Commit Graph

501 Commits

Author SHA1 Message Date
d673bd4fbd docs(phase29aj): isolate phase286 pattern9 legacy smoke 2025-12-29 14:18:31 +09:00
57e0fdb0a0 phase29aj(p8): planner-first pattern9 accum const loop subset 2025-12-29 14:08:53 +09:00
5ba77cb62a phase29aj(p7): planner-first pattern8 bool predicate scan subset 2025-12-29 13:57:26 +09:00
d819fe0dae docs(phase29aj): define joinir regression gate; isolate phase143 legacy 2025-12-29 13:44:53 +09:00
d4a9969fa3 phase29aj(p5): planner-first pattern5 infinite early-exit subset 2025-12-29 13:39:30 +09:00
3a7378a747 phase29aj(p4): planner-first pattern4 continue subset 2025-12-29 13:22:55 +09:00
31e4ac8c99 phase29aj(p3): planner-first pattern3 if-phi facts subset 2025-12-29 13:06:17 +09:00
97d5d3ef77 phase29aj(p2): planner-first pattern1 simplewhile subset 2025-12-29 12:54:32 +09:00
ff3af90b4c phase29aj(p1): remove legacy_rules via plan extractors 2025-12-29 12:39:15 +09:00
5988374ecf phase29aj(p0): expose planner outcome facts for strict observability 2025-12-29 12:17:57 +09:00
2e1bb938b4 phase29ai(p15): observe pattern2 promotion hint under strict 2025-12-29 11:37:31 +09:00
19f2c6b7f6 phase29af(p0): pattern2 boundary hygiene ssot 2025-12-29 05:12:15 +09:00
a04b48416e fix(joinir): Phase 287 P2 - Pattern6 nested loop latch overwrite fix
Fix infinite loop in Pattern6 (nested loop minimal) caused by main→loop_step
overwriting k_inner_exit→loop_step latch values.

Root cause: JoinIR main entry block was incorrectly treated as BackEdge,
causing it to overwrite the correct latch incoming values set by the true
back edge (k_inner_exit → loop_step).

Solution:
- Restrict latch recording to TailCallKind::BackEdge only
- Treat only MAIN's entry block as entry-like (not loop_step's entry block)
- Add debug_assert! to detect double latch set in future

Refactoring:
- Extract latch recording to latch_incoming_recorder module (SSOT)
- Add boundary.loop_header_func_name for explicit header identification
- Strengthen tail_call_classifier with is_source_entry_like parameter

Tests: apps/tests/phase1883_nested_minimal.hako → RC:9 (was infinite loop)
Smoke: 154/154 PASS, no regressions

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-27 09:39:29 +09:00
1d24e9a106 feat(joinir): Phase 286 P2.3 + Phase 287 - Pattern9 Plan化 + Router table-driven
## Phase 286 P2.3: Pattern9 AccumConstLoop Plan化 PoC

- DomainPlan::Pattern9AccumConstLoop 追加
- PlanNormalizer::normalize_pattern9_accum_const_loop() 実装
  - PHI 2本(loop_var, acc_var)
  - const/var 両方 OK(sum = sum + 1 または sum = sum + i)
- Pattern9 は Pattern1 より優先(より具体的なパターン)
- Integration test: phase286_pattern9_frag_poc PASS (return: 3)
- Regression: quick 154 PASS

## Phase 287: Router table-driven Plan extraction

- PLAN_EXTRACTORS static table で Pattern6/7/4/9/1 を統一管理
- PlanExtractorEntry/PlanExtractorVariant 構造体追加
- try_plan_extractors() で ~100行 → 3行に集約
- メンテナンス性向上(新 Pattern 追加はテーブル1行追加のみ)

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-26 02:38:09 +09:00
6656098c95 docs(phase-286): P2.2 完了 + P2.3 準備中の進捗記載
更新内容:
- Phase 286 README に P2.2  COMPLETE セクション追加
  - extractor helper化(extract_loop_increment_plan 統一)の内容
  - router helper化(lower_via_plan() 追加)の内容
  - 削減行数(~65行)と検証結果(quick 154 PASS, PoC 両方 PASS)

- 10-Now.md の Current Focus を Phase 286 P2.3 に更新
  - P2/P2.1/P2.2 の完了を時系列記載
  - Pattern9 (AccumConstLoop) を次の PoC 対象に決定

- CURRENT_TASK.md の Handoff(現状)を更新
  - P2.2 COMPLETE の記載(hygiene 内容 + 成果)
  - P2.3 準備中(Pattern9 PoC 準備)の1段落を追加

- pattern2-deferred.md 新規作成
  - Pattern2 の複雑性(break値再接続)を説明
  - 再開条件(after_bb PHI + compose::loop_ + Break wiring)を明文化
  - Phase 286 P2.2 として後続タスク化

関連 issues: Phase 286 JoinIR Line Absorption

🤖 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
2025-12-26 02:03:34 +09:00
55d30c9845 docs(repl): record Phase 288.1 contract and results
更新内容:
- docs/reference/language/repl.md
  - Phase 288.1 完了ステータスに更新
  - Implementation Architecture セクション追加(AST rewrite 方式説明)
  - AST Rewriter の動作フロー追加(Variable/Assignment 変換ロジック)
  - ExternCall Bridge の仕組み追加(__repl.get/set → VMValue)
  - Expression Detection ロジック追加(wrapper AST 判定)
  - 動作例を完全に更新(全機能が動作済み)

- docs/development/current/main/phases/phase-288/README.md
  - Phase 288.1 完了セクション追加
  - 変更ファイル一覧(8ファイル, +592行)記録
  - 確認コマンド 4種 記録(変数永続化/式表示/_変数/リセット)
  - 回帰テスト結果記録(154/154 PASS)

