Files
hakorune/src/mir/join_ir/normalized.rs
nyash-codex 6bcc70e07e refactor(joinir): Phase 89 リファクタリング - 5) fixture名SSOT化
変更内容:
- 新規ファイル: src/mir/join_ir/normalized/dev_fixtures.rs (SSOT)
  - NormalizedDevFixture enum で fixture 名・パス・ルーティング統一管理
  - ALL_DEV_FIXTURES 配列で一覧化
  - fixture_content() / load_and_lower() ヘルパー実装
- FunctionRoute を route.rs に分離
  - ast_lowerer/route.rs 新規作成
  - resolve_function_route() を route.rs に移動
  - dev fixtures を SSOT から自動登録
- fixtures.rs を簡潔化
  - 4つの builder 関数を SSOT 呼び出しに変更
  - 散在していた include_str! パスを削除

メリット:
- typo・不一致によるルーティングミスを防止
- 新しい fixture 追加時は1箇所のみ変更
- 責務の明確化(route.rs / dev_fixtures.rs)

テスト結果:
- lib tests: 993 passed (回帰なし)
- normalized_dev tests: 61 passed / 1 failed (ベースライン維持)

Generated with Claude Code

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2025-12-14 03:07:53 +09:00

1277 lines
46 KiB
Rust
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

//! Minimal Normalized JoinIR model (Phase 26-H.B).
//!
//! テスト専用の極小サブセット。Pattern1 の while だけを Structured → Normalized に
//! 変換して遊ぶための足場だよ。本線の Structured→MIR 経路には影響しない。
use std::collections::{BTreeMap, HashSet};
use crate::mir::join_ir::{
BinOpKind, CompareOp, ConstValue, JoinContId, JoinFuncId, JoinFunction, JoinInst, JoinIrPhase,
JoinModule, MirLikeInst, UnaryOp,
};
use crate::mir::ValueId;
#[cfg(feature = "normalized_dev")]
use std::collections::HashMap;
#[cfg(feature = "normalized_dev")]
use std::panic::{catch_unwind, AssertUnwindSafe};
#[cfg(feature = "normalized_dev")]
pub mod fixtures;
#[cfg(feature = "normalized_dev")]
pub mod dev_env;
#[cfg(feature = "normalized_dev")]
pub mod dev_fixtures;
#[cfg(feature = "normalized_dev")]
pub mod shape_guard;
#[cfg(feature = "normalized_dev")]
use crate::mir::join_ir::normalized::shape_guard::NormalizedDevShape;
/// 環境レイアウト(最小)。
#[derive(Debug, Clone)]
pub struct EnvLayout {
pub id: u32,
pub fields: Vec<EnvField>,
}
#[derive(Debug, Clone)]
pub struct EnvField {
pub name: String,
pub ty: Option<crate::mir::MirType>,
pub value_id: Option<ValueId>,
}
/// 正規化済み関数 ID
#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
pub struct JpFuncId(pub u32);
/// 正規化済み関数Kont 兼用、is_kont で区別)。
#[derive(Debug, Clone)]
pub struct JpFunction {
pub id: JpFuncId,
pub name: String,
pub env_layout: Option<u32>,
pub body: Vec<JpInst>,
pub is_kont: bool,
}
/// 正規化済み命令(最小セット)。
#[derive(Debug, Clone)]
pub enum JpInst {
Let {
dst: ValueId,
op: JpOp,
args: Vec<ValueId>,
},
EnvLoad {
dst: ValueId,
env: ValueId,
field: usize,
},
EnvStore {
env: ValueId,
field: usize,
src: ValueId,
},
TailCallFn {
target: JpFuncId,
env: Vec<ValueId>,
},
TailCallKont {
target: JpFuncId,
env: Vec<ValueId>,
},
If {
cond: ValueId,
then_target: JpFuncId,
else_target: JpFuncId,
env: Vec<ValueId>,
},
}
/// 演算Let 用の最小サブセット)。
#[derive(Debug, Clone)]
pub enum JpOp {
Const(ConstValue),
BinOp(BinOpKind),
Unary(UnaryOp),
Compare(CompareOp),
BoxCall { box_name: String, method: String },
/// 三項演算子cond ? then : else
Select,
}
/// Normalized JoinIR モジュール(テスト専用)。
#[derive(Debug, Clone)]
pub struct NormalizedModule {
pub functions: BTreeMap<JpFuncId, JpFunction>,
pub entry: Option<JpFuncId>,
pub env_layouts: Vec<EnvLayout>,
pub phase: JoinIrPhase,
/// Structured に戻すためのスナップショット(テスト専用)。
pub structured_backup: Option<JoinModule>,
}
impl NormalizedModule {
pub fn to_structured(&self) -> Option<JoinModule> {
self.structured_backup.clone()
}
}
#[cfg(feature = "normalized_dev")]
fn verify_normalized_pattern1(module: &NormalizedModule) -> Result<(), String> {
if module.phase != JoinIrPhase::Normalized {
return Err("[joinir/normalized-dev] pattern1: phase must be Normalized".to_string());
}
// Env field bounds check
if let Some(env) = module.env_layouts.get(0) {
let field_count = env.fields.len();
for func in module.functions.values() {
for inst in &func.body {
match inst {
JpInst::EnvLoad { field, .. } | JpInst::EnvStore { field, .. } => {
if *field >= field_count {
return Err(format!(
