- Routing: expressions.rs -> expr_parse_match()
- Module: src/parser/expr/match_expr.rs (+OR pattern, comma/newline separators)
- No MIR changes; existing PeekExpr lower handles match result via decision tree
- optimizer: remove unreachable old bodies in normalize_legacy_instructions/normalize_ref_field_access and delegate to optimizer_passes only
- parser/box: route public/private single-line fields through members::fields (supports = init, => computed, {..}+postfix) and keep visibility vectors consistent
- box_def facade: silence dead_code warning for scaffold
Smokes:
- PyVM Stage-2 PASS
- Stage-2 short-circuit PASS
- PyVM collections PASS
(NOTE) using_e2e_smoke is optional and currently failing; unrelated to this change
Major updates:
- Updated progress section (2025-09-14) with recent achievements:
- Python LLVM backend reaching production level (all tests passing)
- Nyash parser implementation started by ChatGPT5
- peek expression rediscovery (when→peek rename)
- Box theory simplifying SSA construction (650→100 lines)
- AI collaboration paper completed
Documentation improvements:
- Changed 500-line limit to "guideline" (not strict requirement)
- Upgraded Python LLVM backend from "experimental" to "production level"
- Added peek expression examples and usage patterns
- Updated todo list status (parser planning completed, compiler MVP in progress)
These updates reflect the significant progress towards self-hosting,
especially the parser implementation using peek expressions to avoid
19-line if-else nesting.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Parser improvements:
- Added expression statement fallback in parse_statement() for flexible syntax
- Fixed ternary operator to use PeekExpr instead of If AST (better lowering)
- Added peek_token() check to avoid ?/?: operator conflicts
LLVM Python improvements:
- Added optional ESC_JSON_FIX environment flag for string concatenation
- Improved PHI generation with better default handling
- Enhanced substring tracking for esc_json pattern
Documentation updates:
- Updated language guide with peek expression examples
- Added box theory diagrams to Phase 15 planning
- Clarified peek vs when syntax differences
These changes enable cleaner parser implementation for self-hosting,
especially for handling digit conversion with peek expressions instead
of 19-line if-else chains.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Tokenizer: add tokens for << >> & | ^ (SHIFT_LEFT/RIGHT, BIT_AND/OR/XOR); keep ||, &&, |>
- Parser: precedence layers for bit ops (| ^ &) and shift (<< >>); remove >> Arrow production
- AST: add BitAnd/BitOr/BitXor/Shr with Display
- MIR builder: map bit ops and Shr to MIR BinaryOp
- Interpreter: implement integer-only bit ops and shifts, mask shift count to 0..63
Compatibility: legacy >> Arrow removed from parser; use |> for pipeline.
- Tokenizer: add SHIFT_LEFT for '<<' (before <= detection)
- Parser: introduce parse_shift() layer; left-associative; only '<<' handled in Phase 1
- AST: add BinaryOperator::Shl and display
- MIR builder: map Shl -> BinaryOp::Shl
- Interpreter: implement Shl in both execution paths (integer-only, type error otherwise)
No change to '>>' (legacy ARROW remains). No runtime gate yet for SHL as it does not collide with existing syntax.
## Major Features Added
### Field Visibility System
- Added `private { ... }` and `public { ... }` blocks in box declarations
- Default visibility is now handled explicitly (fields must be in either block)
- Visibility checks enforced at both interpreter and VM levels
### Parser Enhancements
- Extended AST with public_fields and private_fields vectors
- Added parsing for visibility blocks in box definitions
- Maintained backward compatibility with existing `init { ... }` syntax
### Interpreter Implementation
- Added visibility checks in field access (get_field/set_field)
- External access to private fields now throws appropriate errors
- Internal access (within methods) always allowed
### VM Implementation
- Extended VM with object_class tracking for visibility checks
- RefGet/RefSet instructions now enforce field visibility
- Fixed nested box declaration collection (boxes defined inside methods)
### Test Examples Added
- docs/examples/visibility_ok.nyash - demonstrates correct usage
- docs/examples/visibility_error.nyash - tests private field access errors
## Technical Details
### Error Messages
- Interpreter: "Field 'X' is private in Y"
- VM: Same error message for consistency
### Current Limitations
- All RefGet/RefSet treated as external access in VM (internal flag future work)
- Legacy `init { ... }` fields treated as having unspecified visibility
## Test Results
✅ Interpreter: Both test cases pass correctly
✅ VM: Both test cases pass correctly after nested declaration fix
This implements the foundation for proper encapsulation in Nyash,
following the "explicit is better than implicit" philosophy.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Create items module with global_vars, functions, static_items
- Move parse_global_var from mod.rs to items/global_vars.rs
- Move parse_function_declaration to items/functions.rs
- Move parse_static_declaration and parse_static_function to items/static_items.rs
- Clean up mod.rs by removing 197 lines of code
- Build passes successfully
Next: Step 5 - Final cleanup and documentation
Successfully extracted utility methods from parser/mod.rs to common.rs:
- Created ParserUtils trait with token manipulation methods
- Extracted current_token, peek_token, advance, skip_newlines, etc.
- All parser modules now use trait-based utilities
- Reduced mod.rs by ~60 lines
✅ Build successful with no errors
✅ All functionality preserved - tested with using nyashstd
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