- docs/development/current/main/10-Now.md
  - "Phase 288.1 完了" に更新
  - 次の候補(REPL UX 改善 / JoinIR 設計作業)を追記

- CURRENT_TASK.md
  - 1段落サマリー更新(288.1 完了、次の方向性)

Phase 288.1 成果(SSOT記録):
 変数永続化(session → eval bridge)
 式自動表示(pure expression auto-output)
 _ 変数(last displayed value)
 Fail-Fast 未定義エラー + ヒント
 セッションリセット(.reset)
 154/154 smoke tests PASS(file mode 不変)

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-25 23:05:20 +09:00
edc7355937 refactor(joinir): unify boundary join_inputs SSOT (pattern4/6/7)
Apply Phase 256.8 SSOT fix to Pattern4/6/7:
- Use join_module.entry.params.clone() instead of hardcoded ValueIds
- Add fail-fast validation for params count mismatch
- Remove ValueId(0), ValueId(PARAM_MIN + k) patterns
- Clean up unused PARAM_MIN imports

This prevents entry_param_mismatch errors structurally and maintains
consistency with Pattern2/3.

Changes:
- pattern4_with_continue.rs: Lines 442-476 (SSOT extraction + validation)
- pattern6_scan_with_init.rs: Lines 447-471 (SSOT extraction + validation)
- pattern7_split_scan.rs: Lines 495-526 (SSOT extraction + validation)

All patterns now use the same SSOT principle:
1. Extract entry function (priority: join_module.entry → fallback "main")
2. Use params as SSOT: join_inputs = entry_func.params.clone()
3. Build host_inputs in expected order (pattern-specific)
4. Fail-fast validation: join_inputs.len() == host_inputs.len()

Verification:
- cargo build --release:  PASS (no PARAM_MIN warnings)
- Quick profile:  First FAIL still json_lint_vm (baseline maintained)
- Pattern6 smoke:  PASS (index_of test)
- Pattern7 smoke: Pre-existing phi pred mismatch (not introduced by SSOT)

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-20 20:05:11 +09:00
9336785680 feat(anf): Phase 146/147 - Loop/If Condition ANF with Compare support
## Phase 146 P0: ANF Routing SSOT Unified

**Goal**: Unify ANF routing in `lower_expr_with_scope()` L54-84, remove legacy lowering

**Changes**:
- expr_lowerer_box.rs: Added scope check (PureOnly → skip ANF, WithImpure → try ANF)
- post_if_post_k.rs: Removed legacy inline lowering (L271-285), added `lower_condition_legacy()` helper
- contract.rs: Already had `CondLoweringFailed` out-of-scope reason

**Test Results**:  Phase 146 P0 smoke (exit 7), 0 regressions

## Phase 146 P1: Compare Operator Support

**Goal**: Enable ANF for condition expressions with Compare operators

**Changes**:
- joinir_dev.rs: Added `anf_allow_pure_enabled()` (HAKO_ANF_ALLOW_PURE=1)
- expr_lowerer_box.rs: PureOnly scope ANF support (L56-66)
- execute_box.rs: Compare operator support (+122 lines)
  - `execute_compare_hoist()`, `execute_compare_recursive()`, `ast_compare_to_joinir()`
  - Extended `normalize_and_lower()` for Compare

**Test Results**:  Phase 146 P1 smoke (exit 7 with flags), 0 regressions

## Phase 147 P0: Recursive Comparison ANF

**Goal**: Extend recursive ANF to Compare operators

**Changes**:
- contract.rs: Added `AnfParentKind::Compare` variant
- plan_box.rs: Compare case in BinaryOp routing (L68-79, L134-139)
  - Distinguishes Compare vs arithmetic BinaryOp

**Benefits**: Enables recursive ANF for comparisons
- `s.length() == 3` → `t = s.length(); if (t == 3)` 
- `s1.length() < s2.length()` → `t1 = s1.length(); t2 = s2.length(); if (t1 < t2)` 

## Implementation Summary

**Files Modified** (9 files, +253 lines, -25 lines = +228 net):
1. src/config/env/joinir_dev.rs (+28 lines)
2. src/mir/control_tree/normalized_shadow/anf/contract.rs (+2 lines)
3. src/mir/control_tree/normalized_shadow/anf/execute_box.rs (+122 lines)
4. src/mir/control_tree/normalized_shadow/anf/plan_box.rs (+18 lines)
5. src/mir/control_tree/normalized_shadow/common/expr_lowerer_box.rs (+18 lines, -0 lines)
6. src/mir/control_tree/normalized_shadow/post_if_post_k.rs (+44 lines, -25 lines)
7. CURRENT_TASK.md
8. docs/development/current/main/10-Now.md
9. docs/development/current/main/30-Backlog.md

**Files Created** (7 files):
- apps/tests/phase146_p0_if_cond_unified_min.hako
- apps/tests/phase146_p1_if_cond_intrinsic_min.hako
- tools/smokes/.../phase146_p0_if_cond_unified_vm.sh
- tools/smokes/.../phase146_p0_if_cond_unified_llvm_exe.sh
- tools/smokes/.../phase146_p1_if_cond_intrinsic_vm.sh
- tools/smokes/.../phase146_p1_if_cond_intrinsic_llvm_exe.sh
- docs/development/current/main/phases/phase-146/README.md

**Acceptance Criteria**:  All met
- cargo build --release: PASS (0 errors, 0 warnings)
- Phase 145 regressions: PASS (exit 12, 18, 5)
- Phase 146 P0: PASS (exit 7)
- Phase 146 P1: PASS (exit 7 with HAKO_ANF_ALLOW_PURE=1)

