[kariy]

Summary

Adds end-to-end coverage of the messaging service's gather/insert/checkpoint loop against real settlement nodes, and relocates the existing L1→L2 e2e test from katana-rpc-server so all messaging integration coverage lives in one place.

• tests/eth_settlement.rs — Anvil + StarknetMessagingLocal, isolated MessagingService (no L2 sequencer). Asserts the L1Handler tx lands in an in-memory pool, the L1→L2 index records the mapping, and the checkpoint advances. Two cases: single message + two messages in different blocks.
• tests/starknet_settlement.rs — A second Katana (TestNode) used purely as the settlement chain, hosting piltover_messaging_mock. The isolated MessagingService polls it and collects the MessageSent event emitted by send_message_to_appchain.
• tests/e2e.rs — Moved verbatim from crates/rpc/rpc-server/tests/messaging.rs::test_messaging. Still uses Anvil + a full Katana L2; complements the isolated tests above by exercising the full pipeline including the executor.
• MessagingService refactor — Generalized over any TransactionPool<Transaction = ExecutableTxWithHash> + Clone (defaults to TxPool). Lets the new tests use a Pool with NoopValidator instead of standing up a full stateful validator. Production callers (sequencer) require no changes thanks to the default type parameter.
• rpc-server cleanup — Removed test_messaging and now-unused imports from rpc/rpc-server/tests/messaging.rs. Kept estimate_message_fee and mod messages_status (RPC-method tests, not messaging-service tests). Dropped 5 alloy dev-deps from rpc-server/Cargo.toml that only the moved test needed.

Heads up — StarknetCollector page-size hardcode

crates/messaging/src/stream/collector/starknet.rs hardcodes chunk_size = 200 for get_events, but TestNode's default max_event_page_size is Some(100). The new Starknet test bumps the page size on its TestNode config; in production Katana with a strict event page cap below 200, the collector would silently fail every gather. Worth a follow-up to make the chunk size configurable or smaller by default.

Test plan

• cargo nextest run -p katana-messaging --tests — 24/24 pass (20 pre-existing unit + 4 new integration)
• cargo nextest run -p katana-rpc-server --test messaging — 4/4 pass (estimate_message_fee + 3 messages_status)
• cargo +nightly-2025-02-20 fmt --all
• ./scripts/clippy.sh

🤖 Generated with Claude Code

[github-actions]

Codec benchmark diff vs main

Benchmark	Baseline (ns)	Current (ns)	Δ
CompiledClass(fixture)/compress	2700139	2564207	-5.03%
CompiledClass(fixture)/decompress	2948520	2902683	-1.55%
ExecutionCheckpoint/compress	34	37	+8.82%
ExecutionCheckpoint/decompress	25	27	+8.00%
PruningCheckpoint/compress	35	35	+0.00%
PruningCheckpoint/decompress	25	26	+4.00%
VersionedHeader/compress	678	661	-2.51%
VersionedHeader/decompress	816	872	+6.86%
StoredBlockBodyIndices/compress	83	81	-2.41%
StoredBlockBodyIndices/decompress	36	40	+11.11%
StorageEntry/compress	154	167	+8.44%
StorageEntry/decompress	140	159	+13.57%
ContractNonceChange/compress	150	161	+7.33%
ContractNonceChange/decompress	236	259	+9.75%
ContractClassChange/compress	216	211	-2.31%
ContractClassChange/decompress	266	283	+6.39%
ContractStorageEntry/compress	161	169	+4.97%
ContractStorageEntry/decompress	332	351	+5.72%
GenericContractInfo/compress	141	139	-1.42%
GenericContractInfo/decompress	123	110	-10.57%
Felt/compress	82	90	+9.76%
Felt/decompress	57	60	+5.26%
BlockHash/compress	81	90	+11.11%
BlockHash/decompress	56	61	+8.93%
TxHash/compress	81	90	+11.11%
TxHash/decompress	57	60	+5.26%
ClassHash/compress	81	90	+11.11%
ClassHash/decompress	57	60	+5.26%
CompiledClassHash/compress	81	90	+11.11%
CompiledClassHash/decompress	57	60	+5.26%
BlockNumber/compress	47	50	+6.38%
BlockNumber/decompress	25	26	+4.00%
TxNumber/compress	47	50	+6.38%
TxNumber/decompress	25	26	+4.00%
FinalityStatus/compress	1	0	-100.00%
FinalityStatus/decompress	12	12	+0.00%
TypedTransactionExecutionInfo/compress	15185	14605	-3.82%
TypedTransactionExecutionInfo/decompress	3633	3676	+1.18%
VersionedContractClass/compress	370	411	+11.08%
VersionedContractClass/decompress	825	854	+3.52%
MigratedCompiledClassHash/compress	147	160	+8.84%
MigratedCompiledClassHash/decompress	142	159	+11.97%
ContractInfoChangeList/compress	1813	1520	-16.16%
ContractInfoChangeList/decompress	2287	2411	+5.42%
BlockChangeList/compress	700	659	-5.86%
BlockChangeList/decompress	955	978	+2.41%
ReceiptEnvelope/compress	29041	28641	-1.38%
ReceiptEnvelope/decompress	6144	6568	+6.90%
TrieDatabaseValue/compress	164	158	-3.66%
TrieDatabaseValue/decompress	219	239	+9.13%
TrieHistoryEntry/compress	296	313	+5.74%
TrieHistoryEntry/decompress	265	276	+4.15%

[github-actions]

Runner: AMD EPYC 9V74 80-Core Processor (4 cores) · 15Gi RAM

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 68.63%. Comparing base (9bde0ae) to head (00d1dab).
⚠️ Report is 424 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #574      +/-   ##
==========================================
- Coverage   73.32%   68.63%   -4.69%     
==========================================
  Files         209      321     +112     
  Lines       23132    45264   +22132     
==========================================
+ Hits        16961    31069   +14108     
- Misses       6171    14195    +8024     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.