import numpy as np
from PIL import Image, ImageDraw
from facenet_pytorch import MTCNN

class FaceCropper:
    """Detect faces and return (cropped_face, annotated_image)."""
    def __init__(self, device: str | None = None, image_size: int = 224):
        import torch
        self.device = device or ("cuda" if torch.cuda.is_available() else "cpu")
        self.mtcnn = MTCNN(keep_all=True, device=self.device)
        self.image_size = image_size

    def _to_pil(self, img):
        if isinstance(img, Image.Image):
            return img.convert("RGB")
        return Image.fromarray(img).convert("RGB")

    def detect_and_crop(self, img, select="largest"):
        pil = self._to_pil(img)
        boxes, probs = self.mtcnn.detect(pil)

        annotated = pil.copy()
        draw = ImageDraw.Draw(annotated)

        if boxes is None or len(boxes) == 0:
            return None, annotated, {"boxes": np.empty((0,4)), "scores": np.empty((0,))}

        for b, p in zip(boxes, probs):
            x1, y1, x2, y2 = map(float, b)
            draw.rectangle([x1, y1, x2, y2], outline=(255, 0, 0), width=3)
            draw.text((x1, y1-10), f"{p:.2f}", fill=(255, 0, 0))

        idx = 0
        if select == "largest":
            areas = [(b[2]-b[0])*(b[3]-b[1]) for b in boxes]
            idx = int(np.argmax(areas))
        elif isinstance(select, int) and 0 <= select < len(boxes):
            idx = select

        x1, y1, x2, y2 = boxes[idx].astype(int)
        face = pil.crop((x1, y1, x2, y2))

        return face, annotated, {"boxes": boxes, "scores": probs}