"""Recalcula los indicadores del artículo desde el corte original y compara el informe.""" import argparse from collections import Counter import csv from decimal import Decimal import hashlib import io import json import math from pathlib import Path import re import unicodedata SOURCE_SHA256 = "febff55e3612fb60076df9279f1090240cd02a39e9c7b79b822119dfdef84c8d" def has_location(lat, lon): values = [lat.strip(), lon.strip()] if not all(re.fullmatch(r"[+-]?[0-9]+(?:[.,][0-9]+)?", value) for value in values): return False latitude, longitude = [Decimal(value.replace(",", ".")) for value in values] return (latitude, longitude) != (0, 0) and abs(latitude) <= 90 and abs(longitude) <= 180 def main(): parser = argparse.ArgumentParser(description=__doc__) parser.add_argument("csv", type=Path) parser.add_argument("informe", type=Path) args = parser.parse_args() raw = args.csv.read_bytes() if hashlib.sha256(raw).hexdigest() != SOURCE_SHA256: raise SystemExit("La huella del CSV no corresponde al corte del artículo.") reader = csv.reader(io.StringIO(raw.decode("utf-8-sig"), newline=""), delimiter=";") header, *records = (row for row in reader if row) id_col, name_col, region_col, state_col, lat_col, lon_col = [ header.index(name) for name in ("EstablecimientoCodigo", "EstablecimientoGlosa", "RegionCodigo", "EstadoFuncionamiento", "Latitud", "Longitud") ] ids, names, states = Counter(), Counter(), Counter() total_vigentes, sin_ubicacion = Counter(), Counter() missing_by_state = Counter() empty_cells = missing_pairs = invalid_pairs = 0 for row in records: empty_cells += sum(value.strip() == "" for value in row) identifier = row[id_col].strip() ids[identifier] += 1 normalized_name = " ".join(unicodedata.normalize("NFC", row[name_col]).split()).casefold() if normalized_name: names[normalized_name] += 1 lat, lon = row[lat_col], row[lon_col] located = has_location(lat, lon) both_empty = not (lat.strip() or lon.strip()) missing_pairs += both_empty invalid_pairs += not located and not both_empty state = row[state_col].strip().casefold() states[state] += 1 if both_empty: missing_by_state[state] += 1 if state == "vigente en operación habitual": region = row[region_col].strip() total_vigentes[region] += 1 sin_ubicacion[region] += not located expected = { "sha256_fuente": SOURCE_SHA256, "filas": len(records), "columnas": len(header), "celdas_vacias": empty_cells, "grupos_id_repetido": sum(value > 1 for key, value in ids.items() if key), "grupos_nombre_repetido": sum(value > 1 for value in names.values()), "aceptadas": len(records), "en_cuarentena": 0, "coordenadas": {"sin_par_coordenadas": missing_pairs, "par_coordenadas_invalido": invalid_pairs}, "estados_aceptados": {"vigente_habitual": states["vigente en operación habitual"], "cerrado": states["cerrado"]}, "cobertura_vigentes_por_region": { region: {"vigentes": count, "sin_ubicacion": sin_ubicacion[region]} for region, count in total_vigentes.items() }, } report = json.loads(args.informe.read_text(encoding="utf-8")) discrepancies = [field for field, value in expected.items() if report.get(field) != value] completeness = 1 - empty_cells / (len(records) * len(header)) if not math.isclose(report.get("completitud_bruta", -1), completeness, rel_tol=0, abs_tol=1e-12): discrepancies.append("completitud_bruta") if discrepancies: raise SystemExit("Indicadores que no coinciden: " + ", ".join(discrepancies)) print(f"Verificado: {len(records)} registros, {sum(sin_ubicacion.values())} vigentes sin ubicación " f"({sin_ubicacion['04']} en Coquimbo) y {missing_by_state['cerrado']} cerrados sin coordenadas.") if __name__ == "__main__": main()