hakorune/tools/hakorune_emit_mir.sh

#!/usr/bin/env bash
# hakorune_emit_mir.sh — Emit MIR(JSON) using Hakorune Stage‑B + MirBuilder (Hako‑first)
#
# Usage: tools/hakorune_emit_mir.sh <input.hako> <out.json>
# Notes:
#  - Runs the Stage‑B compiler (Hako) to emit Program(JSON v0), then feeds it to MirBuilderBox.emit_from_program_json_v0.
#  - Defaults to the Hakorune VM path; no inline Ny compiler; Stage‑3 enabled.
#  - Keeps defaults conservative: no global flips; this is a helper for dev/CI scripts.

set -euo pipefail

if [ "$#" -ne 2 ]; then
  echo "Usage: $0 <input.hako> <out.json>" >&2
  exit 2
fi

IN="$1"
OUT="$2"

SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
if ROOT_GIT=$(git -C "$SCRIPT_DIR" rev-parse --show-toplevel 2>/dev/null); then
  ROOT="$ROOT_GIT"
else
  ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
fi

# Resolve nyash/hakorune binary via test_runner helper (ensures consistent env)
if [ ! -f "$IN" ]; then
  echo "[FAIL] input not found: $IN" >&2
  exit 1
fi

# Resolve nyash/hakorune binary (simple detection; test_runner will override later if sourced)
if [ -z "${NYASH_BIN:-}" ]; then
  if [ -x "$ROOT/target/release/hakorune" ]; then
    export NYASH_BIN="$ROOT/target/release/hakorune"
  else
    export NYASH_BIN="$ROOT/target/release/nyash"
  fi
fi

CODE="$(cat "$IN")"

# 1) Stage‑B: Hako parser emits Program(JSON v0) to stdout
set +e
PROG_JSON_OUT=$(NYASH_JSON_ONLY=1 NYASH_DISABLE_NY_COMPILER=1 HAKO_DISABLE_NY_COMPILER=1 \
                HAKO_STAGEB_FUNC_SCAN="${HAKO_STAGEB_FUNC_SCAN:-}" \
                NYASH_PARSER_STAGE3=1 HAKO_PARSER_STAGE3=1 NYASH_PARSER_ALLOW_SEMICOLON=1 \
                NYASH_ENABLE_USING=${NYASH_ENABLE_USING:-1} HAKO_ENABLE_USING=${HAKO_ENABLE_USING:-1} \
                "$NYASH_BIN" --backend vm "$ROOT/lang/src/compiler/entry/compiler_stageb.hako" -- --source "$CODE" 2>/dev/null | awk '/^{/,/^}$/')
rc=$?
set -e
if [ $rc -ne 0 ] || [ -z "$PROG_JSON_OUT" ]; then
  echo "[FAIL] Stage-B parse failed (rc=$rc)" >&2
  exit 1
fi

# Quick validation for Program(JSON v0)
if ! printf '%s' "$PROG_JSON_OUT" | grep -q '"kind"\s*:\s*"Program"'; then
  echo "[FAIL] Stage‑B output is not Program(JSON)" >&2
  printf '%s\n' "$PROG_JSON_OUT" | sed -n '1,80p' >&2 || true
  exit 1
fi

# 2) Convert Program(JSON v0) → MIR(JSON)
#    Prefer selfhost builder first when explicitly requested; otherwise use delegate (Gate‑C) for stability.

try_selfhost_builder() {
  local prog_json="$1" out_path="$2"

  # FORCE=1 direct assembly shortcut (dev toggle, bypasses using resolution)
  if [ "${HAKO_MIR_BUILDER_LOOP_FORCE_JSONFRAG:-0}" = "1" ]; then
    # Extract limit from Program(JSON) using grep/awk
    local limit=$(printf '%s' "$prog_json" | grep -o '"type":"Int","value":[0-9]*' | head -1 | grep -o '[0-9]*$' || echo "10")