**Architecture**:
- SSOT: ANF routing only in `lower_expr_with_scope()` L54-84
- Box-First: Phase 145 `anf/` module extended
- Legacy removed: post_if_post_k.rs unified with SSOT

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-19 17:03:56 +09:00
6a3b6deb20 feat(anf): Phase 145 P0/P1/P2 - ANF (A-Normal Form) transformation
Implement ANF transformation for impure expressions to fix evaluation order:

Phase 145 P0 (Skeleton):
- Add anf/ module with contract/plan/execute 3-layer separation
- AnfDiagnosticTag, AnfOutOfScopeReason, AnfPlan enums
- Stub execute_box (always returns Ok(None))
- 11 unit tests pass

Phase 145 P1 (Minimal success):
- String.length() whitelist implementation
- BinaryOp + MethodCall pattern: x + s.length() → t = s.length(); result = x + t
- Exit code 12 verification (VM + LLVM EXE)
- 17 unit tests pass

Phase 145 P2 (Generalization):
- Recursive ANF for compound expressions
- Left-to-right, depth-first evaluation order
- Patterns: x + s.length() + z, s1.length() + s2.length()
- ANF strict mode (HAKO_ANF_STRICT=1)
- Diagnostic tags (joinir/anf/*)
- 21 unit tests pass, 0 regression

Also includes Phase 143 P2 (else symmetry) completion.

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-19 16:19:49 +09:00
845ae70cb7 chore: Remove unused imports in normalized_shadow modules
Cleaned up unused imports after Phase 143 execution fix (5e662eaaf).

**Priority files (Phase 143)**:
- if_as_last_join_k.rs: removed ValueId, BTreeMap
- loop_true_break_once.rs: added #[cfg(test)] for test-only imports
- post_if_post_k.rs: removed ValueId, BTreeMap
- normalized_helpers.rs: added #[cfg(test)] for Span

**Additional cleanup**:
- contract_checks.rs: removed BasicBlockId
- joinir/mod.rs: removed Info struct re-exports (functions kept)
- patterns/mod.rs: removed Info struct re-exports (functions kept)
- ast_feature_extractor.rs: removed EscapeSkipPatternInfo
- plan_box.rs: added #[cfg(test)] for PlanKind

**Verification**:
- 0 unused import warnings (was 20+)
- All 69 normalized_shadow tests pass
- Clean build with --release

Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-19 08:36:45 +09:00
4082abb30c feat(normalization): Phase 142 P0 - Loop statement-level normalization
Phase 142-loopstmt P0: Statement-level normalization

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

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
2025-12-19 05:28:49 +09:00
f775c0fe01 docs: Phase 131 P2 DirectValue exit reconnection complete
Update completion status for Phase 131 P2:
- CURRENT_TASK.md: Note WSL restart resolved EXDEV issue
- 10-Now.md: Add Phase 131 P2 completion entry
- phase-131/README.md: Document P1.5 and P2 deliverables

Phase 131 Status: DONE 
- loop(true) break-once executes correctly
- Exit values propagate through DirectValue mode
- VM + LLVM EXE parity verified

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-18 17:48:18 +09:00
93e62b1433 docs(phase93): Phase 93 P0完了記録 & ドキュメント整理
## 追加
- docs/development/current/main/phases/phase-93/README.md
  - Phase 93 P0 (ConditionOnly Derived Slot) 完了記録
  - 実装内容・テスト結果の詳細

## 更新
- CURRENT_TASK.md: Phase 93 P0完了に伴う更新
- 10-Now.md: 現在の進捗状況更新
- 30-Backlog.md: Phase 92/93関連タスク整理
- phase-91/92関連ドキュメント: historical化・要約化

## 削減
- 735行削減(historical化により詳細をREADMEに集約)

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-16 23:30:39 +09:00
568619df89 feat(mir): Phase 92 P2-2 - Body-local variable support for ConditionalStep
Phase 92 P2-2完了:ConditionalStepのcondition(ch == '\\'など)でbody-local変数をサポート

## 主要変更

### 1. condition_lowerer.rs拡張
- `lower_condition_to_joinir`に`body_local_env`パラメータ追加
- 変数解決優先度:ConditionEnv → LoopBodyLocalEnv
- すべての再帰ヘルパー(comparison, logical_and, logical_or, not, value_expression)対応

### 2. conditional_step_emitter.rs修正
- `emit_conditional_step_update`に`body_local_env`パラメータ追加
- condition loweringにbody-local環境を渡す

### 3. loop_with_break_minimal.rs修正
- break condition loweringをbody-local init の**後**に移動(line 411)
- header_break_lowering::lower_break_conditionにbody_local_env渡す
- emit_conditional_step_updateにbody_local_env渡す(line 620)

### 4. header_break_lowering.rs修正
- `lower_break_condition`に`body_local_env`パラメータ追加
- scope_managerにbody-local環境を渡す

### 5. 全呼び出し箇所修正
- expr_lowerer.rs (2箇所)
- method_call_lowerer.rs (2箇所)
- loop_with_if_phi_if_sum.rs (3箇所)
- loop_with_continue_minimal.rs (1箇所)
- carrier_update_emitter.rs (1箇所・legacy)

## アーキテクチャ改善

### Break Condition Lowering順序修正
旧: Header → **Break Cond** → Body-local Init
新: Header → **Body-local Init** → Break Cond

