app.py

import gradio as gr
import pandas as pd
import matplotlib.pyplot as plt
import io
import google.generativeai as genai
from PIL import Image
import ast

def process_file(api_key, file, instructions):
    # Configure Gemini API with correct model
    genai.configure(api_key=api_key)
    model = genai.GenerativeModel('gemini-2.5-pro-preview-03-25')

    try:
        # Read uploaded file
        if file.name.endswith('.csv'):
            df = pd.read_csv(file.name)
        else:
            df = pd.read_excel(file.name)
    except Exception as e:
        print(f"File Error: {str(e)}")
        return [None, None, None]

    # Enhanced prompt with strict coding rules
    prompt = f"""Generate 3 distinct matplotlib visualization codes with these rules:
    1. Use ONLY these variables: df (existing DataFrame), plt
    2. No imports or additional data loading
    3. Each visualization must:
       - Start with: plt.figure(figsize=(16,9), dpi=120)
       - Use plt.style.use('ggplot')
       - Include title, axis labels, and grid
       - End with plt.tight_layout()
    4. Different chart types (bar, line, scatter, etc)
    
    Dataset columns: {list(df.columns)}
    Sample data: {df.head(3).to_dict()}
    User instructions: {instructions or 'None'}
    
    Format EXACTLY as:
    # Visualization 1
    plt.figure(figsize=(16,9), dpi=120)
    plt.style.use('ggplot')
    # Visualization code using df
    plt.tight_layout()
    """

    try:
        response = model.generate_content(prompt)
        code_blocks = response.text.split("# Visualization ")[1:4]
    except Exception as e:
        print(f"Gemini Error: {str(e)}")
        return [None, None, None]

    visualizations = []
    for i, block in enumerate(code_blocks, 1):
        buf = io.BytesIO()
        try:
            # Clean and validate generated code
            cleaned_code = '\n'.join([
                line.replace('data', 'df').split('#')[0].strip()
                for line in block.split('\n')[1:]
                if line.strip() and
                not any(s in line.lower() for s in ['import', 'data=', 'data ='])
            ])
            
            # Syntax check
            ast.parse(cleaned_code)
            
            # Execute code in controlled environment
            exec_env = {'df': df, 'plt': plt}
            plt.figure(figsize=(16, 9), dpi=120)
            exec(cleaned_code, exec_env)
            
            # Save HD image
            plt.savefig(buf, format='png', bbox_inches='tight')
            plt.close()
            buf.seek(0)
            visualizations.append(Image.open(buf))
        except Exception as e:
            print(f"Visualization {i} Error: {str(e)}")
            print(f"Problematic Code:\n{cleaned_code}")
            visualizations.append(None)

    # Ensure exactly 3 outputs
    return visualizations + [None]*(3-len(visualizations))

# Gradio interface
with gr.Blocks(theme=gr.themes.Default(spacing_size="lg")) as demo:
    gr.Markdown("# 🔍 Data Visualization Generator")
    
    with gr.Row():
        api_key = gr.Textbox(
            label="🔑 Gemini API Key",
            type="password",
            placeholder="Enter your API key here"
        )
        file = gr.File(
            label="📁 Upload Dataset",
            file_types=[".csv", ".xlsx"],
            type="filepath"
        )
    
    instructions = gr.Textbox(
        label="💡 Custom Instructions",
        placeholder="E.g.: Compare sales trends across regions..."
    )
    
    submit = gr.Button("🚀 Generate Visualizations", variant="primary")
    
    with gr.Row():
        outputs = [
            gr.Image(
                label=f"Visualization {i+1}",
                width=600,
                height=400
            ) for i in range(3)
        ]

    submit.click(
        process_file,
        inputs=[api_key, file, instructions],
        outputs=outputs
    )

if __name__ == "__main__":
    demo.launch()