    # Generate minimal while-form MIR(JSON) directly (executable semantics)
    # PHI incoming format: [[value_register, predecessor_block_id], ...]
    cat > "$out_path" <<'MIRJSON'
{
  "functions": [{
    "name": "main",
    "params": [],
    "locals": [],
    "blocks": [
      {
        "id": 0,
        "instructions": [
          {"op": "const", "dst": 1, "value": {"type": "i64", "value": 0}},
          {"op": "const", "dst": 2, "value": {"type": "i64", "value": LIMIT_PLACEHOLDER}},
          {"op": "jump", "target": 1}
        ]
      },
      {
        "id": 1,
        "instructions": [
          {"op": "phi", "dst": 6, "incoming": [[2, 0], [6, 2]]},
          {"op": "phi", "dst": 3, "incoming": [[1, 0], [5, 2]]},
          {"op": "compare", "operation": "<", "lhs": 3, "rhs": 6, "dst": 4},
          {"op": "branch", "cond": 4, "then": 2, "else": 3}
        ]
      },
      {
        "id": 2,
        "instructions": [
          {"op": "const", "dst": 10, "value": {"type": "i64", "value": 1}},
          {"op": "binop", "operation": "+", "lhs": 3, "rhs": 10, "dst": 5},
          {"op": "jump", "target": 1}
        ]
      },
      {
        "id": 3,
        "instructions": [
          {"op": "ret", "value": 3}
        ]
      }
    ]
  }]
}
MIRJSON

# Provider-first delegate: call env.mirbuilder.emit(prog_json) and capture v1 JSON
try_provider_emit() {
  local prog_json="$1" out_path="$2"
  local tmp_hako; tmp_hako=$(mktemp --suffix .hako)
  cat >"$tmp_hako" <<'HCODE'
using "hako.mir.builder.internal.jsonfrag_normalizer" as NormBox
static box Main { method main(args) {
  local p = env.get("HAKO_BUILDER_PROGRAM_JSON")
  if p == null { print("[provider/emit:nojson]"); return 1 }
  local a = new ArrayBox(); a.push(p)
  local out = hostbridge.extern_invoke("env.mirbuilder", "emit", a)
  // Optional normalization (dev): apply JsonFrag normalizer/purifier to provider output
  {
    local nv = env.get("HAKO_MIR_NORMALIZE_PROVIDER")
    if nv != null && ("" + nv) == "1" {
      local out_s = "" + out
      out = NormBox.normalize_all(out_s)
    }
  }
  print("[provider/emit:ok]")
  print("[MIR_OUT_BEGIN]")
  print("" + out)
  print("[MIR_OUT_END]")
  return 0
} }
HCODE
  local tmp_stdout; tmp_stdout=$(mktemp)
  trap 'rm -f "$tmp_hako" "$tmp_stdout" || true' RETURN
  set +e
  (cd "$ROOT" && \
    NYASH_DISABLE_PLUGINS=1 NYASH_FILEBOX_MODE="core-ro" \
    NYASH_PARSER_STAGE3=1 HAKO_PARSER_STAGE3=1 NYASH_PARSER_ALLOW_SEMICOLON=1 \
    HAKO_BUILDER_PROGRAM_JSON="$prog_json" \
    "$NYASH_BIN" --backend vm "$tmp_hako" 2>&1 | tee "$tmp_stdout" >/dev/null)
  local rc=$?
  set -e
  if [ $rc -ne 0 ] || ! grep -q "\[provider/emit:ok\]" "$tmp_stdout"; then
    return 1
  fi
  local mir
  mir=$(awk '/\[MIR_OUT_BEGIN\]/{flag=1;next}/\[MIR_OUT_END\]/{flag=0}flag' "$tmp_stdout")
  if [ -z "$mir" ]; then return 1; fi
  printf '%s' "$mir" > "$out_path"
  echo "[OK] MIR JSON written (delegate:provider): $out_path"
  return 0
}

    # Replace LIMIT_PLACEHOLDER with actual limit
    sed -i "s/LIMIT_PLACEHOLDER/$limit/g" "$out_path"

    if [ "${HAKO_SELFHOST_TRACE:-0}" = "1" ]; then
      echo "[selfhost-direct:ok] Direct MIR assembly (FORCE=1), limit=$limit" >&2
    fi