"[joinir/normalized-dev] pattern1: env field out of range: {} (fields={})",
field, field_count
));
}
}
_ => {}
}
}
}
}
for func in module.functions.values() {
for inst in &func.body {
match inst {
JpInst::Let { .. }
| JpInst::EnvLoad { .. }
| JpInst::EnvStore { .. }
| JpInst::TailCallFn { .. }
| JpInst::TailCallKont { .. }
| JpInst::If { .. } => {}
}
}
// Tail: allow TailCall or If only
if let Some(last) = func.body.last() {
match last {
JpInst::TailCallFn { .. } | JpInst::TailCallKont { .. } | JpInst::If { .. } => {}
_ => {
return Err(format!(
"[joinir/normalized-dev] pattern1: function '{}' does not end with tail call/if",
func.name
))
}
}
}
}
Ok(())
}
/// Pattern1 専用: Normalized → Structured への簡易逆変換。
pub fn normalized_pattern1_to_structured(norm: &NormalizedModule) -> JoinModule {
assert_eq!(
norm.phase,
JoinIrPhase::Normalized,
"normalized_pattern1_to_structured expects Normalized phase"
);
let env_layout = norm
.env_layouts
.get(0)
.expect("normalized_pattern1_to_structured: missing env layout");
let mut module = JoinModule::new();
for (jp_id, jp_fn) in &norm.functions {
let params: Vec<ValueId> = jp_fn
.env_layout
.and_then(|id| norm.env_layouts.iter().find(|layout| layout.id == id))
.unwrap_or(env_layout)
.fields
.iter()
.enumerate()
.map(|(idx, f)| f.value_id.unwrap_or(ValueId(idx as u32)))
.collect();
let mut func = JoinFunction::new(JoinFuncId(jp_id.0), jp_fn.name.clone(), params);
for inst in &jp_fn.body {
match inst {
JpInst::Let { dst, op, args } => match op {
JpOp::Const(v) => func.body.push(JoinInst::Compute(MirLikeInst::Const {
dst: *dst,
value: v.clone(),
})),
JpOp::BoxCall { box_name, method } => {
func.body.push(JoinInst::Compute(MirLikeInst::BoxCall {
dst: Some(*dst),
box_name: box_name.clone(),
method: method.clone(),
args: args.clone(),
}))
}
JpOp::BinOp(op) => func.body.push(JoinInst::Compute(MirLikeInst::BinOp {
dst: *dst,
op: *op,
lhs: args.get(0).copied().unwrap_or(ValueId(0)),
rhs: args.get(1).copied().unwrap_or(ValueId(0)),
})),
JpOp::Unary(op) => func.body.push(JoinInst::Compute(MirLikeInst::UnaryOp {
dst: *dst,
op: *op,
operand: args.get(0).copied().unwrap_or(ValueId(0)),
})),
JpOp::Select => func.body.push(JoinInst::Compute(MirLikeInst::Select {
dst: *dst,
cond: args.get(0).copied().unwrap_or(ValueId(0)),
then_val: args.get(1).copied().unwrap_or(ValueId(0)),
else_val: args.get(2).copied().unwrap_or(ValueId(0)),
})),
JpOp::Compare(op) => func.body.push(JoinInst::Compute(MirLikeInst::Compare {
dst: *dst,
op: *op,
lhs: args.get(0).copied().unwrap_or(ValueId(0)),
rhs: args.get(1).copied().unwrap_or(ValueId(0)),
})),
},
JpInst::TailCallFn { target, env } => func.body.push(JoinInst::Call {
func: JoinFuncId(target.0),
args: env.clone(),
k_next: None,
dst: None,
}),
JpInst::TailCallKont { target, env } => func.body.push(JoinInst::Jump {
cont: JoinContId(target.0),
args: env.clone(),
cond: None,
}),
JpInst::If {
cond,
then_target,
else_target,
env,
} => {
// Jump to then_target on cond, else jump to else_target (Pattern1 minimal)
func.body.push(JoinInst::Jump {
cont: JoinContId(then_target.0),
args: env.clone(),
cond: Some(*cond),
});
func.body.push(JoinInst::Jump {
cont: JoinContId(else_target.0),
args: env.clone(),
cond: None,
});
}
JpInst::EnvLoad { .. } | JpInst::EnvStore { .. } => {
// Not used in Pattern1 minimal; ignore for now
}
}
}
module.add_function(func);
}
module.entry = norm.entry.map(|e| JoinFuncId(e.0));
module.phase = JoinIrPhase::Structured;
module
}
/// Pattern2 専用のミニ変換(最小サブセット: ループ変数1つ + break、acc などの LoopState を 1 個まで+ホスト 1 個まで)。
///
/// 制約:
/// - structured.phase は Structured であること
/// - main/loop_step/k_exit の 3 関数構成joinir_min_loop 相当)
pub fn normalize_pattern2_minimal(structured: &JoinModule) -> NormalizedModule {
assert!(
structured.is_structured(),
"normalize_pattern2_minimal: expected Structured JoinIR"
);
// Minimal guardrail: Pattern2 mini should have main/loop_step/k_exit only, with 1 loop param.