理由:break conditionが`ch == '\"'`のようにbody-local変数を参照する場合、body-local initが先に必要

### 変数解決優先度(Phase 92 P2-2)
1. ConditionEnv(ループパラメータ、captured変数)
2. LoopBodyLocalEnv(body-local変数like `ch`)

## テスト

### ビルド
 cargo build --release成功(30 warnings、0 errors)

### E2E
⚠️ body-local promotion問題でブロック(Phase 92範囲外)
- Pattern2はbody-local変数をcarrier promotionする必要あり
- 既存パターン(A-3 Trim, A-4 DigitPos)に`ch = get_char(i)`が該当しない
- **Phase 92 P2-2目標(condition loweringでbody-local変数サポート)は達成**

## 次タスク(Phase 92 P3以降)
- body-local variable promotion拡張(Pattern2で`ch`のような変数を扱う)
- P5b E2Eテスト完全動作確認

## Phase 92 P2-2完了
 Body-local変数のcondition lowering対応完了
 ConditionalStepでbody-local変数参照可能
 Break condition lowering順序修正
2025-12-16 17:08:15 +09:00
42f8f4d088 docs: Update CURRENT_TASK.md - Phase 91 planning complete
## P1 Status Update
- Phase 91 Step 1:  COMPLETE
  - Loop inventory: 34 loops across 6 files
  - Pattern P5b design: Complete
  - Test fixture: Created
  - Capability taxonomy: Extended

- Phase 91 Step 2: NEXT (Implementation)
  - detect_escape_pattern() in Canonicalizer
  - Unit tests + parity verification
  - No lowering in Step 2 (deferred to Step 3)

## Key Findings
- Current JoinIR readiness: 47% (16/30 loops)
- Pattern P5b unlocks: All escape sequence parsers
- Effort estimate: 2-3 days for recognition phase

## Next Priority
Phase 91 Step 2: Implement Canonicalizer recognition for P5b pattern

🤖 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
2025-12-16 14:22:55 +09:00
521e58d061 docs: mark loop canonicalizer Phase 137-141 complete 2025-12-16 07:54:33 +09:00
5edd81e373 refactor(mir): Phase 138-P1-A - loop_canonicalizer を4モジュール分割
## 概要
- 931行の mod.rs を 4モジュール + 161行 mod.rs に分割
- 全14テスト PASS、退行なし

## モジュール構成
- skeleton_types.rs (213行) - LoopSkeleton/SkeletonStep/UpdateKind/CarrierSlot/ExitContract
- capability_guard.rs (104行) - RoutingDecision/capability_tags
- pattern_recognizer.rs (249行) - try_extract_skip_whitespace_pattern
- canonicalizer.rs (414行) - canonicalize_loop_expr + 統合テスト
- mod.rs (161行) - 型定義と re-export

## ファイルサイズ達成
- 最大ファイル: canonicalizer.rs 414行(目標250行を一部超過するが許容範囲)
- mod.rs: 931行 → 161行 (83%削減)
- 合計: 1141行(元の931行 + tests分離で増加)

## テスト結果
- 14 tests passed
- loop_canonicalizer::* 全テスト green
2025-12-16 06:41:46 +09:00
a0009d474d feat(mir): Loop Canonicalizer Phase 3 - skip_whitespace pattern recognition
## Summary
skip_whitespace パターンを Skeleton→Decision で認識可能に。
dev-only 観測で chosen=Pattern3IfPhi / missing_caps=[] を固定。

## Changes
- src/mir/loop_canonicalizer/mod.rs:
  - try_extract_skip_whitespace_pattern() 追加
    - loop(cond) { ... if check { p = p + 1 } else { break } } パターン認識
    - carrier name, delta, body statements を抽出
  - canonicalize_loop_expr() 拡張(skip_whitespace 対応)
    - Pattern3IfPhi 成功時は RoutingDecision::success 返却
    - Skeleton に HeaderCond, Body, Update ステップ追加
    - CarrierSlot に Counter role 設定
    - ExitContract に has_break=true 設定
  - Phase 3 unit tests 追加
    - test_skip_whitespace_pattern_recognition: 基本パターン
    - test_skip_whitespace_with_body_statements: body 付きパターン
    - test_skip_whitespace_fails_without_else: else なし失敗
    - test_skip_whitespace_fails_with_wrong_delta: 減算パターン失敗
  - Phase 2 obsolete tests 削除
- src/mir/builder/control_flow/joinir/routing.rs:
  - Debug 出力拡張(chosen pattern 表示)

## Tests
- cargo test --release --lib loop_canonicalizer::tests: PASS(11 tests)
- cargo test --release --lib: PASS(1044 tests, 退行なし)
- HAKO_JOINIR_DEBUG=1 test_pattern3_skip_whitespace.hako:
  - chosen=Pattern3IfPhi 
  - missing_caps=[] 

## Validation
-  dev-only 観測(HAKO_JOINIR_DEBUG=1)のときだけログ出力
-  フラグ OFF 時は完全不変
-  skip_whitespace パターンで SUCCESS 固定
-  unit tests で全パターン固定

Phase 137-3 complete
2025-12-16 05:38:18 +09:00
ec1ff5b766 docs: update Phase 137-1 completion status and Phase 2 roadmap
## Summary
Loop Canonicalizer Phase 1(型定義)の完了と、Phase 2(dev-only 観測)への
ロードマップを記録。

## Changes

### loop-canonicalizer.md
- Status: Design (P0) → Phase 1 done(型定義まで)
- 実装の入口を明記(src/mir/loop_canonicalizer/mod.rs)
- NYASH_LOOP_ROUTING_TRACE → joinir_dev_enabled() に変更
  (新 ENV を増やさず、既存のスイッチに寄せる)