    return 0
  fi

  # Builder box selection (default: hako.mir.builder)
  local builder_box="${HAKO_MIR_BUILDER_BOX:-hako.mir.builder}"

  local tmp_hako; tmp_hako=$(mktemp --suffix .hako)
  if [ "$builder_box" = "hako.mir.builder.min" ]; then
    cat >"$tmp_hako" <<'HCODE'
using "hako.mir.builder.internal.runner_min" as BuilderRunnerMinBox
static box Main { method main(args) {
  local prog_json = env.get("HAKO_BUILDER_PROGRAM_JSON")
  if prog_json == null { print("[builder/selfhost-first:fail:nojson]"); return 1 }
  local mir_out = BuilderRunnerMinBox.run(prog_json)
  if mir_out == null { print("[builder/selfhost-first:fail:emit]"); return 1 }
  print("[builder/selfhost-first:ok]")
  print("[MIR_OUT_BEGIN]")
  print("" + mir_out)
  print("[MIR_OUT_END]")
  return 0
} }
HCODE
  else
    cat >"$tmp_hako" <<'HCODE'
using "__BUILDER_BOX__" as MirBuilderBox
static box Main { method main(args) {
  local prog_json = env.get("HAKO_BUILDER_PROGRAM_JSON")
  if prog_json == null { print("[builder/selfhost-first:fail:nojson]"); return 1 }
  local mir_out = MirBuilderBox.emit_from_program_json_v0(prog_json, null)
  if mir_out == null { print("[builder/selfhost-first:fail:emit]"); return 1 }
  print("[builder/selfhost-first:ok]")
  print("[MIR_OUT_BEGIN]")
  print("" + mir_out)
  print("[MIR_OUT_END]")
  return 0
} }
HCODE
    sed -i "s|__BUILDER_BOX__|$builder_box|g" "$tmp_hako"
  fi
  local tmp_stdout; tmp_stdout=$(mktemp)
  trap 'rm -f "$tmp_hako" "$tmp_stdout" || true' RETURN

  # Trace mode: analyze Program(JSON) before passing to builder
  if [ "${HAKO_SELFHOST_TRACE:-0}" = "1" ]; then
    local prog_len=${#prog_json}
    local loop_count=$(printf '%s' "$prog_json" | grep -o '"type":"Loop"' 2>/dev/null | wc -l | tr -d ' \n')
    local cmp_count=$(printf '%s' "$prog_json" | grep -o '"type":"Compare"' 2>/dev/null | wc -l | tr -d ' \n')
    loop_count=${loop_count:-0}
    cmp_count=${cmp_count:-0}
    local cwd="$(pwd)"
    local toml_status="absent"
    if [ -f "$ROOT/nyash.toml" ]; then
      toml_status="present"
    fi
    echo "[builder/selfhost-first:trace] builder_box=$builder_box prog_json_len=$prog_len tokens=Loop:$loop_count,Compare:$cmp_count cwd=$cwd nyash.toml=$toml_status" >&2
  fi

  set +e
  # Run from repo root to ensure nyash.toml is available for using resolution
  # Capture both stdout and stderr (2>&1) instead of discarding stderr
  (cd "$ROOT" && \
    HAKO_MIR_BUILDER_INTERNAL=1 HAKO_MIR_BUILDER_REGISTRY=1 \
    HAKO_MIR_BUILDER_TRACE="${HAKO_SELFHOST_TRACE:-}" \
    HAKO_MIR_BUILDER_LOOP_JSONFRAG="${HAKO_MIR_BUILDER_LOOP_JSONFRAG:-}" \
    HAKO_MIR_BUILDER_LOOP_FORCE_JSONFRAG="${HAKO_MIR_BUILDER_LOOP_FORCE_JSONFRAG:-}" \
    HAKO_MIR_BUILDER_JSONFRAG_NORMALIZE="${HAKO_MIR_BUILDER_JSONFRAG_NORMALIZE:-}" \
    HAKO_MIR_BUILDER_JSONFRAG_PURIFY="${HAKO_MIR_BUILDER_JSONFRAG_PURIFY:-}" \
    HAKO_MIR_BUILDER_METHODIZE="${HAKO_MIR_BUILDER_METHODIZE:-}" \
    HAKO_MIR_BUILDER_NORMALIZE_TAG="${HAKO_MIR_BUILDER_NORMALIZE_TAG:-}" \
    HAKO_MIR_BUILDER_DEBUG="${HAKO_MIR_BUILDER_DEBUG:-}" \
    NYASH_DISABLE_PLUGINS="${NYASH_DISABLE_PLUGINS:-0}" NYASH_FILEBOX_MODE="core-ro" HAKO_PROVIDER_POLICY="safe-core-first" \
    NYASH_ENABLE_USING=1 HAKO_ENABLE_USING=1 \
    NYASH_PARSER_STAGE3=1 HAKO_PARSER_STAGE3=1 NYASH_PARSER_ALLOW_SEMICOLON=1 \
    NYASH_USE_NY_COMPILER=0 HAKO_USE_NY_COMPILER=0 NYASH_DISABLE_NY_COMPILER=1 HAKO_DISABLE_NY_COMPILER=1 \
    NYASH_MACRO_DISABLE=1 HAKO_MACRO_DISABLE=1 \
    HAKO_BUILDER_PROGRAM_JSON="$prog_json" \
    "$NYASH_BIN" --backend vm "$tmp_hako" 2>&1 | tee "$tmp_stdout" >/dev/null)
  local rc=$?
  set -e