let func_count = structured.functions.len();
let loop_func = structured
.functions
.values()
.find(|f| f.name == "loop_step")
.or_else(|| structured.functions.get(&JoinFuncId::new(1)))
.expect("normalize_pattern2_minimal: loop_step not found");
assert!(
func_count == 3,
"normalize_pattern2_minimal: expected 3 functions (entry/loop_step/k_exit) but got {}",
func_count
);
let param_max = {
#[allow(unused_mut)]
let mut max = 3;
#[cfg(feature = "normalized_dev")]
{
let shapes = shape_guard::supported_shapes(structured);
if shapes
.iter()
.any(|s| matches!(s, NormalizedDevShape::JsonparserAtoiMini))
{
max = max.max(8);
}
if shapes
.iter()
.any(|s| matches!(s, NormalizedDevShape::JsonparserAtoiReal))
{
max = max.max(10);
}
if shapes
.iter()
.any(|s| matches!(s, NormalizedDevShape::JsonparserParseNumberReal))
{
max = max.max(12);
}
if shapes
.iter()
.any(|s| matches!(s, NormalizedDevShape::JsonparserSkipWsReal))
{
max = max.max(6);
}
if shapes.iter().any(|s| {
matches!(
s,
NormalizedDevShape::Pattern4ContinueMinimal
| NormalizedDevShape::JsonparserParseArrayContinueSkipWs
| NormalizedDevShape::JsonparserParseObjectContinueSkipWs
)
}) {
max = max.max(6);
}
if shapes.iter().any(|s| {
matches!(
s,
NormalizedDevShape::Pattern3IfSumMinimal
| NormalizedDevShape::Pattern3IfSumMulti
| NormalizedDevShape::Pattern3IfSumJson
| NormalizedDevShape::SelfhostIfSumP3
| NormalizedDevShape::SelfhostIfSumP3Ext
| NormalizedDevShape::SelfhostStmtCountP3
| NormalizedDevShape::SelfhostDetectFormatP3
)
}) {
max = max.max(6);
}
}
max
};
assert!(
(1..=param_max).contains(&loop_func.params.len()),
"normalize_pattern2_minimal: expected 1..={} params (loop var + carriers + optional host)",
param_max
);
let jump_conds = loop_func
.body
.iter()
.filter(|inst| matches!(inst, JoinInst::Jump { cond: Some(_), .. }))
.count();
let tail_calls = loop_func
.body
.iter()
.filter(|inst| matches!(inst, JoinInst::Call { k_next: None, .. }))
.count();
assert!(
jump_conds >= 1 && tail_calls >= 1,
"normalize_pattern2_minimal: expected at least one conditional jump and one tail call"
);
let mut functions = BTreeMap::new();
let mut env_layouts = Vec::new();
for (fid, func) in &structured.functions {
let env_layout_id = if func.params.is_empty() {
None
} else {
let id = env_layouts.len() as u32;
env_layouts.push(EnvLayout {
id,
fields: func
.params
.iter()
.enumerate()
.map(|(idx, vid)| EnvField {
name: format!("field{}", idx),
ty: None,
value_id: Some(*vid),
})
.collect(),
});
Some(id)
};
let mut body = Vec::new();
for inst in &func.body {
match inst {
JoinInst::Compute(MirLikeInst::Const { dst, value }) => body.push(JpInst::Let {
dst: *dst,
op: JpOp::Const(value.clone()),
args: vec![],
}),
JoinInst::Compute(MirLikeInst::BoxCall {
dst,
box_name,
method,
args,
}) => {
if let Some(dst) = dst {
body.push(JpInst::Let {
dst: *dst,
op: JpOp::BoxCall {
box_name: box_name.clone(),
method: method.clone(),
},
args: args.clone(),
})
}
}
JoinInst::Compute(MirLikeInst::BinOp { dst, op, lhs, rhs }) => {
body.push(JpInst::Let {
dst: *dst,
op: JpOp::BinOp(*op),
args: vec![*lhs, *rhs],
})
}
JoinInst::Compute(MirLikeInst::UnaryOp { dst, op, operand }) => {
body.push(JpInst::Let {
dst: *dst,
op: JpOp::Unary(*op),
args: vec![*operand],
})
}
JoinInst::Compute(MirLikeInst::Compare { dst, op, lhs, rhs }) => {
body.push(JpInst::Let {
dst: *dst,
op: JpOp::Compare(*op),
args: vec![*lhs, *rhs],
})
}
JoinInst::Compute(MirLikeInst::Select {
dst,
cond,
then_val,
else_val,
}) => {
body.push(JpInst::Let {
dst: *dst,
op: JpOp::Select,
args: vec![*cond, *then_val, *else_val],
})
}
JoinInst::Jump { cont, args, cond } => {
if let Some(cond_val) = cond {
body.push(JpInst::If {
cond: *cond_val,
then_target: JpFuncId(cont.0),
else_target: JpFuncId(loop_func.id.0),
env: args.clone(),
});
} else {
body.push(JpInst::TailCallKont {
target: JpFuncId(cont.0),
env: args.clone(),
});
}
}
JoinInst::Select {
dst,
cond,
then_val,
else_val,
..
