Spaces:
Running
Running
| """Valida arenas/*.yaml y, si se indica la carpeta del dataset, también el manifiesto. | |
| Uso: | |
| python tools/validate.py # solo los yaml | |
| python tools/validate.py /ruta/al/dataset # yaml + manifiesto + imágenes | |
| """ | |
| from __future__ import annotations | |
| import sys | |
| from pathlib import Path | |
| sys.path.insert(0, str(Path(__file__).resolve().parent.parent)) | |
| import yaml # noqa: E402 | |
| from app.config import DEFAULT_ARENAS_DIR, ArenaConfig, validate_arena_dict # noqa: E402 | |
| from app.db import Database # noqa: E402 | |
| from app.ingest import ingest # noqa: E402 | |
| def main() -> int: | |
| rc = 0 | |
| print(f"Leyendo arenas de: {DEFAULT_ARENAS_DIR}") | |
| paths = sorted(DEFAULT_ARENAS_DIR.glob("*.yaml")) | |
| if not paths: | |
| print(" (no hay ningún .yaml en esa carpeta)") | |
| return 1 | |
| for path in paths: | |
| raw = yaml.safe_load(path.read_text(encoding="utf-8")) | |
| errs = validate_arena_dict(raw) | |
| if errs: | |
| rc = 1 | |
| print(f"✗ {path}\n - " + "\n - ".join(errs)) | |
| continue | |
| cfg = ArenaConfig(raw) | |
| cols = cfg.columns | |
| print(f"✓ {path}: yaml válido") | |
| print(f" candidatos: {cfg.candidate_keys}") | |
| print(f" columnas esperadas en el CSV (según el bloque `columns` de este fichero): " | |
| + ", ".join(f"{cols[k]}→{k}" for k in sorted(cols))) | |
| if len(sys.argv) > 1: | |
| root = Path(sys.argv[1]) | |
| manifest = root / cfg.data["manifest"] | |
| images = root / cfg.data.get("images_dir", "") | |
| print(f" manifiesto: {manifest}") | |
| rep = ingest(cfg, Database(), manifest, images, "salt") | |
| for w in rep.warnings: | |
| print(f" ! {w}") | |
| if rep.ok: | |
| print(f" ✓ manifiesto ok: {rep.n_cases} casos, {rep.n_assets} imágenes, " | |
| f"{rep.n_comparisons} comparaciones, segmentos {rep.segments}, " | |
| f"parámetros {rep.param_keys}") | |
| else: | |
| rc = 1 | |
| shown = rep.errors[:15] | |
| print(" ✗ manifiesto inválido:\n - " + "\n - ".join(shown)) | |
| if len(rep.errors) > len(shown): | |
| print(f" ... y {len(rep.errors) - len(shown)} errores más") | |
| return rc | |
| if __name__ == "__main__": | |
| sys.exit(main()) | |