### CURRENT_TASK.md
- Phase 137-1 完了を追記
- P0 を「設計」→「Phase 2(dev-only 観測)」へ更新

### phases/README.md
- Phase 137 を追加

### phases/phase-137/README.md (NEW)
- Phase 1 完了サマリ
- Phase 2 予定を最小で記録

### joinir-design-map.md / 01-JoinIR-Selfhost-INDEX.md
- loop-canonicalizer への導線を追加

## Next Steps

Phase 2: dev-only 観測の導入
- canonicalize_loop_expr() の薄い入口を追加
- joinir_dev_enabled() 配下でログ出し
- 既定挙動は完全不変

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-16 05:10:29 +09:00
84ded8ca6c docs: add loop canonicalizer SSOT + MirBuilder entrypoint 2025-12-16 04:37:47 +09:00
455d511187 feat(mir): Phase 136 P0 - ValueId allocator SSOT 徹底(関数内経路から value_gen.next() 掃討)
## Summary
Eliminates remaining `value_gen.next()` calls from function-context code paths,
unifying all ValueId allocation through `MirBuilder::next_value_id()` SSOT allocator.

## Changes

### 1. Fixed `new_typed_value()` (src/mir/builder.rs:1068)
**Before**: `let id = self.value_gen.next();` (bypasses function context)
**After**: `let id = self.next_value_id();` (respects function context)

This is a public API used in function context, so must use SSOT allocator
to avoid collisions with reserved PHI dsts and function params.

### 2. Fixed test code (src/mir/builder.rs)
**test_shadowing_binding_restore** (lines 1161, 1171):
- Simulates function scope with `push_lexical_scope()`
- Changed to `builder.next_value_id()` for function scope simulation

**test_valueid_binding_parallel_allocation** (lines 1196-1216):
- Tests ValueId/BindingId independence
- Changed to `builder.next_value_id()` with note that Module context fallback preserves test intent

### 3. Verified Module context fallbacks (OK, no change needed)
These already check `current_function.is_some()` and use `value_gen.next()` only as Module context fallback:
- `src/mir/builder/utils.rs:43` - next_value_id() SSOT implementation
- `src/mir/builder/utils.rs:436` - pin_to_slot()
- `src/mir/builder/utils.rs:467` - materialize_local()
- `src/mir/utils/phi_helpers.rs:69` - insert_phi_unified()

## Verification
```bash
rg -n "value_gen\.next\(" src/mir --type rust | grep -v "Module context" | grep -v "//"
# Result: Only comments/docs remain
```

## Acceptance
 cargo test --release --lib - 997 passed
 phase135_trim_mir_verify.sh - PASS
 phase132_exit_phi_parity.sh - 3/3 PASS
 All function-context `value_gen.next()` eliminated

## Effect
- **Collision prevention**: No more ValueId collisions between function-level allocations and reserved PHI dsts
- **SSOT compliance**: All ValueId allocation flows through single allocator
- **Contract enforcement**: Phase 135 P1 contract_checks will catch violations immediately

🤖 Generated with Claude Code

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-15 19:37:30 +09:00
d82c332a40 feat(joinir): Phase 135 P0 - ConditionLoweringBox allocator SSOT (ValueId collision fix)
## Summary
Root cause: ConditionLoweringBox was bypassing ConditionContext.alloc_value (SSOT allocator),
causing ValueId collisions between JoinIR condition params and lowered instructions.

## Changes
1. **ConditionLoweringBox (expr_lowerer.rs)**: Must use ConditionContext.alloc_value
   - Pass &mut ConditionContext to lower_condition (SSOT allocator)
   - Eliminates internal counter usage

2. **Allocator unification (condition_lowerer.rs, method_call_lowerer.rs)**:
   - Accept &mut dyn FnMut() -> ValueId as allocator parameter
   - Ensures all lowering paths use same SSOT allocator

3. **Boundary Copy deduplication (joinir_inline_boundary_injector.rs)**:
   - Deduplicate condition_bindings by dst
   - Fail-Fast if different sources target same dst (MIR SSA violation)

4. **Trim pattern fixes (trim_loop_lowering.rs, trim_pattern_validator.rs, stmts.rs)**:
   - Use builder.next_value_id() instead of value_gen.next() in function context
   - Ensures function-level ValueId allocation respects reserved PHI dsts

## Acceptance
 ./target/release/hakorune --verify apps/tests/phase133_json_skip_whitespace_min.hako
 Smoke: phase135_trim_mir_verify.sh - MIR SSA validation PASS
 Regression: phase132_exit_phi_parity.sh - 3/3 PASS
 Regression: phase133_json_skip_whitespace_llvm_exe.sh - compile-only PASS

🤖 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
2025-12-15 18:49:08 +09:00
c8970bd4d4 docs: Phase 132-P3 verifier note 2025-12-15 06:08:19 +09:00
771bf6b0d1 feat(joinir): Phase 132-P3 - Exit PHI collision early detection
Added verify_exit_phi_no_collision() to contract_checks.rs for early detection
of ValueId collisions between exit PHIs and other instructions.

Problem detected:
- If exit_phi_builder uses builder.value_gen.next() (module-level) instead of
  func.next_value_id() (function-level), ValueIds can collide:

  Example:
  - bb0: %1 = const 0   (counter init)
  - bb3: %1 = phi ...   (exit PHI - collision!)

