tomoaki/hakorune
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,235 Commits 1 Branch 0 Tags
f9d100ce013c0a64a4b2bf1ad9e1fbf7dfc6affc
Commit Graph

243 Commits

Author SHA1 Message Date
Moe Charm
af32896574 fix: Fix NyashParser import path and improve BoxCall fallback 
- Fix NyashParser import path in vm.rs tests
- Improve BoxCall fallback logic for plugin/builtin methods
- Add proper function existence checks before lowering to Call

This ensures plugin Box methods correctly fall back to BoxCall
when the corresponding user-defined function doesn't exist.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-20 23:07:16 +09:00
Moe Charm
b3ac83de36 feat(phase-9.78b): ChatGPT5 VM unified Box handling + MIR parameter fix 
Phase 9.78b Step 1-2完了 + ChatGPT5による修正:
- ✅ MIRパラメータ解決修正（ValueId reset）
- ✅ VMでExternCall/Call実装
- ✅ プラグインローダーv2統合
- ✅ 3種類のBox完全動作（UserDefined/Builtin/Plugin）
- ✅ VM E2Eテスト成功

次期作業:
- Phase 9.78b Step 3: BoxFactory dyn化
- Phase 9.78b Step 4以降: アーキテクチャ改善

Co-authored-by: ChatGPT5 <noreply@openai.com>
 2025-08-20 20:01:55 +09:00
Moe Charm
8aef29eaef 🎉 feat: MIRパラメータ解決完全修正 + VM E2Eテスト成功！ 
Phase 9.78b関連修正:
- MirBuilder::lower_method_as_function でValueIdリセット実装
  - me → %0, 引数 → %1... の正しいマッピング
- build_me_expression() で変数テーブル優先解決
- VM E2Eテスト test_vm_user_box_birth_and_method 成功
  - new Person("Alice").greet() → "Hello, Alice"

ドキュメント:
- mir-unified-reference.md を正式MIRドキュメントとして配置
  - 現在の35命令実装状態を正確に記載
  - RefGet/RefSet仕様含む

ChatGPT5によるMIRビルダー修正で、VMバックエンドでのユーザー定義Box完全動作！
 2025-08-20 19:34:09 +09:00
Moe Charm
41832b2c88 🚀 feat: VM ExternCall実装 - ChatGPT5による plugin loader v2 統合 
Phase 9.78b の続き:
- VM::execute_instruction に ExternCall 完全実装
- plugin loader v2 経由でenv.console.log等を処理
- MirBuilder::build_method_call に最適化追加（new Class().method() → 直接Call）
- extern_call メソッドを plugin_loader_v2 に追加

ChatGPT5によるVMとプラグインシステムの統合作業
 2025-08-20 19:10:30 +09:00
Moe Charm
41361a2f50 🚀 feat: Phase 9.78b Step 1&2完了 - ChatGPT5による実装 
ChatGPT5が実装したPhase 9.78b Step 1&2の統合

## 🎯 実装内容
1. Phase 9.78b Step 1: BoxDeclarationをcore::modelへ移動
   - src/core/mod.rs, model.rs 新規作成
   - BoxDeclarationを純粋データモデルとして分離

2. Phase 9.78b Step 2: NyashRuntime骨組み作成
   - src/runtime/nyash_runtime.rs 追加
   - 統一Box管理の基盤

3. ビルドエラー修正
   - Arc重複インポート修正
   - as_str() → as_ref() 変更
   - parent_name.to_string() 型変換
   - インポートパス調整

## 📊 ビルド結果
- ✅ フルビルド成功 (47.34秒)
- ✅ ユニットテスト: 145/145成功
- ✅ 統合テスト: 16/16成功
- ✅ WASMビルド成功 (1.9MB)
- ❌ MIRテスト: 1失敗 (ref_new命令)

## 🚀 次のステップ
- Phase 9.78b Step 3: BoxFactory dyn化
- Codexの設計に基づく段階的実装継続
 2025-08-20 18:57:10 +09:00
Moe Charm
86b9f7719b 🏗️ feat: Phase 9.78b準備 - インタープリター・VM統合アーキテクチャ設計 
Phase 9.78b: Codexの天才的分析に基づくアーキテクチャ再設計準備

