app.py

import gradio as gr
import pandas as pd
from io import StringIO
from smolagents import CodeAgent, HfApiModel

# Initialize the AI agent
agent = CodeAgent(
    tools=[],
    model=HfApiModel(model_id="Qwen/Qwen2.5-Coder-32B-Instruct"),
)

def process_text(content):
    """Handle text processing without database dependency"""
    # Get CSV conversion from AI
    csv_output = agent.run(f"Convert to CSV:\n{content}\nReturn ONLY valid CSV:")
    
    # Process CSV data
    try:
        df = pd.read_csv(StringIO(csv_output), keep_default_na=False)
        return df.head(10), csv_output
    except Exception as e:
        return pd.DataFrame(), f"Error processing data: {str(e)}"

def analyze_content(full_text):
    """Analyze text content for reporting"""
    analysis_prompt = f"""
    Analyze this text and generate a structured report:
    {full_text[:5000]}
    
    Include:
    1. Key themes/topics
    2. Important entities
    3. Summary statistics
    4. Recommendations/insights
    
    Use markdown formatting with headers.
    """
    return agent.run(analysis_prompt)

def handle_upload(*files):
    """Process uploaded files"""
    all_dfs = []
    full_text = ""
    
    for file in files:
        content = file.read().decode()
        df, _ = process_text(content)
        all_dfs.append(df)
        full_text += f"\n\n--- {file.name} ---\n{content}"
    
    combined_df = pd.concat(all_dfs, ignore_index=True) if all_dfs else pd.DataFrame()
    report = analyze_content(full_text) if full_text else "No content to analyze"
    
    return combined_df, report

with gr.Blocks() as demo:
    gr.Markdown("# Document Analysis System")
    
    with gr.Row():
        file_input = gr.File(file_count="multiple", file_types=[".txt"])
        upload_btn = gr.Button("Process Files", variant="primary")
    
    with gr.Row():
        data_output = gr.Dataframe(label="Structured Data Preview")
        report_output = gr.Markdown(label="Analysis Report")

    upload_btn.click(
        handle_upload,
        inputs=file_input,
        outputs=[data_output, report_output]
    )

if __name__ == "__main__":
    demo.launch(
        server_name="0.0.0.0",
        server_port=7860,
        show_error=True
    )