} => {
body.push(JpInst::Let {
dst: *dst,
op: JpOp::Select,
args: vec![*cond, *then_val, *else_val],
});
}
JoinInst::Call { func, args, k_next, .. } => {
if k_next.is_none() {
body.push(JpInst::TailCallFn {
target: JpFuncId(func.0),
env: args.clone(),
});
}
}
JoinInst::MethodCall {
dst,
receiver,
method,
args,
..
} => {
let mut call_args = Vec::with_capacity(args.len() + 1);
call_args.push(*receiver);
call_args.extend(args.iter().copied());
body.push(JpInst::Let {
dst: *dst,
op: JpOp::BoxCall {
box_name: "unknown".to_string(),
method: method.clone(),
},
args: call_args,
});
}
_ => {
// Ret / other instructions are ignored in this minimal prototype
}
}
}
functions.insert(
JpFuncId(fid.0),
JpFunction {
id: JpFuncId(fid.0),
name: func.name.clone(),
env_layout: env_layout_id,
body,
is_kont: func.name.starts_with("k_"),
},
);
}
let norm = NormalizedModule {
functions,
entry: structured.entry.map(|e| JpFuncId(e.0)),
env_layouts,
phase: JoinIrPhase::Normalized,
structured_backup: Some(structured.clone()),
};
#[cfg(feature = "normalized_dev")]
{
verify_normalized_pattern2(&norm, param_max).expect("normalized Pattern2 verifier");
}
norm
}
#[cfg(feature = "normalized_dev")]
fn normalize_pattern2_shape(
structured: &JoinModule,
target_shape: NormalizedDevShape,
) -> Result<NormalizedModule, String> {
if !structured.is_structured() {
return Err("[normalize_p2] Not structured JoinIR".to_string());
}
let shapes = shape_guard::supported_shapes(structured);
if !shapes.contains(&target_shape) {
return Err(format!(
"[normalize_p2] shape mismatch: expected {:?}, got {:?}",
target_shape, shapes
));
}
Ok(normalize_pattern2_minimal(structured))
}
/// Phase 50: selfhost token-scan P2 を Normalized に載せるdev-only
#[cfg(feature = "normalized_dev")]
pub fn normalize_selfhost_token_scan_p2(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern2_shape(structured, NormalizedDevShape::SelfhostTokenScanP2)
}
/// Phase 51: selfhost token-scan P2accum 拡張)を Normalized に載せるdev-only
#[cfg(feature = "normalized_dev")]
pub fn normalize_selfhost_token_scan_p2_accum(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern2_shape(structured, NormalizedDevShape::SelfhostTokenScanP2Accum)
}
/// Phase 47-A/B: Normalize Pattern3 if-sum shapes to Normalized JoinIR
#[cfg(feature = "normalized_dev")]
pub fn normalize_pattern3_if_sum_minimal(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern3_if_sum_shape(structured, NormalizedDevShape::Pattern3IfSumMinimal)
}
/// Phase 47-B: Normalize Pattern3 if-sum multi-carrier (sum+count) shape.
#[cfg(feature = "normalized_dev")]
pub fn normalize_pattern3_if_sum_multi_minimal(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern3_if_sum_shape(structured, NormalizedDevShape::Pattern3IfSumMulti)
}
/// Phase 47-B: Normalize JsonParser if-sum (mini) shape.