Previous behavior:
- Error only detected at LLVM backend runtime
- Cryptic error: "Cannot overwrite PHI dst=1"

New behavior:
- Panic at Rust compile time (debug build)
- Clear error message with fix suggestion:
  "Exit PHI dst %1 collides with instruction in block 0
   Fix: Use func.next_value_id() in exit_phi_builder.rs"

Benefits:
- 🔥 Fail-Fast: Catch errors during Rust compilation, not LLVM execution
- 📋 Clear messages: Exact collision point + fix suggestion
- 🧪 Testable: verify_exit_phi_no_collision() can be unit tested

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-15 06:00:48 +09:00
447d4ea246 feat(llvm): Phase 132 - Pattern 1 exit value parity fix + Box-First refactoring
## Phase 132: Exit PHI Value Parity Fix

### Problem
Pattern 1 (Simple While) returned 0 instead of final loop variable value (3)
- VM: RC: 3  (correct)
- LLVM: Result: 0  (wrong)

### Root Cause (Two Layers)
1. **JoinIR/Boundary**: Missing exit_bindings → ExitLineReconnector not firing
2. **LLVM Python**: block_end_values snapshot dropping PHI values

### Fix
**JoinIR** (simple_while_minimal.rs):
- Jump(k_exit, [i_param]) passes exit value

**Boundary** (pattern1_minimal.rs):
- Added LoopExitBinding with carrier_name="i", role=LoopState
- Enables ExitLineReconnector to update variable_map

**LLVM** (block_lower.py):
- Use predeclared_ret_phis for reliable PHI filtering
- Protect builder.vmap PHIs from overwrites (SSOT principle)

### Result
-  VM: RC: 3
-  LLVM: Result: 3
-  VM/LLVM parity achieved

## Phase 132-Post: Box-First Refactoring

### Rust Side
**JoinModule::require_function()** (mod.rs):
- Encapsulate function search logic
- 10 lines → 1 line (90% reduction)
- Reusable for Pattern 2-5

### Python Side
**PhiManager Box** (phi_manager.py - new):
- Centralized PHI lifecycle management
- 47 lines → 8 lines (83% reduction)
- SSOT: builder.vmap owns PHIs
- Fail-Fast: No silent overwrites

**Integration**:
- LLVMBuilder: Added phi_manager
- block_lower.py: Delegated to PhiManager
- tagging.py: Register PHIs with manager

### Documentation
**New Files**:
- docs/development/architecture/exit-phi-design.md
- docs/development/current/main/investigations/phase132-llvm-exit-phi-wrong-result.md
- docs/development/current/main/phases/phase-132/

**Updated**:
- docs/development/current/main/10-Now.md
- docs/development/current/main/phase131-3-llvm-lowering-inventory.md

### Design Principles
- Box-First: Logic encapsulated in classes/methods
- SSOT: Single Source of Truth (builder.vmap for PHIs)
- Fail-Fast: Early explicit failures, no fallbacks
- Separation of Concerns: 3-layer architecture

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-15 03:17:31 +09:00
18bf35e6d4 fix(llvm): Phase 131-10 - Smart console.log routing(Segfault修正)
## 問題
- Integer値をi8*ポインタに変換 → Segfault(Exit 139)

## 解決策
- String literal → nyash.console.log(i8*)
- Integer/Handle → nyash.console.log_handle(i64)

## 結果
- Case B (loop_min_while): LLVM outputs `0,1,2` 
- VM/LLVM完全パリティ達成

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-14 09:19:00 +09:00
5709026812 docs(phase131): Phase 131-3 完了 - LLVM lowering 棚卸し(3ケース)
Phase 131-3 完了: MIR→LLVM lowering 棚卸し

テスト結果マトリックス:
- Case A (phase87_llvm_exe_min.hako):  PASS (baseline)
- Case B (loop_min_while.hako):  TAG-EMIT (PHI after terminator)
- Case B2 (print(42) simple):  PASS (BoxCall works)
- Case C (llvm_stage3_loop_only.hako):  TAG-EMIT (JoinIR pattern gap)

Critical Bugs:
1. Bug #1: PHI After Terminator (Case B)
   - 原因: function_lower.py が terminator を PHI より先に emit
   - 修正: 4-pass block emission (2-3h)

2. Bug #2: JoinIR Pattern Gap (Case C)
   - 原因: loop(true) { break } パターンが JoinIR 未対応
   - 修正: Pattern 5 設計・実装 (3-4h)

Next Actions:
- P1 (推奨): PHI ordering 修正 → 80% のループを有効化
- P2: JoinIR Pattern 5 → infinite loop 対応

ドキュメント:
- phase131-3-llvm-lowering-inventory.md: 詳細棚卸し結果
- phase87-selfhost-llvm-exe-line.md: LLVM IR parsing error 追記
- CURRENT_TASK.md: phase131-3 参照追加

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 05:55:21 +09:00
e912ef9134 docs(phase131): SSOT整備 - VM入口/env var/次の入口を1箇所に固定
Phase 131 SSOT 整備完了:

1) VM Box 解決の入口SSOT(phase131-2-box-resolution-map.md)
   - 📍 入口3行を先頭に追記(handle_new_box/UnifiedRegistry/BoxFactoryRegistry)
   - NYASH_VM_FAST を bench/profile-only 特例と明記

2) 環境変数のSSOTを1箇所へ(environment-variables.md)
   - ## LLVM Build Pipeline セクション新設(14変数)
   - phase87 の Environment Variables セクションを参照リンクに置き換え
   - 重複表を削除、SSOTドキュメントへの導線確立

3) 次の開発の入口を明確化(CURRENT_TASK.md)
   - P1(Loop Canonicalizer): 🔶 設計待ち(外部検討中)
   - P2(JoinIR / Selfhost depth-2):  実装可能
   - 次に触るSSOTを1行で指定:
     - Loop系: joinir-architecture-overview.md
     - VM Box系: phase131-2-box-resolution-map.md

Impact:
- 迷子防止: 各領域のSSOTが1行で分かる
- 重複削減: env var 表を environment-variables.md に集約
- 状態明確化: P1/P2 の「設計待ち/実装可能」が一目瞭然

