VoxelForge/Standalone/Research/WorldAssemblerM3/Linux135/RunL135Profiles.py

#!/usr/bin/env python3
"""Collect final-stamp cave, candidate, proof, and startup counters by stratum."""
from __future__ import annotations

import json
import re
import subprocess
import time
from collections import defaultdict
from pathlib import Path

ROOT = Path(__file__).resolve().parents[4]
OUT = ROOT / "Saved/Linux/l135"
BIN = OUT / "bin/VoxelForgeCoreHeadless-clang-l135"
DATA = ROOT / "Standalone/data/owner_world.json"
STAMP = "8dc4294ce0250a591ee4515471c3c547e796ffa4346ab44f0ae1740bbc3797f5"
CAMERAS = {
    0: (8.0, 16.0, -372.2626953125),
    1: (25.8369140625, 3.19921875, -737.0),
    2: (19.166015625, -14.2001953125, -753.3740234375),
    3: (-20.0625, -5.068359375, -900.353515625),
}
STAMP_RE = re.compile(r"^\[VoxelForgeSourceStamp\] sha256=([0-9a-f]+)$", re.M)
PLAN_RE = re.compile(r"^\[WorldAssemblerPlanTiming\] total_ms=([0-9.]+) status=([^ ]+)$", re.M)
CAVE_RE = re.compile(r"^\[CavernPlannerV1\].* plan_ms=([0-9.]+) status=([^ ]+)$", re.M)
TILE_RE = re.compile(r"^\[WorldAssemblerTilePipeline\] (.*)$", re.M)
CLASS_RE = re.compile(r"^\[WorldAssemblerLODClassification\] (.*)$", re.M)
COMPOSER_RE = re.compile(r"^\[PassageComposerTile\] (.*)$", re.M)
FIELDS_RE = re.compile(r"([a-zA-Z0-9_]+)=([^ ]+)")


def fields(line: str) -> dict[str, str]:
    return dict(FIELDS_RE.findall(line))


def main() -> int:
    if not BIN.is_file() or not DATA.is_file():
        raise SystemExit("missing L135 profile binary or owner fixture")
    rows = []
    for strate, xyz in CAMERAS.items():
        camera = ",".join(f"{value:.12f}" for value in xyz) + ",0,0"
        log = OUT / f"profile-final-speed2-s{strate}.log"
        command = ["taskset", "-c", "0-11", "prlimit", "--as=6442450944", "--",
            "timeout", "1740s", str(BIN), str(DATA), "--world-assembler-prototype",
            "--stream-benchmark", "--stream-pipeline", "--workers=8",
            "--clip-radius=3", "--max-clip-level=8", "--max-level=8",
            f"--camera={camera}", "--profile-lod-work", "--profile-passage-composer",
            "--stream-scheduler-policy=3"]
        started = time.monotonic()
        with log.open("w", encoding="utf-8") as output:
            result = subprocess.run(command, cwd=ROOT, stdout=output,
                stderr=subprocess.STDOUT, check=False, timeout=1740)
        content = log.read_text(encoding="utf-8", errors="replace")
        found_stamp = STAMP_RE.search(content)
        if result.returncode or found_stamp is None or found_stamp.group(1) != STAMP:
            raise SystemExit(f"final-stamp profile failed: {log}, exit={result.returncode}")
        plan, cave = PLAN_RE.search(content), CAVE_RE.search(content)
        tile = TILE_RE.search(content)
        if plan is None or cave is None or tile is None:
            raise SystemExit(f"profile counters missing: {log}")
        aggregate = defaultdict(lambda: defaultdict(int))
        for match in CLASS_RE.finditer(content):
            v = fields(match.group(1)); lod = int(v["lod"])
            for key in ("tiles", "tile_skips", "density_point_calls", "field_eval_sample_points",
                        "proof_air_cells", "proof_solid_cells"):
                aggregate[lod][key] = int(v.get(key, "0"))
            aggregate[lod]["field_eval_ms_est"] = float(v.get("field_eval_ms_est", "0"))
        for match in COMPOSER_RE.finditer(content):
            v = fields(match.group(1)); lod = int(v["key"].split("lod", 1)[1].rstrip(")"))
            aggregate[lod]["composer_tiles"] += 1
            for key in ("samples", "native_cave_samples", "room_candidates", "edge_candidates",
                        "features", "route_candidates", "passage_iterations"):
                aggregate[lod][key] += int(v.get(key, "0"))
            aggregate[lod]["composer_worker_ms"] += float(v.get("worker_ms", "0"))
        for lod, values in aggregate.items():
            calls = values.get("density_point_calls", 0)
            values["field_eval_ns_per_density_call_est"] = round(
                values.get("field_eval_ms_est", 0.0) * 1.0e6 / calls, 1) if calls else None
            composer_tiles = values.get("composer_tiles", 0)
            values["room_candidates_per_composer_tile"] = round(
                values.get("room_candidates", 0) / composer_tiles, 3) if composer_tiles else None
            values["edge_candidates_per_composer_tile"] = round(
                values.get("edge_candidates", 0) / composer_tiles, 3) if composer_tiles else None
        tile_fields = fields(tile.group(1))
        tile_summary = {}
        for key, value in tile_fields.items():
            try:
                tile_summary[key] = int(value)
            except ValueError:
                try:
                    tile_summary[key] = float(value)
                except ValueError:
                    continue
        row = {"strate": strate, "source_stamp": found_stamp.group(1),
            "world_assembler_plan_ms": float(plan.group(1)),
            "cavern_planner_ms": float(cave.group(1)),
            "tile_pipeline": tile_summary,
            "lod": {str(k): dict(v) for k, v in sorted(aggregate.items())},
            "wall_s": round(time.monotonic() - started, 3),
            "log": str(log.relative_to(ROOT))}
        rows.append(row)
        print("[L135ProfileRun] " + json.dumps(row, sort_keys=True), flush=True)
    summary = {"status": "PASS", "source_stamp": STAMP, "strata": rows,
        "field_eval_ns_per_density_call_note":
            "Estimate includes all shape-field density calls at this view, not native-cave-exclusive time.",
        "tile_skip_note":
            "tile_pipeline.whole_tile_skips and lod.tile_skips are aggregate classifiers; they are not cave-only attribution."}
    (OUT / "profile-summary-speed2.json").write_text(
        json.dumps(summary, indent=2) + "\n", encoding="utf-8")
    print("[L135ProfileSummary] strata=4 status=PASS")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())