import os
from io import BytesIO
import gradio as gr
import google.generativeai as genai
from PIL import Image
from dotenv import load_dotenv

# Load environment variables
load_dotenv()
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
input_prompt="""
 Analyze the image and predict type of soil "
"1. Soil Type: Sandy/Clayey/Loamy,"
"2. ",
"""


# Function to get a response from the Google Gemini Vision API
def get_gemini_response(image):
    model = genai.GenerativeModel('gemini-1.5-pro')

    # Convert PIL image to bytes
    bytes_io = BytesIO()
    image.save(bytes_io, format='PNG')
    bytes_data = bytes_io.getvalue()
    response=model.generate_content([input_prompt,image])
    # Placeholder logic for now: replace with actual API usage
    # You would pass the image bytes data to the API
    # Currently this is a mock response
    crop_name = "Wheat"  # Example: Replace this with the actual API result
    disease_type = "Wheat Septoria"  # Example: Replace this with the actual API result

    # Return disease info
    #return get_disease_info(crop_name, disease_type)
    return response.text

# Function to handle the uploaded image and predict crop health
def predict_crop_health(uploaded_image):
    # Pass the image to the Gemini API to get prediction
    return get_gemini_response(uploaded_image)

# Define the Gradio interface: Inputs and Outputs
inputs = gr.Image(type="pil", label="Upload Crop Image")
outputs = gr.Markdown(label="Prediction Results")

# Launch the Gradio interface
gr.Interface(
    fn=predict_crop_health,
    inputs=inputs,
    outputs=gr.Textbox(label="Crop Disease Predictor"),
    title="Soil Type Classifier",
    description="Upload The Image Of Soil",
    live=False
).launch()