🗄️ rullst-orm

ORM Local & Comparative Benchmarks

Pushes to main run Criterion against selected Rullst paths and equivalent Rullst, Diesel, and SeaORM SQLite operations. Results describe the exact commit, runner, and benchmark input; they are not production database latency claims.

🏠 Home ⚡ Core Benchmarks 📊 CI History
🔨

Query Builder

Measures construction of a builder using where_eq, order_by, limit, and offset. The suite does not currently measure allocations.

🗺️

Model JSON

Measures generated to_json and from_json conversion for the benchmark model.

🏗️

Identifier Validation

Measures valid, qualified, and rejected structural identifier inputs used by schema/query safety checks.

🔐

SQLite Round Trips

Measures insert/delete and several read shapes locally. The comparative target gives the three pinned ORMs one typed connection and equivalent schema, index, seed, and SQLite policy.

📝 What's Being Measured

// The current CPU benchmark constructs this query-builder value. #[derive(Debug, FromRow, Orm)] #[orm(table = "users")] struct User { id: i32, name: String, email: String } // bench_query_builder: builds SQL without executing — pure generation cost let sql = User::query() .where_eq("email", "alice@example.com") .order_by("name") .limit(10) .offset(20);
⏳ Loading ORM benchmark data...