from pathlib import Path
import argparse,json
import pandas as pd
import joblib

ROOT=Path(__file__).resolve().parents[1]
MODELS=ROOT/"models"

def load_artifacts():
    metadata=json.loads((MODELS/"model_metadata.json").read_text(encoding="utf-8"))
    model=joblib.load(MODELS/"final_model.joblib")
    medians={}
    median_file=MODELS/"catboost_numeric_medians.json"
    if median_file.exists():
        medians=json.loads(median_file.read_text(encoding="utf-8"))
    preprocessor=None
    pp=MODELS/"final_preprocessor.joblib"
    if pp.exists():
        preprocessor=joblib.load(pp)
    return metadata,model,medians,preprocessor

def _prepare_catboost(features,metadata,medians):
    required=metadata["feature_columns"]
    row=pd.DataFrame([{c:features.get(c) for c in required}])
    for c in metadata["numeric_columns"]:
        row[c]=pd.to_numeric(row[c],errors="coerce").fillna(float(medians[c]))
    for c in metadata["categorical_columns"]:
        row[c]=row[c].astype("object").where(row[c].notna(),"MISSING").astype(str)
    return row[required]

def predict_next7d_occupancy(features:dict)->dict:
    metadata,model,medians,preprocessor=load_artifacts()
    required=metadata["feature_columns"]
    missing=[c for c in required if c not in features]
    if missing:
        raise ValueError(f"Missing required features: {missing}")

    if metadata["model_kind"]=="catboost_native":
        row=_prepare_catboost(features,metadata,medians)
        pred=float(model.predict(row)[0])
    else:
        row=pd.DataFrame([{c:features.get(c) for c in required}])
        pred=float(model.predict(preprocessor.transform(row))[0])

    pred=max(0.0,min(1.0,pred))
    on_books=float(features["on_books_occupancy_next_7d"])
    expected_increment=max(0.0,pred-on_books)

    return {
        "predicted_next_7d_occupancy_rate":round(pred,6),
        "predicted_next_7d_occupancy_percent":round(pred*100,2),
        "on_books_occupancy_rate":round(on_books,6),
        "expected_additional_occupancy_rate":round(expected_increment,6),
        "model_name":metadata["selected_algorithm"],
        "forecast_window":"next 7 days",
        "decision_support_only":True,
        "synthetic_training_data_warning":metadata["synthetic_training_data"],
        "production_warning":metadata["production_warning"]
    }

def main():
    parser=argparse.ArgumentParser()
    parser.add_argument("json_file")
    args=parser.parse_args()
    payload=json.loads(Path(args.json_file).read_text(encoding="utf-8"))
    print(json.dumps(
        predict_next7d_occupancy(payload.get("features",payload)),
        ensure_ascii=False,indent=2
    ))

if __name__=="__main__":
    main()