import gradio as gr
import pytesseract
from PIL import Image

# Function to extract text from the uploaded image
def extract_text(image):
    text = pytesseract.image_to_string(image, lang='eng+hin')  # Extract text using Tesseract
    return text

# Function to extract text and perform keyword search
def ocr_and_search(image, keyword):
    extracted_text = extract_text(image)  # Get the extracted text
    if keyword:
        # Search for the keyword in the extracted text
        matching_lines = [line for line in extracted_text.splitlines() if keyword.lower() in line.lower()]
        return extracted_text, matching_lines  # Return both extracted text and matching lines
    else:
        return extracted_text, []  # If no keyword, return empty list for matches

# Create a Gradio interface
interface = gr.Interface(
    fn=ocr_and_search,
    inputs=[
        gr.Image(type="pil", label="Upload Image"),  # Input for image upload
        gr.Textbox(label="Enter Keyword")  # Input for keyword search
    ],
    outputs=[
        gr.Textbox(label="Extracted Text"),  # Output for extracted text
        gr.Textbox(label="Matching Lines")  # Output for matching lines based on keyword search
    ],
    title="OCR with Keyword Search",
    description="Upload an image and enter a keyword to search within the extracted text."
)

# Launch the app with share=True to create a public link
if __name__ == "__main__":
    interface.launch(share=True)  # Set share=True to generate a public URL