Code Editor for app.py

import streamlit as st
import google.generativeai as genai
from PIL import Image
import json
import os

# Configure the page
st.set_page_config(page_title="CNPJ Extractor", layout="wide")

# Initialize Gemini API
# Note: In a production environment, use st.secrets or environment variables
GOOGLE_API_KEY = st.text_input("Enter your Google API Key", type="password")
if GOOGLE_API_KEY:
    genai.configure(api_key=GOOGLE_API_KEY)

def extract_cnpj(image):
    """Use Gemini to extract CNPJ from image"""
    model = genai.GenerativeModel('gemini-pro-vision')
    
    prompt = """
    Please analyze this image and extract any CNPJ numbers you find.
    A CNPJ is a Brazilian company registration number in the format XX.XXX.XXX/XXXX-XX.
    Return the result as a JSON string with the following structure:
    {
        "cnpj": "XX.XXX.XXX/XXXX-XX",
        "confidence": "high/medium/low"
    }
    If no CNPJ is found, return {"cnpj": null, "confidence": null}
    """
    
    response = model.generate_content([prompt, image])
    
    try:
        # Extract JSON from response
        response_text = response.text
        # Find JSON-like string in response
        start = response_text.find('{')
        end = response_text.rfind('}') + 1
        json_str = response_text[start:end]
        result = json.loads(json_str)
        return result
    except Exception as e:
        return {"cnpj": None, "confidence": None, "error": str(e)}

def main():
    st.title("🔍 CNPJ Extractor from Images")
    
    st.write("""
    Upload an image containing a CNPJ (Brazilian company registration number) and 
    this app will attempt to extract it using Google's Gemini AI.
    """)
    
    # File uploader
    uploaded_file = st.file_uploader(
        "Choose an image file", 
        type=['png', 'jpg', 'jpeg']
    )
    
    if uploaded_file and GOOGLE_API_KEY:
        # Display the uploaded image
        image = Image.open(uploaded_file)
        st.image(image, caption="Uploaded Image", use_column_width=True)
        
        with st.spinner("Analyzing image..."):
            try:
                result = extract_cnpj(image)
                
                # Display results in a nice format
                st.subheader("Results")
                if result.get("cnpj"):
                    st.success(f"CNPJ found: {result['cnpj']}")
                    st.info(f"Confidence: {result['confidence']}")
                else:
                    st.warning("No CNPJ found in the image")
                
                # Display raw JSON
                st.subheader("Raw JSON Output")
                st.json(result)
                
            except Exception as e:
                st.error(f"An error occurred: {str(e)}")
    elif not GOOGLE_API_KEY:
        st.warning("Please enter your Google API Key to proceed")

if __name__ == "__main__":
    main()