SolvraScript

/* The command-line entrypoint and standard library of the Solvra scripting layer. */

SolvraScript follows a disciplined artifact pipeline that keeps the runtime deterministic while allowing tiered performance:

1. .svs — Source authoring format. Write your scripts here.
2. solvra_compile converts .svs into .svc portable bytecode that the Tier-0 interpreter consumes; this is the deterministic artifact that is shared across Hive nodes and caches.
3. .svc modules load through solvra_core (Tier-0) and can be rematerialized into higher tiers dynamically.
4. solvra_aot compiles .svs/.svc into .saot native binaries for Tier-3 execution when zero runtime compilation and native throughput are required; .saot files are CPU/ABI specific and are not distributed like .svc.

# Compile to portable bytecode
cargo run -p solvra_core --bin solvra_compile -- script.svs script.svc

# Run the `.svc` bytecode with the Tier-0 interpreter
cargo run -p solvra_core -- script.svc --tier=0

# Optionally produce an AOT binary for native execution
cargo run -p solvra_core --bin solvra_aot -- compile script.svs -o script.saot -O2
cargo run -p solvra_core -- script.saot

This pipeline keeps the interpreter deterministic (.svc), while the optional .saot binaries deliver maximum performance when local execution demands native code.