## 📋 実施内容
1. Codex分析結果のアーカイブ
   - 実装詳細共有 → モデル共有・実行時共有への転換提案
   - 8ステップの段階的実装計画

2. Phase 9.78a作業の保存
   - MIR生成でのNewBox命令統一（保持）
   - ScopeTracker基本実装（一時コメントアウト）
   - VM拡張の方向性（TODOコメント付き）

3. ビルドエラー修正
   - ScopeTrackerインポート問題を一時的に解決
   - ビルド成功（警告のみ）

## 📚 作成ドキュメント
- architecture-redesign-proposal.md - Codexの設計提案
- phase_9_78b_interpreter_architecture_refactoring.md - 実装計画
- codex-analysis/* - 分析結果アーカイブ

## 🎯 次のステップ
Phase 9.78b Step 1: BoxDeclarationをcore::modelへ移動
 2025-08-20 17:58:51 +09:00
Moe Charm
bab57e7c07 🎉 feat: ChatGPT5による奇跡の全テスト修復完了！ 
- instance_v2移行で破綻していた440個のテストを30分で全修正
- git巻き戻しレベルの状況から完全復活
- 人間には不可能な速度での大規模整合性修正

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-20 04:45:26 +09:00
Moe Charm
5b6cf828af feat: Phase 9.78b完了 - UnifiedBoxRegistry統合とビルド最適化 
🏭 **Phase 9.78b: 統合レジストリ実装完了**

**主要な変更:**
- execute_new()に統合レジストリを統合、レガシーmatch文への自動フォールバック
- 全importエラー解決（RuntimeError, モジュールパス修正）
- runner.rs起動時に統合レジストリ初期化
- 20+種類のビルトインBoxが統合ファクトリ経由で作成可能

**ビルド時間最適化:**
- wasmtime/wabt依存を"wasm-backend"フィーチャーでオプション化
- デフォルトビルド 4分 → 43秒の劇的高速化達成

**技術的達成:**
- 600+行match文 → 30行ファクトリパターンへの移行基盤完成
- プラグイン・ユーザー定義・ビルトインBox統一アーキテクチャ確立
- Everything is Box哲学の実装レベル体現

**次のステップ:**
Phase 9.78c: プラグインBox統合、Phase 9.78d: ユーザー定義Box統合

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-19 16:56:44 +09:00
Moe Charm
840c1b85ef feat: Fix VM SSA loop execution with proper phi node handling 
Fixed infinite loop issue in VM by addressing phi node caching problem.
The phi node was caching the initial value and returning it for all
subsequent iterations, preventing loop variable updates.

Changes:
- Created vm_phi.rs module to separate loop execution logic (similar to mir/loop_builder.rs)
- Disabled phi node caching to ensure correct value selection each iteration
- Added LoopExecutor to track block transitions and handle phi nodes properly
- Fixed VM to correctly track previous_block for phi input selection

The VM now correctly executes SSA-form loops with proper variable updates:
- Loop counter increments correctly
- Phi nodes select the right input based on control flow
- Test case now completes successfully (i=1,2,3,4)

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-18 23:36:40 +09:00
Moe Charm
9d78ecd089 feat(phase-9.78): Improve MIR generation for built-in Box types and enhance LLVM mock 
- Fix MIR builder to handle built-in Box types (IntegerBox, StringBox, etc.) specially
- Generate MIR with actual constant values instead of ref_new instructions
- Enhance LLVM mock compiler with MIR interpreter foundation
- Add value storage HashMap for future MIR instruction interpretation

This enables proper MIR generation where 'new IntegerBox(42)' becomes '%0 = const 42'
instead of '%0 = ref_new "IntegerBox"'. This is essential for future LLVM code generation.