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 05:27:39 +09:00
05c5021147 docs(phase131): LLVM SSOT強化 + ConsoleBox調査完了 + Phase86-90要約
Phase 131-1 完了: LLVM exe line SSOT 強化
- phase87-selfhost-llvm-exe-line.md に 4セクション追加(+300行)
  - 環境変数リファレンス(14変数)
  - 成功/失敗基準(exit code 0/1/2/3)
  - コンパイラモード説明(harness vs crate)
  - デバッグセクション拡張
- "1コマンドで再現" 可能な状態を確立

Phase 131-2 完了: ConsoleBox 問題調査
- VM の 3重登録経路を特定(BoxFactoryRegistry/UnifiedRegistry/Builtin)
- LLVM backend は Phase 133 で解決済み
- 3つのドキュメント作成:
  - phase131-2-consolebox-investigation.md(詳細調査)
  - phase131-2-summary.md(エグゼクティブサマリ)
  - phase131-2-box-resolution-map.md(Box 解決マップ)

Phase 86-90 完了: Loop frontends 要約
- phase86-90-loop-frontends-summary.md 追加
- Pattern4/ContinueReturn/ParseStringComposite の経緯を1枚に集約
- INDEX から導線追加

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 05:24:31 +09:00
5b5f3b55d4 docs: Phase 88-89 完了 + Phase 90 P0 追加
## CURRENT_TASK.md
- Phase 88-89 完了を反映
- Phase 90 P0 追加(_parse_string 合成 fixture)
- P1/P2 候補を Phase 91/92 に繰り上げ
- 最終更新: 2025-12-14

## 10-Now.md
- Phase 88 到達点を追記(continue + 可変ステップ)
- Phase 89 P0/P1 到達点を追記(ContinueReturn detector + lowering)
- 未検証の断定は書かない(dev-only fixture で固定、と明記)

Impact:
- 正確な進捗記録
- Phase 90 の明確な方針(fixture → detector → lowering)

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 02:31:39 +09:00
4e3fc4ad49 feat(joinir): Phase 89 P0 - Continue + Early Return pattern detector
## Pattern4 Detector 締め
- is_pattern4_continue_minimal() を厳しく
- Select 必須 + conditional Jump exactly 1
- loop 内 return を P4 と誤認しない

## 新パターン箱
- LoopPattern::ContinueReturn enum 追加
- has_return_in_loop_body() helper 追加
- Fail-Fast: UnimplementedPattern error

## Normalized-dev 統合
- NormalizedDevShape::PatternContinueReturnMinimal
- detector: Select + conditional Jumps >= 2
- canonical には入れない(dev-only)

## Documentation
- 10-Now.md, CURRENT_TASK.md 更新

Impact:
- 987 lib tests PASS
- 60 normalized_dev tests PASS
- Pattern4 誤爆防止

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 00:59:58 +09:00
8a8c90fc74 docs(joinir): Phase 86 SSOT 追記完了 + Phase 88 準備
Phase 86 SSOT 追記(P0 完了):
- 10-Now.md: carrier_init_builder / error_tags 確立を追記
- CURRENT_TASK.md: P0→P1 繰り上げ、Done 節追加

Phase 88 準備:
- nyash_kernel/lib.rs: AOT 実行器で Ring0Context 初期化
- nyash.toml: プラグインパス正規化(plugins/...)
- auto_detect.conf: integration タイムアウト 120秒に延長

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 00:05:16 +09:00
cd12a2569e feat(llvm): Phase 87 - LLVM exe line SSOT + integration smoke
Established single pipeline for .hako → executable generation.

SSOT: tools/build_llvm.sh
- Standard procedure: build_llvm.sh input.hako -o output_exe
- Prerequisites: llvm-config-18, llvmlite, LLVM features enabled
- Pipeline: .hako → MIR → LLVM IR → .o → executable

New files:
- docs: phase87-selfhost-llvm-exe-line.md (SSOT procedure doc)
  * Full troubleshooting guide (llvm-config, llvmlite, linking)
  * Advanced usage (custom output, debugging, performance notes)
  * Anti-patterns documentation (no duplication/fragmentation)
- fixture: apps/tests/phase87_llvm_exe_min.hako (5 lines, return 42)
- smoke: integration/apps/phase87_llvm_exe_min.sh
  * Exit code verification (stdout-independent testing)
  * SKIP if LLVM unavailable (graceful degradation)
  * Integration profile only (not in quick)
- index: 10-Now.md, 01-JoinIR-Selfhost-INDEX.md (Phase 87 entries)
- task: CURRENT_TASK.md (Phase 74-87 status update)

Integration smoke: demonstrates full pipeline
- Build: .hako → .o → exe (successful )
- Runtime: Ring0 initialization issue (known limitation, out of scope)
Quick profile: unchanged (integration only, per policy)
Policy: No duplicate build paths, SSOT maintained

987/987 tests PASS (production stable)
2025-12-13 22:51:13 +09:00
f858a5fbd0 docs: archive CURRENT_TASK and add lightweight view 2025-12-13 22:21:04 +09:00
8b48bec962 feat(joinir): Phase 78 - BindingId infrastructure for promoted carriers (dev-only)
Phase 78 adds infrastructure to assign BindingIds to synthetic promoted
carriers (e.g., is_digit_pos, is_ch_match), enabling type-safe promoted
variable lookup without string-based naming conventions.

