How can i run this cloud function in python.

[finite rampart]

i'm kinda new in python and i wanna know why i can't run this cloud function in my self-hosted appwrite cloud function.
this is my code for image processing, it works on locale runs but i don't know how to make it work on cloud.

import cv2
import dlib
import numpy as np
import pytesseract
from PIL import Image


# Load the detector
detector = dlib.get_frontal_face_detector()

def extract_face(image_path, output_path, padding=35):
    # Load the image
    img = cv2.imread(image_path)
    # Convert to grayscale
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    # Use the detector to find faces
    faces = detector(gray)
    for i, face in enumerate(faces):
        # Extract the face with some padding to make the image larger
        x1, y1, x2, y2 = face.left() - padding, face.top() - padding, face.right() + padding, face.bottom() + padding
        # Ensure the coordinates are within the image boundaries
        x1, y1 = max(0, x1), max(0, y1)
        x2, y2 = min(img.shape[1]-1, x2), min(img.shape[0]-1, y2)
        # Crop the face out of the image
        crop = img[y1:y2, x1:x2]
        return crop

In the console in the appwrite self-host it logs
Docker Error: tar: short read

also part 2 of the code:

def passportProcessing(image_path):
   # Reading the image
    img = cv2.imread(image_path)

    # Convert to gray scale
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    # Adaptive thresholding
    adaptive_thresh = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
                                            cv2.THRESH_BINARY, 11, 2)

    # Morphological operations to enhance text
    kernel = np.ones((1, 1), np.uint8)
    img_morph = cv2.morphologyEx(adaptive_thresh, cv2.MORPH_OPEN, kernel)
    img_morph = cv2.morphologyEx(img_morph, cv2.MORPH_CLOSE, kernel)

    if img is not None:
        text = pytesseract.image_to_string(img_morph,lang='eng+ara')
        texts = text.split('\n')
        # Applying the cleaning function to each text and filtering out single character or empty entries
        filtered_texts_corrected = [clean_text(text) for text in texts if text.strip() and len(text.strip()) > 1]

        return filtered_texts_corrected
    else:
        print("Image not loaded correctly. Check the file path and file format: {image_path}")
        
def clean_text(text):
    return ''.join(char for char in text if char not in ('\u200f', '\u200e'))

def main(req, res):
  client = Client()
  
  print('Hello from Appwrite Function!')
  print("Variables",req.variables)

  # You can remove services you don't use
  database = Databases(client)
  users = Users(client)

  if not req.variables.get('APPWRITE_FUNCTION_ENDPOINT') or not req.variables.get('APPWRITE_FUNCTION_API_KEY'):
    print('Environment variables are not set. Function cannot use Appwrite SDK.')
  else:
    (
    client
      .set_endpoint(req.variables.get('APPWRITE_FUNCTION_ENDPOINT', None))
      .set_project(req.variables.get('APPWRITE_FUNCTION_PROJECT_ID', None))
      .set_key(req.variables.get('APPWRITE_FUNCTION_API_KEY', None))
      .set_self_signed(True)
    )
  
  return res.json({
    "areDevelopersAwesome": True,
  })

[lament bridge]

This error is showing up when deploying or executing?

[finite rampart]

when depoloying

[lament bridge]

How are you deploying? And have you deployed any Function from this device before?

What version of Appwrite?

[finite rampart]

i deployed cloud function by node runtimes not with python, my version of the appwrite is 1.4.13.

[lament bridge]

And how are you deploying this Function?

[finite rampart]

i just grabbed it from github repository to deploy it.

[lament bridge]

Right, but how did you get the code into Appwrite? With the CLI? Direct upload to the Console? Connect to Github?

[finite rampart]

From connect to Github.

[lament bridge]

Does the Function name have any spaces in it?

[finite rampart]

no i build it like every nodejs function i built before so no unnecessary spaces.

[lament bridge]

It looks like this error might be triggered by setting an incorrect root path sometimes?

[finite rampart]

{
            "name": "ImageProcessing",
            "runtime": "python-3.11",
            "path": "functions/ImageProcessing",
            "entrypoint": "src/index.py",
            "ignore": [
                "__pypackages__"
            ],
            "execute": [],
            "events": [],
            "schedule": "",
            "timeout": 15
        }

Weird Cause my Root Path is Correct. but question how will it read those packages i'll use with it like pytesseract and cmake?
cause those packages mostly have to be in users personal computer so it could access it.

[lament bridge]

Any packages that can be installed with pip should be added to your requirements.txt, which will automatically be installed during deployment.

For OS packages, the best method would be to include the binaries in your Function. There's been dicussion about allowing you to set your own Docker container, but that's unlikely to happen any time soon