Example MIR improvement:
Before:
  %1 = const "IntegerBox"
  %0 = ref_new %1

After:
  %0 = const 42

🤖 Generated with Claude Code

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-18 19:29:24 +09:00
copilot-swe-agent[bot]
313ad2a46d Complete LLVM PoC mock implementation - demonstrates integration structure 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-18 09:25:57 +00:00
copilot-swe-agent[bot]
e74a9f621e Implement LLVM backend basic structure - Phase 9.78 Week 1 foundation 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-18 09:14:39 +00:00
copilot-swe-agent[bot]
fa1f188187 Phase 3 complete: Memory optimization - VM now 22.80x faster than interpreter 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-18 07:44:34 +00:00
copilot-swe-agent[bot]
230bd7a1d2 Phase 1 complete: Remove debug overhead - VM now 18.84x faster than interpreter 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-18 07:33:49 +00:00
Moe Charm
c6c3c8e2f9 fix(bid-ffi): Fix HostVtable lifetime issue causing segfault 
🚨 Critical memory safety fix:
- HostVtable was created on stack and destroyed after init
- Plugin stored reference to destroyed memory → NULL pointer access
- Changed to static LazyLock storage for lifetime safety

✅ Results:
- Segfault completely eliminated
- Plugin logging now works properly
- Type info system confirmed working
- Full E2E FileBox plugin operation successful

🔧 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-18 14:10:41 +09:00
Moe Charm
bec0e9bc92 revert: 古いプラグインシステム実装前の状態に巻き戻し 
- ソースコードをcommit 3f7d71fの状態に復元（古いプラグインシステム実装前）
- docsフォルダは最新の状態を維持（BID-FFI設計ドキュメント含む）
- nyashバイナリの基本動作確認済み
- BID-FFIシステムをクリーンに再実装する準備完了

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-18 08:34:19 +09:00
Moe Charm
5b7ef32277 feat(phase-5.2): Complete backend cleanup - reject deprecated MIR instructions 
Phase 5.2 of MIR 35→26 reduction project:
- VM backend: Reject all 17 deprecated instructions with proper error messages
- WASM backend: Reject deprecated instructions with migration guidance
- All deprecated instructions now properly handled in both backends
- Build successful with expected deprecation warnings

Deprecated instructions removed:
- UnaryOp → Use BinOp
- Load/Store → Use BoxFieldLoad/BoxFieldStore
- ArrayGet/ArraySet → Use BoxFieldLoad/BoxFieldStore or intrinsics
- Print/Debug → Use Call @print/@debug
- TypeCheck/Cast → Use Call @type_check/@cast
- Throw/Catch → Use Call @throw/@catch
- RefNew → References handled implicitly
- BarrierRead/BarrierWrite → Use AtomicFence
- FutureNew/FutureSet/Await → Use NewBox/BoxCall
- Copy/Nop → Not needed in final IR
 2025-08-17 14:06:47 +09:00
Moe Charm
3df87fb1ce fix(phase-4.3c-3): Fix StringBox literal handling in MIR builder 
Phase 4-3c-3 Complete: WASM host functions now correctly output string content

## Changes:
- Fixed MIR builder to handle StringBox with string literal arguments
- Special case for  to generate proper string constants
- Removed debug output after successful verification
- WASM now correctly outputs "Hello MIR!" instead of "StringBox"

## Test Results:
- MIR generation: ✅ Generates  correctly
- WASM compilation: ✅ String data correctly placed at offset 4096
- WASM execution: ✅ Outputs "Hello MIR\!" as expected

## Technical Details:
- Modified build_new_expression() to detect StringBox with literal arguments
- Generates Const instruction with actual string content
- Host function reads StringBox memory layout correctly

This completes the WASM string output functionality for Phase 4.

