import os
os.system('pip install paddlepaddle==2.4.2')
# os.system('pip install paddlepaddle==0.0.0 -f https://www.paddlepaddle.org.cn/whl/linux/cpu-mkl/develop.html')
os.system('pip install paddleocr')
from paddleocr import PaddleOCR, draw_ocr
from PIL import Image
from typing import  Dict, List, Any
import base64
from io import BytesIO
import numpy as np

class EndpointHandler():
    def __init__(self, path=""):
        
        self.pipeline = PaddleOCR(lang="en",ocr_version="PP-OCRv4",
                                    show_log = False,use_gpu=False,
                                    det_model_dir=path,
                                    cls_model_dir=path,
                                    rec_model_dir=path
                                    )


    def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
        """
        Args:
            data (:obj:):
                includes the input data and the parameters for the inference.
        Return:
            A :obj:`list`:. The object returned should be a list of one list like [[{"label": 0.9939950108528137}]] containing :
                - "label": A string representing what the label/class is. There can be multiple labels.
                - "score": A score between 0 and 1 describing how confident the model is for this label/class.
        """
        
        inputs = data.pop("inputs", data)
        #parameters = data.pop("parameters", None)
        
        receipt_image = Image.open(BytesIO(base64.b64decode(inputs)))
        receipt_image_array = np.array(receipt_image.convert('RGB'))
        result = self.pipeline.ocr(receipt_image_array,cls=True)
        txts = [line[1][0] for line in result[0]]
        # pass inputs with all kwargs in data
        extract = "".join(txts)
        # postprocess the prediction
        return extract