Capstone: Source to Bytecode to VM

Why this module exists

This capstone connects every stage:

1. Lex source into tokens
2. Parse tokens into a program AST
3. Run semantic analysis
4. Compile to bytecode
5. Encode bytecode to a file format
6. Decode it back
7. Execute in the VM

That is the full pipeline of a real compiler + runtime.

1) Honest compilation

The key idea is separation:

• The compiler produces bytecode
• The VM consumes bytecode

They are separate components with a clear contract. This is what makes it "honest."

2) Error handling

Each stage can fail:

• Lex and parse errors
• Semantic errors (undefined names, const misuse, arity mismatch)
• Runtime errors (type errors, division by zero)

The capstone should return the first error it encounters.

3) Open-ended next steps

From here, you can extend:

• Closures and captured variables
• Bytecode optimizations
• A binary bytecode format
• A simple JIT