🤖 Generated with Claude Code
Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-17 13:49:35 +09:00
Moe Charm
bb3f2e8032 feat: Complete Phase 3 of MIR 35→26 reduction - optimization pass migration 
- Remove old instructions from VM/WASM backends (UnaryOp, Print, Load/Store, RefGet/RefSet)
- Add comprehensive MIR optimizer with Effect System based optimizations
- Implement dead code elimination, CSE, pure instruction reordering
- Add intrinsic function support in VM backend
- Update backends to use new BoxFieldLoad/Store and Call intrinsics

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-17 12:27:12 +09:00
Moe Charm
cde961defc feat(phase-9.77): Implement BoxCall instructions and fix wasmtime version 
Phase 9.77 WASM Emergency Recovery Progress:
- ✅ Task 1.1: Implement BoxCall instruction for toString(), print(), equals(), clone(), log()
- ✅ Task 1.2: Update wasmtime 18.0 → 35.0.0 and add runtime imports
- 🔄 Task 1.3: Working on UTF-8 encoding error fix

Changes:
- Add generate_box_call() method in codegen.rs with 5 helper methods
- Update wasmtime dependency to 35.0.0 for AOT compatibility
- Add BoxCall runtime imports (box_to_string, box_print, box_equals, box_clone)
- Implement wat_to_wasm() with UTF-8 validation and debug output
- Update CURRENT_TASK.md with Copilot handoff notes

Current issue: 'Generated WASM is not valid UTF-8' error source unknown
Next: Copilot to investigate error origin and complete Task 1.3

🤖 Generated with Claude Code

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-16 21:21:51 +09:00
Moe Charm
fa1a3ad644 feat(Phase 9.75j): Complete warning elimination - 106→0 warnings (100% improvement) 
✨ Major code quality improvements:
• Fixed all unused variable and parameter warnings
• Added appropriate #[allow(dead_code)] annotations
• Renamed constants to follow Rust naming conventions
• Achieved completely warning-free codebase

🔧 Key changes:
• Parser expressions: Fixed arg_count and statement_count usage
• MIR modules: Added dead_code allows for future-use code
• Backend modules: Prefixed unused parameters with underscore
• Effect constants: Renamed 'read' to 'READ_ALIAS' for conventions

📊 Results:
• Before: 106 warnings (noisy build output)
• After: 0 warnings (clean build)
• Improvement: 100% warning reduction

🎯 This continues the bug fixing initiative "つづきのしゅうせいおねがいにゃ！"
from Phase 9.75i (birth() fixes, static methods, Copilot apps).

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-16 17:39:04 +09:00
copilot-swe-agent[bot]
5bcaa14b52 Phase 9.75-B: Convert StreamBox from Arc<Mutex> to RwLock, progress on DebugBox 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 23:59:11 +00:00
copilot-swe-agent[bot]
2091462441 Implement Phase 9.7: ExternCall instruction and WASM runtime imports 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 08:56:39 +00:00
copilot-swe-agent[bot]
85a5ce053d Fix WASM Jump/Branch implementation and HTTPServer listen() functionality 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 07:14:45 +00:00
copilot-swe-agent[bot]
626572a012 Fix WASM import expectations and validate complete implementation 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 05:35:17 +00:00
copilot-swe-agent[bot]
c1ae84ce61 Implement basic WASM string constant support 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 05:31:36 +00:00
copilot-swe-agent[bot]
db9bcd3b45 Implement AOT backend infrastructure with CLI integration 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 04:38:20 +00:00
copilot-swe-agent[bot]
bf51d3c71b Expand VM BoxCall support - add StringBox substring/concat and ArrayBox methods 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 03:46:06 +00:00
copilot-swe-agent[bot]
d26b365791 Fix VM BoxCall implementation - now returns actual method results 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-14 03:39:08 +00:00
Moe Charm
c7365d3dc9 feat(merge): integrate Phase 8.3 WASM Box Operations + Benchmark System 
🎉 Successful merge of Copilot and Claude implementations:

**Copilot Contributions (Phase 8.3):**
- ✅ WASM Box Operations: RefNew/RefGet/RefSet complete implementation
- ✅ Memory management: BoxLayout, MemoryManager with standard types
- ✅ WASM codegen: Box allocation, field access, type-safe operations
- ✅ Runtime support: malloc, heap management, type ID system

**Claude Contributions (Benchmark System):**
- ✅ Comprehensive benchmark framework (src/benchmarks.rs)
- ✅ CLI integration: --benchmark, --iterations, --output options
- ✅ 3-backend performance comparison (Interpreter/VM/WASM)
- ✅ 280x WASM speedup verification system
- ✅ Golden dump testing infrastructure

