Adding Error Codes

Overview

Vole uses structured error codes: E0xxx (lexer), E1xxx (parser), E2xxx (semantic).

Checklist

1. •

   Get next available code:

   bash

   just dev-next-error sema    # For semantic errors (E2xxx)
   just dev-next-error parser  # For parser errors (E1xxx)
   just dev-next-error lexer   # For lexer errors (E0xxx)
   

2. •

   List existing errors (for reference):

   bash

   just dev-list-errors sema
   just dev-list-errors parser
   

3. •

   Add error variant to the appropriate file:

   Error Type	File
   Lexer	src/errors/lexer.rs
   Parser	src/errors/parser.rs
   Semantic	src/errors/sema.rs

4. •

   Emit the error in the relevant analyzer/parser code:

   Error Type	Emission Location
   Lexer	src/frontend/lexer.rs
   Parser	src/frontend/parser/*.rs
   Semantic (expr)	src/sema/analyzer/expr.rs
   Semantic (stmt)	src/sema/analyzer/stmt.rs
   Semantic (patterns)	src/sema/analyzer/patterns.rs

5. •

   Add snapshot test for error message:

   • •Create test/snapshot/check/your_error.vole with code that triggers the error
   • •Run cargo run --bin vole-snap -- bless test/snapshot/check/your_error.vole
6. •

   Verify:

   bash

   just ci
   

Error Code Format

// In errors/sema.rs (example)
E2061 => "cannot use `break` outside of a loop",

Diagnostic Builder

Use the diagnostic builder for rich error messages:

self.error(SemaError::YourError)
    .span(span)
    .label(span, "explanation here")
    .emit();