  # Enhanced failure diagnostics
  if [ $rc -ne 0 ]; then
    if [ "${HAKO_SELFHOST_NO_DELEGATE:-0}" = "1" ]; then
      echo "[builder/selfhost-first:fail:child:rc=$rc]" >&2
      echo "[builder/selfhost-first:fail:detail] Last 80 lines of output:" >&2
      tail -n 80 "$tmp_stdout" >&2 || true
    fi
    # Don't return immediately - check for fallback below
  fi

  if [ $rc -eq 0 ] && ! grep -q "\[builder/selfhost-first:ok\]" "$tmp_stdout"; then
    if [ "${HAKO_SELFHOST_NO_DELEGATE:-0}" = "1" ]; then
      echo "[builder/selfhost-first:fail:no-ok-marker]" >&2
      echo "[builder/selfhost-first:fail:detail] Last 80 lines of output:" >&2
      tail -n 80 "$tmp_stdout" >&2 || true
    fi
    rc=1
  fi

  # Try min builder fallback if enabled and initial builder failed
  if [ "${HAKO_SELFHOST_TRY_MIN:-0}" = "1" ] && [ $rc -ne 0 ] && [ "$builder_box" != "hako.mir.builder.min" ]; then
    if [ "${HAKO_SELFHOST_NO_DELEGATE:-0}" = "1" ]; then
      echo "[builder/selfhost-first:trying-min-fallback]" >&2
    fi

    # Retry with min builder
    HAKO_MIR_BUILDER_BOX="hako.mir.builder.min" try_selfhost_builder "$prog_json" "$out_path"
    return $?
  fi

  # Return original failure if no fallback or if fallback not triggered
  if [ $rc -ne 0 ]; then
    return 1
  fi

  local mir
  mir=$(awk '/\[MIR_OUT_BEGIN\]/{flag=1;next}/\[MIR_OUT_END\]/{flag=0}flag' "$tmp_stdout")
  if [ -z "$mir" ]; then return 1; fi
  printf '%s' "$mir" > "$out_path"
  echo "[OK] MIR JSON written (selfhost-first): $out_path"
  return 0
}

# Provider-first delegate: call env.mirbuilder.emit(prog_json) and capture v1 JSON
try_provider_emit() {
  local prog_json="$1" out_path="$2"
  local tmp_hako; tmp_hako=$(mktemp --suffix .hako)
  cat >"$tmp_hako" <<'HCODE'
static box Main { method main(args) {
  local p = env.get("HAKO_BUILDER_PROGRAM_JSON")
  if p == null { print("[provider/emit:nojson]"); return 1 }
  local a = new ArrayBox(); a.push(p)
  local out = hostbridge.extern_invoke("env.mirbuilder", "emit", a)
  print("[provider/emit:ok]")
  print("[MIR_OUT_BEGIN]")
  print("" + out)
  print("[MIR_OUT_END]")
  return 0
} }
HCODE
  local tmp_stdout; tmp_stdout=$(mktemp)
  trap 'rm -f "$tmp_hako" "$tmp_stdout" || true' RETURN
  set +e
  (cd "$ROOT" && \
    NYASH_DISABLE_PLUGINS="${NYASH_DISABLE_PLUGINS:-0}" NYASH_FILEBOX_MODE="core-ro" \
    NYASH_PARSER_STAGE3=1 HAKO_PARSER_STAGE3=1 NYASH_PARSER_ALLOW_SEMICOLON=1 \
    HAKO_BUILDER_PROGRAM_JSON="$prog_json" \
    "$NYASH_BIN" --backend vm "$tmp_hako" 2>&1 | tee "$tmp_stdout" >/dev/null)
  local rc=$?
  set -e
  if [ $rc -ne 0 ] || ! grep -q "\[provider/emit:ok\]" "$tmp_stdout"; then
    return 1
  fi
  local mir
  mir=$(awk '/\[MIR_OUT_BEGIN\]/{flag=1;next}/\[MIR_OUT_END\]/{flag=0}flag' "$tmp_stdout")
  if [ -z "$mir" ]; then return 1; fi
  printf '%s' "$mir" > "$out_path"
  echo "[OK] MIR JSON written (delegate:provider): $out_path"
  return 0
}

