{
CompressedDelegationRecord::try_from_slice(
acc.data(),
)
.ok()
} else {
- error!("Failed to parse compressed delegation record for {pubkey} directly from the data.");
+ error!("Failed to fetch or parse compressed delegation record for {pubkey}.");
None
}
}
};🤖 Prompt for AI Agents
In magicblock-chainlink/src/chainlink/fetch_cloner.rs around lines 536 to 562,
the initial CompressedDelegationRecord::try_from_slice parse error is discarded
and later error messaging doesn't differentiate between a remote fetch failure
and a parse failure; update the block to capture and debug-log the first parse
Err, then when attempting the remote fetch capture its result and if parsing the
fetched data fails log an error stating "fetched account parsed failed for
{pubkey}: {parse_err}", and if the remote fetch itself fails log "remote fetch
failed for {pubkey}: {fetch_err}"; ensure both parse errors use .to_string() or
Debug formatting and use debug for non-fatal parse details and error for final
failures so callers can distinguish fetch vs parse causes.
| let delegated_to_other = if delegation_record.authority.eq(&self.validator_pubkey) || delegation_record.authority.eq(&Pubkey::default()) { | ||
| None | ||
| } else { | ||
| Some(delegation_record.authority) | ||
| }; |
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Optional: Consider reusing delegation authority check logic.
The inline logic for computing delegated_to_other duplicates the pattern in get_delegated_to_other (lines 747-756). While this works correctly, extracting a small helper function could reduce duplication and ensure consistency.
This is a minor maintainability suggestion and was noted in a previous review comment.
Based on past review comments.
🤖 Prompt for AI Agents
In magicblock-chainlink/src/chainlink/fetch_cloner.rs around lines 1129 to 1133,
the inline computation of delegated_to_other duplicates the delegation-authority
check already implemented in get_delegated_to_other (lines 747-756); refactor to
reuse that logic by either calling the existing get_delegated_to_other with the
delegation_record and self.validator_pubkey here or extract the shared check
into a small private helper used by both sites so the semantics remain identical
and you avoid duplicated code.
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.