NanoLang User Guide#

I am NanoLang. I use explicit boundaries, prefix calls, equal-precedence operators, and tests beside the functions they exercise. I compile to C and to verified NanoISA. I also host a secure runtime: service contracts, unforgeable capabilities, a POSIX fabric, and a trap journal. I do not claim a kernel.

fn factorial(n: int) -> int {
    if (<= n 1) {
        return 1
    }
    return (* n (factorial (- n 1)))
}

shadow factorial {
    assert (== (factorial 0) 1)
    assert (== (factorial 5) 120)
}

Start Here#

1. Build me and run a first program.
2. Learn my language.
3. Use structured data and typed errors.
4. Work with modules, foreign code, and resources.
5. Understand shadows, tests, and my verified boundary.
6. Choose a tool or backend.
7. Measure native performance and tune from evidence.
8. Read the secure runtime: NSI, capabilities, fabric, policy, journal.

Reference#

• Examples are generated from every .nano file under examples/.
• Builtins come from the mechanically checked standard-library reference.
• Modules are generated from the module tree and manifests.
• Compiler CLI comes from the compiler used to build this guide.
• NanoISA is my shared typed VM boundary.

What I Promise#

• Function calls use (function argument) syntax.
• Function parameters and return types are explicit.
• Local bindings may use inference when the initializer makes the type plain.
• All infix operators have equal precedence and associate left to right.
• Project policy requires useful shadows for changed named functions. Compiler enforcement has documented exemptions.
• A passing shadow is a test, not a proof.
• My C backend is the production compilation path. Other backends have narrower documented subsets.

The parser, typechecker, builtin registry, and tests are the authority when an old document disagrees. I would rather correct a guide than preserve a confident mistake.