#[cfg(feature = "normalized_dev")]
pub fn normalize_pattern3_if_sum_json_minimal(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern3_if_sum_shape(structured, NormalizedDevShape::Pattern3IfSumJson)
}
/// Phase 50: selfhost if-sum P3 を Normalized に載せるdev-only
#[cfg(feature = "normalized_dev")]
pub fn normalize_selfhost_if_sum_p3(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern3_if_sum_shape(structured, NormalizedDevShape::SelfhostIfSumP3)
}
/// Phase 51: selfhost if-sum P3ext 拡張)を Normalized に載せるdev-only
#[cfg(feature = "normalized_dev")]
pub fn normalize_selfhost_if_sum_p3_ext(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern3_if_sum_shape(structured, NormalizedDevShape::SelfhostIfSumP3Ext)
}
#[cfg(feature = "normalized_dev")]
fn normalize_pattern3_if_sum_shape(
structured: &JoinModule,
target_shape: NormalizedDevShape,
) -> Result<NormalizedModule, String> {
if !structured.is_structured() {
return Err("[normalize_p3] Not structured JoinIR".to_string());
}
let shapes = shape_guard::supported_shapes(structured);
if !shapes.contains(&target_shape) {
return Err(format!(
"[normalize_p3] shape mismatch: expected {:?}, got {:?}",
target_shape, shapes
));
}
// Phase 47-B: P3 if-sum は既存の P2 ミニ正規化器で十分に表現できる
// Select/If/Compare/BinOp をそのまま JpInst に写す)。
Ok(normalize_pattern2_minimal(structured))
}
/// Phase 48-A: Pattern4 (continue) minimal ループの正規化。
///
/// ガード:
/// - structured.phase は Structured であること
/// - 対象は Pattern4ContinueMinimal のシンプル continue パターン
///
/// 実装方針:
/// - Phase 48-A minimal: P2 正規化に委譲P4 は P2 の逆制御フローなので同じインフラ利用可)
/// - TODO (Phase 48-B): P4 固有の正規化実装
/// - EnvLayout for i, count carriers
/// - HeaderCond → ContinueCheck → Updates → Tail step sequence
/// - continue = early TailCallFn (skip Updates)
#[cfg(feature = "normalized_dev")]
pub fn normalize_pattern4_continue_minimal(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern4_continue_shape(structured, NormalizedDevShape::Pattern4ContinueMinimal)
}
/// Phase 48-B: JsonParser _parse_array continue skip_ws を Normalized に載せるdev-only
#[cfg(feature = "normalized_dev")]
pub fn normalize_jsonparser_parse_array_continue_skip_ws(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern4_continue_shape(
structured,
NormalizedDevShape::JsonparserParseArrayContinueSkipWs,
)
}
/// Phase 48-B: JsonParser _parse_object continue skip_ws を Normalized に載せるdev-only
#[cfg(feature = "normalized_dev")]
pub fn normalize_jsonparser_parse_object_continue_skip_ws(
structured: &JoinModule,
) -> Result<NormalizedModule, String> {
normalize_pattern4_continue_shape(
structured,
NormalizedDevShape::JsonparserParseObjectContinueSkipWs,
)
}
#[cfg(feature = "normalized_dev")]
fn normalize_pattern4_continue_shape(
structured: &JoinModule,
target_shape: NormalizedDevShape,
) -> Result<NormalizedModule, String> {
if !structured.is_structured() {
return Err("[normalize_p4] Not structured JoinIR".to_string());
}
// Use shape detection to verify P4 shape
let shapes = shape_guard::supported_shapes(structured);
if !shapes.contains(&target_shape) {
return Err(format!(
"[normalize_p4] shape mismatch: expected {:?}, got {:?}",
target_shape, shapes
));
}
// Phase 48-B: reuse Pattern2 minimal normalizer (continue is early tail-call).
Ok(normalize_pattern2_minimal(structured))
}
/// Pattern2 専用: Normalized → Structured への簡易逆変換。
pub fn normalized_pattern2_to_structured(norm: &NormalizedModule) -> JoinModule {
if let Some(backup) = norm.to_structured() {
return backup;
}
let mut module = JoinModule::new();
for (jp_id, jp_fn) in &norm.functions {
let params: Vec<ValueId> = jp_fn
.env_layout
.and_then(|id| norm.env_layouts.iter().find(|layout| layout.id == id))
.map(|layout| {
layout
.fields
.iter()
.enumerate()
.map(|(idx, f)| f.value_id.unwrap_or(ValueId(idx as u32)))
.collect()
})
.unwrap_or_default();
let mut func = JoinFunction::new(JoinFuncId(jp_id.0), jp_fn.name.clone(), params);
for inst in &jp_fn.body {
match inst {
JpInst::Let { dst, op, args } => match op {
JpOp::Const(v) => func.body.push(JoinInst::Compute(MirLikeInst::Const {
dst: *dst,
value: v.clone(),
})),
JpOp::BoxCall { box_name, method } => {
func.body.push(JoinInst::Compute(MirLikeInst::BoxCall {
dst: Some(*dst),
box_name: box_name.clone(),
method: method.clone(),
args: args.clone(),
}))
}
JpOp::BinOp(op) => func.body.push(JoinInst::Compute(MirLikeInst::BinOp {
dst: *dst,
op: *op,
lhs: args.get(0).copied().unwrap_or(ValueId(0)),