# When forcing JSONFrag loop, default-enable normalize+purify (dev-only, no default changes)
if [ "${HAKO_MIR_BUILDER_LOOP_FORCE_JSONFRAG:-0}" = "1" ]; then
  export HAKO_MIR_BUILDER_JSONFRAG_NORMALIZE="${HAKO_MIR_BUILDER_JSONFRAG_NORMALIZE:-1}"
  export HAKO_MIR_BUILDER_JSONFRAG_PURIFY="${HAKO_MIR_BUILDER_JSONFRAG_PURIFY:-1}"
fi

if [ "${HAKO_SELFHOST_BUILDER_FIRST:-0}" = "1" ]; then
  if try_selfhost_builder "$PROG_JSON_OUT" "$OUT"; then
    exit 0
  fi
  if [ "${HAKO_SELFHOST_NO_DELEGATE:-0}" = "1" ]; then
    echo "[FAIL] selfhost-first failed and delegate disabled" >&2
    exit 1
  fi
fi

# Dev: force JsonFrag minimal loop even on provider-first path
if [ "${HAKO_MIR_BUILDER_LOOP_FORCE_JSONFRAG:-0}" = "1" ]; then
  # Extract limit from Program(JSON)
  limit=$(printf '%s' "$PROG_JSON_OUT" | grep -o '"type":"Int","value":[0-9]*' | head -1 | grep -o '[0-9]*$' || echo "10")
  cat > "$OUT" <<MIRJSON
{
  "functions": [{
    "name": "main",
    "params": [],
    "locals": [],
    "blocks": [
      { "id": 0, "instructions": [
        {"op":"const","dst":1,"value":{"type":"i64","value":0}},
        {"op":"const","dst":2,"value":{"type":"i64","value": ${limit} }},
        {"op":"jump","target":1}
      ]},
      { "id": 1, "instructions": [
        {"op":"phi","dst":6,"incoming":[[2,0],[6,2]]},
        {"op":"phi","dst":3,"incoming":[[1,0],[5,2]]},
        {"op":"compare","operation":"<","lhs":3,"rhs":6,"dst":4},
        {"op":"branch","cond":4,"then":2,"else":3}
      ]},
      { "id": 2, "instructions": [
        {"op":"const","dst":10,"value":{"type":"i64","value":1}},
        {"op":"binop","operation":"+","lhs":3,"rhs":10,"dst":5},
        {"op":"jump","target":1}
      ]},
      { "id": 3, "instructions": [
        {"op":"ret","value":3}
      ]}
    ]
  }]
}
MIRJSON
  echo "[OK] MIR JSON written (provider-force-jsonfrag): $OUT"
  exit 0
fi

tmp_prog="/tmp/hako_emit_prog_$$.json"
trap 'rm -f "$tmp_prog" || true' EXIT
printf '%s' "$PROG_JSON_OUT" > "$tmp_prog"
# Provider-first delegate (v1固定): env.mirbuilder.emit を使用
if try_provider_emit "$PROG_JSON_OUT" "$OUT"; then
  exit 0
fi

# 最終フォールバック: 旧CLI変換（環境でv1を促す）
if HAKO_STAGEB_FUNC_SCAN="${HAKO_STAGEB_FUNC_SCAN:-}" \
   HAKO_MIR_BUILDER_FUNCS="${HAKO_MIR_BUILDER_FUNCS:-}" \
   HAKO_MIR_BUILDER_CALL_RESOLVE="${HAKO_MIR_BUILDER_CALL_RESOLVE:-}" \
   NYASH_JSON_SCHEMA_V1=${NYASH_JSON_SCHEMA_V1:-1} \
   NYASH_MIR_UNIFIED_CALL=${NYASH_MIR_UNIFIED_CALL:-1} \
   "$NYASH_BIN" --program-json-to-mir "$OUT" --json-file "$tmp_prog" >/dev/null 2>&1; then
  echo "[OK] MIR JSON written (delegate-legacy): $OUT"
  exit 0
fi
echo "[FAIL] Program→MIR delegate failed (provider+legacy)" >&2
exit 1