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):
    """Convert text content to structured format"""
    csv_output = agent.run(f"Convert to CSV (include headers):\n{content}\nOutput ONLY valid CSV:")
    try:
        return pd.read_csv(StringIO(csv_output), keep_default_na=False)
    except:
        return pd.DataFrame()

def analyze_content(full_text):
    """Generate comprehensive report"""
    report = agent.run(f"""
    Create detailed analysis report from this data:
    {full_text[:5000]}
    
    Include:
    1. Key insights and patterns
    2. Important statistics
    3. Actionable recommendations
    4. Potential anomalies
    
    Use markdown formatting with headers.
    """)
    return report

def handle_upload(files):
    """Handle multiple file uploads correctly"""
    all_dfs = []
    full_content = []
    
    for file in files:
        try:
            content = file.name + "\n" + file.read().decode()
            df = process_text(content)
            all_dfs.append(df)
            full_content.append(content)
        except Exception as e:
            print(f"Error processing {file.name}: {str(e)}")
    
    combined_df = pd.concat(all_dfs, ignore_index=True) if all_dfs else pd.DataFrame()
    report = analyze_content("\n\n".join(full_content)) if full_content else "No valid content found"
    
    return combined_df, report

with gr.Blocks() as demo:
    gr.Markdown("# Multi-Document Analyzer")
    
    with gr.Row():
        file_input = gr.File(
            file_count="multiple",
            file_types=[".txt"],
            label="Upload Text Files"
        )
        process_btn = gr.Button("Analyze Documents", variant="primary")
    
    with gr.Row():
        data_output = gr.Dataframe(label="Structured Data Preview", wrap=True)
        report_output = gr.Markdown(label="Analysis Report")

    process_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,
        share=True  # Enable public sharing
    )