rhs: args.get(1).copied().unwrap_or(ValueId(0)),
})),
JpOp::Unary(op) => func.body.push(JoinInst::Compute(MirLikeInst::UnaryOp {
dst: *dst,
op: *op,
operand: args.get(0).copied().unwrap_or(ValueId(0)),
})),
JpOp::Select => func
.body
.push(JoinInst::Compute(MirLikeInst::Select {
dst: *dst,
cond: args.get(0).copied().unwrap_or(ValueId(0)),
then_val: args.get(1).copied().unwrap_or(ValueId(0)),
else_val: args.get(2).copied().unwrap_or(ValueId(0)),
})),
JpOp::Compare(op) => func.body.push(JoinInst::Compute(MirLikeInst::Compare {
dst: *dst,
op: *op,
lhs: args.get(0).copied().unwrap_or(ValueId(0)),
rhs: args.get(1).copied().unwrap_or(ValueId(0)),
})),
},
JpInst::TailCallFn { target, env } => func.body.push(JoinInst::Call {
func: JoinFuncId(target.0),
args: env.clone(),
k_next: None,
dst: None,
}),
JpInst::TailCallKont { target, env } => func.body.push(JoinInst::Jump {
cont: JoinContId(target.0),
args: env.clone(),
cond: None,
}),
JpInst::If {
cond,
then_target,
else_target,
env,
} => {
func.body.push(JoinInst::Jump {
cont: JoinContId(then_target.0),
args: env.clone(),
cond: Some(*cond),
});
func.body.push(JoinInst::Jump {
cont: JoinContId(else_target.0),
args: env.clone(),
cond: None,
});
}
JpInst::EnvLoad { .. } | JpInst::EnvStore { .. } => {
// Not used in minimal pattern2 bridge
}
}
}
module.add_function(func);
}
module.entry = norm.entry.map(|e| JoinFuncId(e.0));
module.phase = JoinIrPhase::Structured;
module
}
#[cfg(feature = "normalized_dev")]
fn verify_normalized_pattern2(
module: &NormalizedModule,
max_env_fields: usize,
) -> Result<(), String> {
if module.phase != JoinIrPhase::Normalized {
return Err(
"[joinir/normalized-dev] pattern2: phase must be Normalized".to_string(),
);
}
let mut layout_sizes: HashMap<u32, usize> = HashMap::new();
for layout in &module.env_layouts {
let size = layout.fields.len();
if !(1..=max_env_fields).contains(&size) {
return Err(format!(
"[joinir/normalized-dev] pattern2: expected 1..={} env fields, got {}",
max_env_fields, size
));
}
layout_sizes.insert(layout.id, size);
}
for func in module.functions.values() {
let expected_env_len = func
.env_layout
.and_then(|id| layout_sizes.get(&id))
.copied();
for inst in &func.body {
match inst {
JpInst::Let { .. }
| JpInst::EnvLoad { .. }
| JpInst::EnvStore { .. }
| JpInst::TailCallFn { .. }
| JpInst::TailCallKont { .. }
| JpInst::If { .. } => {}
}
match inst {
JpInst::TailCallFn { env, .. }
| JpInst::TailCallKont { env, .. }
| JpInst::If { env, .. } => {
if let Some(expected) = expected_env_len {
if env.is_empty() {
return Err(
"[joinir/normalized-dev] pattern2: env must not be empty"
.to_string(),
);
}
let _ = expected;
}
}
_ => {}
}
}
if let Some(last) = func.body.last() {
match last {
JpInst::TailCallFn { .. }
| JpInst::TailCallKont { .. }
| JpInst::If { .. } => {}
_ => {
return Err(format!(
"[joinir/normalized-dev] pattern2: function '{}' does not end with tail call/if",
func.name
));
}
}
}
}
Ok(())
}
/// Pattern1 専用のミニ変換。
///
/// 制約:
/// - structured.phase は Structured であること
/// - 対象は Pattern1 のシンプル whilebreak/continue なし)
pub fn normalize_pattern1_minimal(structured: &JoinModule) -> NormalizedModule {
assert!(
structured.is_structured(),
"normalize_pattern1_minimal: expected Structured JoinIR"
);
// entry/loop_step/k_exit を前提に、loop_step を拾う
let loop_func = structured
.functions
.values()
.find(|f| f.name == "loop_step")
.or_else(|| structured.functions.get(&JoinFuncId::new(1)))
.expect("normalize_pattern1_minimal: loop_step not found");
// EnvLayout をざっくり作る(フィールド名は field0, field1,... で代用)
let env_layout = EnvLayout {
id: 0,
fields: loop_func
.params
.iter()
.enumerate()
.map(|(idx, vid)| EnvField {
name: format!("field{}", idx),
ty: None,
value_id: Some(*vid),
})
.collect(),
};
// loop_step の Compute を Let に写経Pattern1 では Compute/Call/Ret のみ想定)
let mut extra_konts: HashSet<JpFuncId> = HashSet::new();
let mut jp_body = Vec::new();
for inst in &loop_func.body {
match inst {
JoinInst::Compute(MirLikeInst::Const { dst, value }) => jp_body.push(JpInst::Let {
dst: *dst,
op: JpOp::Const(value.clone()),
args: vec![],
}),
JoinInst::Compute(MirLikeInst::BoxCall {
dst,
box_name,
method,
args,
}) => {
if let Some(dst) = dst {
jp_body.push(JpInst::Let {
dst: *dst,
op: JpOp::BoxCall {
box_name: box_name.clone(),
method: method.clone(),
},
args: args.clone(),
})
}
}
JoinInst::Compute(MirLikeInst::BinOp { dst, op, lhs, rhs }) => {
jp_body.push(JpInst::Let {
dst: *dst,
op: JpOp::BinOp(*op),
args: vec![*lhs, *rhs],
