[Fusion] Properly support persisted operations

[tobias-tengler]

No description provided.

[Copilot]

Pull Request Overview

This PR enhances persisted operations support in the Fusion gateway by fixing service resolution and hash computation issues. The changes ensure proper document hash generation and consistent service access patterns throughout the persisted operations pipeline.

Key changes:

• Fixed service resolution to use SchemaServices instead of Services in persisted operations middleware
• Updated document hash computation to handle documents without pre-computed hashes
• Added comprehensive test coverage for persisted operations in the Fusion gateway

Reviewed Changes

Copilot reviewed 32 out of 34 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
WritePersistedOperationMiddleware.cs	Fixed service resolution to retrieve IDocumentHashProvider from SchemaServices
ReadPersistedOperationMiddleware.cs	Fixed service resolution and updated hash computation logic to compute hash when not present
PersistedOperationTests.cs	Added new test file with comprehensive persisted operation scenarios
IOperationDocumentNodeProvider.cs	Moved interface to correct namespace (Execution)
IOperationDocumentHashProvider.cs	Moved interface to correct namespace (Execution)
OperationDocument.cs	Implemented IOperationDocumentNodeProvider interface
TransactionScopeHandlerTests.cs	Updated namespace and fixed whitespace in condition
Various snapshot files	Added test snapshots for persisted operation test cases
Various .txt test files	Removed obsolete test files

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[github-actions]

🚀 Performance Test Results

📊 Response Time

Current

Test	Min	Med	Max	Avg	P90	P95	P99
Single Fetch	1.45ms	2.75ms	58.16ms	3.00ms	4.13ms	4.94ms	7.99ms
DataLoader	2.92ms	4.83ms	24.78ms	5.31ms	7.20ms	8.97ms	12.38ms

Baseline

Test	Min	Med	Max	Avg	P90	P95	P99
Single Fetch	1.36ms	1.75ms	41.90ms	1.99ms	2.58ms	3.06ms	5.83ms
DataLoader	2.63ms	3.34ms	16.53ms	3.70ms	4.80ms	5.90ms	9.21ms

Change vs Baseline

Test	Min	Med	Max	Avg	P90	P95	P99
Single Fetch	⚠️ (6.43% worse)	⚠️ (57.28% worse)	⚠️ (38.79% worse)	⚠️ (50.97% worse)	⚠️ (59.95% worse)	⚠️ (61.57% worse)	⚠️ (37.10% worse)
DataLoader	⚠️ (11.24% worse)	⚠️ (44.82% worse)	⚠️ (49.88% worse)	⚠️ (43.55% worse)	⚠️ (50.20% worse)	⚠️ (52.10% worse)	⚠️ (34.43% worse)

⚡ Throughput

Test	Metric	Current	Baseline	Change
Single Fetch	Requests/sec	78.71 req/s	78.80 req/s	✅ (0.11% better)
DataLoader	Requests/sec	78.41 req/s	78.60 req/s	✅ (0.24% better)

🎯 Reliability

Test	Error Rate
Single Fetch	0.00% ✅
DataLoader	0.00% ✅

🔍 Analysis

⚠️ Performance regression detected. Response times have increased significantly compared to baseline.

---

Run 18754333492 • Commit 5e68c80 • Thu, 23 Oct 2025 16:02:26 GMT