import streamlit as st
from PIL import Image, ImageOps
import numpy as np
import io
import time
import math

# Hugging Face Transformers specific imports
from transformers import AutoModelForSequenceClassification, AutoTokenizer

# Load pre-trained model and tokenizer
model_name = "distilbert-base-uncased-finetuned-sst-2-english"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

def rotate_image(image, angle):
    """
    Rotate the given PIL image by the given angle
    """
    # Add padding to prevent cropping
    image = ImageOps.expand(image, border=int(np.ceil(max(image.size) * 1.5)), fill="white")
    rotated_image = image.rotate(angle, resample=Image.BICUBIC, expand=True)
    return rotated_image

def main():
    st.title("Hugging Face Streamlit Example: Rotating Image")

    # Upload an image
    st.sidebar.title("Upload Image")
    uploaded_file = st.sidebar.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])

    # Rotation speed adjustment
    rotation_speed = st.sidebar.slider("Rotation Speed", min_value=0, max_value=360, value=10, step=1)

    if uploaded_file is not None:
        # Display the image
        image = Image.open(uploaded_file)
        st.image(image, caption="Uploaded Image", use_column_width=True)

        # Rotation angle initialization
        rotation_angle = 0

        # Display an empty placeholder
        rotating_image_placeholder = st.empty()

        while True:
            # Calculate the new rotation angle
            rotation_angle += rotation_speed
            rotation_angle %= 360  # Keep angle within 0-359 range

            # Rotate the image
            rotated_image = rotate_image(image, rotation_angle)

            # Display the rotated image
            rotating_image_placeholder.image(rotated_image, caption="Rotated Image", use_column_width=True)

            # Pause briefly to control the rotation speed
            time.sleep(0.1)

if __name__ == "__main__":
    main()