})
}
JoinInst::Compute(MirLikeInst::UnaryOp { dst, op, operand }) => {
jp_body.push(JpInst::Let {
dst: *dst,
op: JpOp::Unary(*op),
args: vec![*operand],
})
}
JoinInst::Compute(MirLikeInst::Compare { dst, op, lhs, rhs }) => {
jp_body.push(JpInst::Let {
dst: *dst,
op: JpOp::Compare(*op),
args: vec![*lhs, *rhs],
})
}
// Tail recursion / exit は TailCall と If でざっくり表現
JoinInst::Jump { cont, args, cond } => {
if let Some(cond_val) = cond {
extra_konts.insert(JpFuncId(cont.0));
jp_body.push(JpInst::If {
cond: *cond_val,
then_target: JpFuncId(cont.0),
else_target: JpFuncId(loop_func.id.0),
env: args.clone(),
});
} else {
extra_konts.insert(JpFuncId(cont.0));
jp_body.push(JpInst::TailCallKont {
target: JpFuncId(cont.0),
env: args.clone(),
});
}
}
JoinInst::Call { func, args, .. } => jp_body.push(JpInst::TailCallFn {
target: JpFuncId(func.0),
env: args.clone(),
}),
JoinInst::Ret { value } => {
if let Some(v) = value {
let kont_id = JpFuncId(loop_func.id.0 + 1);
extra_konts.insert(kont_id);
jp_body.push(JpInst::TailCallKont {
target: kont_id,
env: vec![*v],
});
}
}
_ => {
// Pattern1 の最小変換なので他は無視(将来拡張)
}
}
}
let loop_fn = JpFunction {
id: JpFuncId(loop_func.id.0),
name: loop_func.name.clone(),
env_layout: Some(env_layout.id),
body: jp_body,
is_kont: false,
};
let mut functions = BTreeMap::new();
functions.insert(loop_fn.id, loop_fn);
for kont_id in extra_konts {
functions.entry(kont_id).or_insert_with(|| JpFunction {
id: kont_id,
name: format!("kont_{}", kont_id.0),
env_layout: Some(env_layout.id),
body: Vec::new(),
is_kont: true,
});
}
let norm = NormalizedModule {
functions,
entry: Some(JpFuncId(loop_func.id.0)),
env_layouts: vec![env_layout],
phase: JoinIrPhase::Normalized,
structured_backup: Some(structured.clone()),
};
#[cfg(feature = "normalized_dev")]
{
verify_normalized_pattern1(&norm).expect("normalized verifier");
}
norm
}
/// Dev helper: Structured → Normalized → Structured roundtrip (Pattern1/2 minis only).
#[cfg(feature = "normalized_dev")]
pub(crate) fn normalized_dev_roundtrip_structured(
module: &JoinModule,
) -> Result<JoinModule, String> {
if !module.is_structured() {
return Err("[joinir/normalized-dev] expected Structured JoinModule".to_string());
}
let shapes = shape_guard::supported_shapes(module);
if shapes.is_empty() {
return Err("[joinir/normalized-dev] module shape is not supported by normalized_dev".into());
}
let debug = dev_env::normalized_dev_logs_enabled() && crate::config::env::joinir_dev_enabled();
for shape in shapes {
if debug {
eprintln!(
"[joinir/normalized-dev/roundtrip] attempting {:?} normalization",
shape
);
}
let attempt = match shape {
NormalizedDevShape::Pattern1Mini => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern1_minimal(module);
normalized_pattern1_to_structured(&norm)
})),
NormalizedDevShape::Pattern2Mini
| NormalizedDevShape::JsonparserSkipWsMini
| NormalizedDevShape::JsonparserSkipWsReal
| NormalizedDevShape::JsonparserAtoiMini
| NormalizedDevShape::JsonparserAtoiReal
| NormalizedDevShape::JsonparserParseNumberReal => catch_unwind(AssertUnwindSafe(
|| {
let norm = normalize_pattern2_minimal(module);
normalized_pattern2_to_structured(&norm)
},
)),
NormalizedDevShape::SelfhostTokenScanP2 => catch_unwind(AssertUnwindSafe(|| {
let norm =
normalize_selfhost_token_scan_p2(module).expect("selfhost P2 normalization failed");
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::SelfhostTokenScanP2Accum => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_selfhost_token_scan_p2_accum(module)
.expect("selfhost P2 accum normalization failed");
normalized_pattern2_to_structured(&norm)
})),
// Phase 47-A: P3 minimal (delegates to P2 for now, but uses proper guard)
NormalizedDevShape::Pattern3IfSumMinimal => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern3_if_sum_minimal(module)
.expect("P3 normalization failed");
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::Pattern3IfSumMulti => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern3_if_sum_multi_minimal(module)
.expect("P3 multi normalization failed");
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::Pattern3IfSumJson => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern3_if_sum_json_minimal(module)
.expect("P3 json normalization failed");
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::SelfhostIfSumP3 => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_selfhost_if_sum_p3(module)
.expect("selfhost P3 normalization failed");
