This experimental package demonstrates compile-time chain composition (StaticChain) as outlined in ../cpp/CHAIN_PERFORMANCE_OPTIMIZATION.md.

• Show conceptual and empirical difference between:
  1. Direct inline function pipeline
  2. StaticChain (compile-time tuple of operations)
  3. Dynamic Chain with virtual dispatch + coroutine awaitable

• include/codeuchain_opt/static_chain.hpp: Header-only StaticChain<Ops...> template and example stateless ops.

cd packages/cpp/build   # assuming core library already configured
cmake --build . --target static_chain_demo -j
./packages/cpp_opt/static_chain_demo

If not yet configured, do:

cd packages/cpp
mkdir -p build && cd build
cmake -DCMAKE_BUILD_TYPE=Release ../..  # top-level if aggregated, else adjust
cmake --build . --target static_chain_demo -j

StaticChain Demo (ns/op)
 direct    : <baseline arithmetic>
 static    : <near-baseline if state overhead dominates>
 dynamic   : <higher due to virtual + coroutine + map copies>

static should be closer to direct than dynamic. Remaining gap is dominated by state mutation + variant access.

For guidance on when to apply advanced optimizations (StaticChain vs dynamic chain, mutability, slot caching, hybrid state) versus leaving code in the default dynamic form, see:

OPTIMIZATION_DECISION_GUIDE.md

Highlights:

• Do NOT optimize unless profiling shows micro-scale hot spots.
• Prefer dynamic chains for clarity and observability.
• Escalate to StaticChain + mutability + slot caching only in high-frequency trivial workloads.
• HybridState + interning (planned) targets memory + lookup churn for further reductions.

This README remains focused on the prototype mechanics; the decision guide captures strategy.

• Hybrid state prototype (HybridState) with inline storage
• Key interning & slot caching toggles
• JSON metrics export integration with main benchmark harness

Prototype code intended for design exploration; APIs may change or be removed.