**Unified Features:**
- 🔧 execute_wasm_mode: Supports both output file and stdout
- 🔧 CLI arguments: All options preserved and functional
- 🔧 Error handling: Improved MIR verification messages
- 🔧 Build system: All modules properly integrated

**Next Steps Ready:**
- 📊 MIR diet planning (35→20 instructions)
- 🚀 Phase 8.4: AOT WASM native compilation
- 🧪 Golden dump automation

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-14 08:12:36 +09:00
copilot-swe-agent[bot]
a0a97620b8 Complete Phase 8.3 WASM Box Operations - Everything is Box in WASM 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 22:20:50 +00:00
Moe Charm
8ec80a35c3 feat(benchmark): add comprehensive performance benchmarking system 
🚀 Phase 8.2 PoC2 Achievement: 280x WASM performance boost proven\!

## New Features:
- Complete benchmark framework (src/benchmarks.rs)
- CLI integration: --benchmark --iterations options
- 3-backend comparison: Interpreter/VM/WASM
- Automated performance measurement & reporting

## Benchmark Results (100 iterations average):
- WASM: 0.17ms (280x faster than Interpreter\!)
- VM: 16.97ms (2.9x faster than Interpreter)
- Interpreter: 48.59ms (baseline)

## Added Files:
- benchmarks/bench_{light,medium,heavy}.nyash - Test cases
- benchmark_summary_20250814.md - Clean results
- wasm_demo/ - Browser execution environment

## Documentation Updates:
- docs/execution-backends.md - Added benchmark usage
- docs/CURRENT_TASK.md - Phase 8.3 Copilot coordination
- CLAUDE.md - Quick benchmark access

## Copilot Integration Ready:
- Phase 8.3 merge conflict avoidance strategy documented
- Benchmark framework ready for Box operation performance validation
- CLI integration preserved for future enhancements

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-08-14 07:19:23 +09:00
copilot-swe-agent[bot]
8a7532a77b 🎉 Phase 8.2 PoC1 COMPLETE: MIR→WASM basic arithmetic working end-to-end! 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 13:19:43 +00:00
copilot-swe-agent[bot]
3335f26812 Phase 8.2 PoC1 Progress: WASM codegen mostly working, debugging Return instruction issue 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 13:15:02 +00:00
copilot-swe-agent[bot]
8276eeeb64 Phase 8.1 Complete: WASM backend foundation implementation 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 13:05:49 +00:00
copilot-swe-agent[bot]
28070ea2f2 Fix VM execution bug - include terminator instructions in execution loop 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 12:18:47 +00:00
copilot-swe-agent[bot]
cefac9051e Phase 7.4: Complete async MIR implementation with comprehensive tests 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 11:59:22 +00:00
copilot-swe-agent[bot]
b59f0ee419 Phase 7.1: Add MIR instructions for async operations (FutureNew, FutureSet, Await) 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 11:46:01 +00:00
copilot-swe-agent[bot]
9adb34ff58 Complete Phase 6.1: AST lowering for New/FieldAccess → MIR Ref ops with tests and VM field storage 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 10:26:46 +00:00
copilot-swe-agent[bot]
af3ad5e5e2 Implement core AST lowering for New/FieldAccess to MIR Ref ops with basic VM field storage 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 10:20:37 +00:00
copilot-swe-agent[bot]
84d2aac8da Implement Phase 6 Box reference operations in MIR/VM - RefNew/RefGet/RefSet/WeakNew/WeakLoad/Barrier* 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 09:45:22 +00:00
copilot-swe-agent[bot]
d3a85b2305 Implement Phase 5: Control flow & exceptions in MIR/VM - Core functionality complete 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 06:23:28 +00:00
copilot-swe-agent[bot]
2f3b453fcb Implement Phase 1: MIR + VM backend with golden tests 
Co-authored-by: moe-charm <217100418+moe-charm@users.noreply.github.com>
 2025-08-13 05:59:10 +00:00