Key Changes:
1. CarrierVar.binding_id field (dev-only):
   - Added Option<BindingId> to track BindingId for each carrier
   - Updated all constructors and struct instantiations

2. CarrierBindingAssigner Box (new file, 273 lines):
   - Allocates BindingIds for promoted carriers via builder.allocate_binding_id()
   - Records original → promoted mapping in promoted_bindings
   - Sets binding_id field on promoted CarrierVar
   - Includes 3 comprehensive unit tests

3. ConditionEnv.register_carrier_binding() (new method):
   - Registers carrier BindingId → ValueId mappings
   - Enables type-safe lookup via binding_id_map

4. Logging cleanup:
   - Gated 6 eprintln! statements with NYASH_JOINIR_DEBUG
   - Unified logging tags to [binding_pilot/*]

Design Decisions:
- Promoters create CarrierInfo, lowering code assigns BindingIds
- CarrierBindingAssigner called from Pattern2/4 lowering (has builder access)
- Clear documentation prevents misuse (promoters lack builder access)

Files modified (18):
- carrier_info.rs: binding_id field added to CarrierVar
- carrier_binding_assigner.rs: New Box for BindingId allocation
- condition_env.rs: register_carrier_binding() method
- mod.rs: Module exports
- pattern2_with_break.rs, pattern4_with_continue.rs: Updated for binding_id
- loop_body_*_promoter.rs: Logging cleanup + binding_id in structs
- phase78-bindingid-promoted-carriers.md: Architecture documentation

Tests: 970/970 PASS (zero regressions)
Status: Infrastructure complete, integration deferred to Phase 79

Next Phase: Wire CarrierBindingAssigner in Pattern2/4 lowering + E2E tests

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

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-13 16:20:33 +09:00
7ef9079aae docs(current): mark Phase 77 done, queue Phase 78 2025-12-13 06:14:00 +09:00
65ecd75529 docs(task): Phase 74-77 status update + Phase 77 implementation readiness 2025-12-13 05:36:39 +09:00
ea7fb607c7 docs(roadmap): Update CURRENT_TASK.md with Phase 70-B/C/72/73
Add entries for completed phases:
- Phase 70-B: Multihop passthrough support (c2df1cac)
- Phase 70-C: Merge relay detection (24cc948f)
- Phase 72: PHI reserved observation (253eb59b)
- Phase 73: ScopeManager BindingId design + PoC (851bf4f8)

Add next roadmap:
- Phase 74-77: BindingId Migration (8-12 hours, 4 phases)
- All phases with links to SSOT docs
- Estimated effort per phase

🤖 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
2025-12-13 03:48:25 +09:00
db9c9055fa obs(joinir): Phase 72 - PHI Reserved Region Observation Complete
## Summary
Observed PHI dst ValueId distribution to determine if verifier can enforce
reserved region (0-99). **Conclusion: Verifier strengthening NOT recommended.**

## Key Finding
PHI dst allocation does NOT architecturally respect reserved region:
- PHI dst comes from `builder.next_value_id()` (host MirBuilder)
- Reserved region (0-99) is a JoinValueSpace contract for JoinIR lowering
- These are separate allocation pools with no enforcement mechanism
- Current stability is accidental (ValueId allocation ordering)

## Evidence
Manual verification (`apps/tests/loop_min_while.hako`):
- PHI dst = %3 (ValueId(3))  in reserved region
- Works because PHI allocated early in function (0-20 typical)
- JoinValueSpace allocates high (100+, 1000+)
- Accidental separation, not enforced

## Implementation
Added observation infrastructure (debug-only):
- `src/mir/join_ir/verify_phi_reserved.rs` (266 lines)
  - PHI dst observer with distribution analyzer
  - Region classifier (Reserved/Param/Local)
  - Human-readable report generator
- Instrumentation at PHI allocation points:
  - loop_header_phi_builder.rs:94, 151
  - Debug-only `observe_phi_dst()` calls

## Test Results
- Unit tests:  4/4 PASS (verify_phi_reserved module)
- Lib tests:  950/950 PASS, 56 ignored
- Normalized tests:  54/54 PASS
- Manual verification:  PHI dst in 0-99 observed

## Decision: Document, Don't Enforce
**Rationale**:
1. No architectural mechanism to enforce PHI dst ∈ [0, 99]
2. Adding verifier creates false assumptions about allocation order
3. Current system stable through separate pools (950/950 tests)
4. Future architectural fix possible (Phase 73+) but not urgent

## Documentation
- PHASE_72_SUMMARY.md: Executive summary and implementation record
- phase72-phi-reserved-observation.md: Detailed findings and analysis
- CURRENT_TASK.md: Phase 72 completion entry

## Next Steps
- Phase 73: Update architecture overview with Phase 72 findings
- Optional: Explicit PHI reserved pool (future enhancement)

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-13 03:23:02 +09:00
7b56a7c01d docs(joinir): Phase 70-A - Relay Runtime Guard (dev-only)
Phase 66で analysis/plan層はmultihop受理可能になったが、runtime lowering
(実行導線)はまだ未対応。このPhaseでは「未対応は同じタグで必ず落ちる」を
docs + テストで固定する。

Key changes:
- phase70-relay-runtime-guard.md: Runtime guard設計doc新規追加
  - 現状(plan OK / runtime NG)の明確化
  - Fail-Fastタグ [ownership/relay:runtime_unsupported] の標準化
  - Phase 70-B以降の解除条件

- pattern3_with_if_phi.rs: エラーメッセージのタグ統一
  - [ownership/relay:runtime_unsupported] 形式に変更
  - var/owner_scope/relay_path の診断情報追加

- normalized_joinir_min.rs: 固定テスト追加
  - test_phase70a_multihop_relay_runtime_unsupported_tag
  - Plan層のmultihop受理確認 + runtime拒否の文書化

Tests: normalized_dev 50/50 PASS (+1), lib 950/950 PASS

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

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-13 02:22:29 +09:00