#!/usr/bin/env python3
"""Build a clean Hugging Face artifact bundle from current repo outputs."""

from __future__ import annotations

import argparse
import json
import shutil
import sys
from pathlib import Path
from typing import Iterable, List

ROOT = Path(__file__).resolve().parent.parent
if str(ROOT) not in sys.path:
    sys.path.insert(0, str(ROOT))

from scripts.hf_training_contract import ARTIFACT_ROOT_DEFAULT, build_artifact_layout
from scripts.generate_defense_packet import _latest_summary


RESULTS_ROOT = ROOT / "training" / "kan_bench_results"
DEFAULT_DOCS = [
    ROOT / "docs" / "defense_packet.md",
    ROOT / "docs" / "defense_packet.json",
    ROOT / "docs" / "patent_evidence_packet.md",
    ROOT / "docs" / "patents" / "CROSS_REFERENCE_MATRIX.md",
    ROOT / "docs" / "patents" / "REPO_PATENT_INVENTORY.md",
]


def _copy_many(paths: Iterable[Path], dest: Path, keep_parent: bool = False) -> List[str]:
    copied: List[str] = []
    dest.mkdir(parents=True, exist_ok=True)
    for path in paths:
        if not path.exists():
            continue
        filename = f"{path.parent.name}__{path.name}" if keep_parent else path.name
        target = dest / filename
        shutil.copy2(path, target)
        copied.append(str(target.resolve().relative_to(ROOT.resolve())))
    return copied


def _latest_verification_summaries() -> List[Path]:
    out: List[Path] = []
    for prefix in ["smoke", "benchmark", "benchmark-matrix", "ablation", "patent-evidence"]:
        latest = _latest_summary(prefix)
        if latest is not None:
            rel, _ = latest
            out.append(ROOT / rel)
    return out


def build_bundle(root: str = ARTIFACT_ROOT_DEFAULT) -> dict:
    layout = build_artifact_layout(root).ensure()
    if layout.root.exists():
        shutil.rmtree(layout.root)
    layout = build_artifact_layout(root).ensure()
    docs_dir = layout.root / "docs"
    verify_dir = layout.root / "verification"

    result_files = [
        RESULTS_ROOT / "text2cypher_v4_results.json",
        RESULTS_ROOT / "spider2_v2_results.json",
        RESULTS_ROOT / "unified_dialect_results.json",
        RESULTS_ROOT / "swebench_results.json",
        RESULTS_ROOT / "sota_comparison_table.json",
    ]
    checkpoint_files = [
        RESULTS_ROOT / "sota_text2cypher_checkpoint.pt",
        RESULTS_ROOT / "sota_spider2_checkpoint.pt",
        RESULTS_ROOT / "sota_unified_checkpoint.pt",
        RESULTS_ROOT / "sota_swebench_checkpoint.pt",
        RESULTS_ROOT / "sota_gaia_checkpoint.pt",
    ]

    copied = {
        "results": _copy_many(result_files, layout.results_dir),
        "checkpoints": _copy_many(checkpoint_files, layout.checkpoints_dir),
        "docs": _copy_many(DEFAULT_DOCS, docs_dir),
        "verification": _copy_many(_latest_verification_summaries(), verify_dir, keep_parent=True),
    }

    manifest = {
        "artifact_root": str(layout.root.resolve().relative_to(ROOT.resolve())),
        "results": copied["results"],
        "checkpoints": copied["checkpoints"],
        "docs": copied["docs"],
        "verification": copied["verification"],
    }
    manifest_path = layout.root / "manifest.json"
    manifest_path.write_text(json.dumps(manifest, indent=2))
    copied["manifest"] = [str(manifest_path.resolve().relative_to(ROOT.resolve()))]
    return copied


def main() -> int:
    parser = argparse.ArgumentParser(description="Package current repo artifacts for HF upload")
    parser.add_argument("--artifact-root", default=ARTIFACT_ROOT_DEFAULT)
    args = parser.parse_args()
    payload = build_bundle(args.artifact_root)
    print(json.dumps(payload, indent=2))
    return 0


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