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::SelfhostIfSumP3Ext => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_selfhost_if_sum_p3_ext(module)
.expect("selfhost P3 ext normalization failed");
normalized_pattern2_to_structured(&norm)
})),
// Phase 53: selfhost P2/P3 practical variations (delegate to existing normalizers)
NormalizedDevShape::SelfhostArgsParseP2 => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern2_minimal(module);
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::SelfhostStmtCountP3 => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_selfhost_if_sum_p3_ext(module)
.expect("selfhost stmt_count P3 normalization failed");
normalized_pattern2_to_structured(&norm)
})),
// Phase 54: selfhost P2/P3 shape growth (delegate to existing normalizers)
NormalizedDevShape::SelfhostVerifySchemaP2 => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern2_minimal(module);
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::SelfhostDetectFormatP3 => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_selfhost_if_sum_p3_ext(module)
.expect("selfhost detect_format P3 normalization failed");
normalized_pattern2_to_structured(&norm)
})),
// Phase 48-A: P4 minimal (delegates to P2 for now, but uses proper guard)
NormalizedDevShape::Pattern4ContinueMinimal => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern4_continue_minimal(module)
.expect("P4 normalization failed");
normalized_pattern2_to_structured(&norm)
})),
NormalizedDevShape::JsonparserParseArrayContinueSkipWs => {
catch_unwind(AssertUnwindSafe(|| {
let norm =
normalize_jsonparser_parse_array_continue_skip_ws(module)
.expect("P4 array normalization failed");
normalized_pattern2_to_structured(&norm)
}))
}
NormalizedDevShape::JsonparserParseObjectContinueSkipWs => {
catch_unwind(AssertUnwindSafe(|| {
let norm =
normalize_jsonparser_parse_object_continue_skip_ws(module)
.expect("P4 object normalization failed");
normalized_pattern2_to_structured(&norm)
}))
}
// Phase 89: Continue + Early Return pattern (dev-only, delegates to P2 for now)
NormalizedDevShape::PatternContinueReturnMinimal => catch_unwind(AssertUnwindSafe(|| {
let norm = normalize_pattern2_minimal(module);
normalized_pattern2_to_structured(&norm)
})),
};
match attempt {
Ok(structured) => {
if debug {
eprintln!(
"[joinir/normalized-dev/roundtrip] {:?} normalization succeeded (functions={})",
shape,
structured.functions.len()
);
}
return Ok(structured);
}
Err(_) => {
if debug {
eprintln!(
"[joinir/normalized-dev/roundtrip] {:?} normalization failed (unsupported)",
shape
);
}
}
}
}
Err("[joinir/normalized-dev] all normalization attempts failed".into())
}
#[cfg(test)]
mod tests {
use super::*;
fn build_structured_pattern1() -> JoinModule {
let mut module = JoinModule::new();
let mut loop_fn = JoinFunction::new(
JoinFuncId::new(1),
"loop_step".to_string(),
vec![ValueId(10)],
);
loop_fn.body.push(JoinInst::Compute(MirLikeInst::Const {
dst: ValueId(11),
value: ConstValue::Integer(0),
}));
loop_fn.body.push(JoinInst::Compute(MirLikeInst::BinOp {
dst: ValueId(12),
op: BinOpKind::Add,
lhs: ValueId(10),
rhs: ValueId(11),
}));
loop_fn.body.push(JoinInst::Jump {
cont: JoinContId(2),
args: vec![ValueId(12)],
cond: Some(ValueId(12)), // dummy
});
let mut k_exit =
JoinFunction::new(JoinFuncId::new(2), "k_exit".to_string(), vec![ValueId(12)]);
k_exit.body.push(JoinInst::Ret {
value: Some(ValueId(12)),
});
module.entry = Some(loop_fn.id);
module.add_function(loop_fn);
module.add_function(k_exit);
module
}
#[test]
fn normalized_pattern1_minimal_smoke() {
let structured = build_structured_pattern1();
let normalized = normalize_pattern1_minimal(&structured);
assert_eq!(normalized.phase, JoinIrPhase::Normalized);
assert!(!normalized.env_layouts.is_empty());
assert!(!normalized.functions.is_empty());
#[cfg(feature = "normalized_dev")]
{
verify_normalized_pattern1(&normalized).expect("verifier should pass");
}
let restored = normalized.to_structured().expect("backup");
assert!(restored.is_structured());
assert_eq!(restored.functions.len(), structured.functions.len());
}
#[test]
fn normalized_pattern1_roundtrip_structured_equivalent() {
let structured = build_structured_pattern1();
let normalized = normalize_pattern1_minimal(&structured);
let reconstructed = normalized_pattern1_to_structured(&normalized);
assert!(reconstructed.is_structured());
assert_eq!(reconstructed.functions.len(), structured.functions.len());
}
}