import gradio as gr
import os
import json
import uuid
from datetime import datetime
from pathlib import Path
import requests
from functools import lru_cache

# Minimal HF Space implementation
DATA_DIR = Path("./data")
PROJECTS_FILE = DATA_DIR / "projects.json"

DATA_DIR.mkdir(exist_ok=True)
if not PROJECTS_FILE.exists():
    with open(PROJECTS_FILE, 'w') as f:
        json.dump({}, f)

def load_projects():
    try:
        with open(PROJECTS_FILE, 'r') as f:
            return json.load(f)
    except:
        return {}

def save_projects(projects):
    with open(PROJECTS_FILE, 'w') as f:
        json.dump(projects, f, indent=2)

@lru_cache(maxsize=32)
def call_hf_api(prompt: str):
    """Call Hugging Face Inference API with caching."""
    try:
        hf_token = os.getenv('HF_TOKEN')
        if not hf_token:
            return None
            
        headers = {"Authorization": f"Bearer {hf_token}"}
        response = requests.post(
            "https://api-inference.huggingface.co/models/microsoft/DialoGPT-medium",
            headers=headers,
            json={"inputs": prompt, "parameters": {"max_length": 300}},
            timeout=10
        )
        
        if response.status_code == 200:
            result = response.json()
            if isinstance(result, list) and len(result) > 0:
                return result[0].get('generated_text', '').replace(prompt, '').strip()
    except Exception as e:
        print(f"HF API error: {e}")
    return None

def generate_static_analysis(project_name: str, description: str, features: str):
    """Fallback static analysis template."""
    return f"""# Technical Analysis for {project_name}

## Project Overview
{description}

## Recommended Technology Stack
- **Backend**: Python with FastAPI
- **Frontend**: React with TypeScript  
- **Database**: PostgreSQL
- **Testing**: Jest, Pytest
- **Deployment**: Hugging Face Spaces

## Key Features Implementation
{features}

*Analysis powered by AgentAI on Hugging Face Spaces*
"""

def analyze_requirements(project_name: str, description: str, features: str):
    if not project_name or not description:
        return "Please provide project name and description.", "", ""
    
    # Try LLM analysis first
    prompt = f"Analyze this software project: {project_name}. Description: {description}. Features: {features}. Provide technical recommendations:"
    llm_analysis = call_hf_api(prompt)
    
    if llm_analysis and len(llm_analysis) > 50:
        analysis = f"""# AI-Generated Technical Analysis for {project_name}

## Project Overview
{description}

## AI Analysis
{llm_analysis}

## Key Features
{features}

*Analysis generated using Hugging Face LLM*
"""
        status_msg = f"✅ Project '{project_name}' analyzed with AI!"
    else:
        # Fallback to static template
        analysis = generate_static_analysis(project_name, description, features)
        status_msg = f"✅ Project '{project_name}' created (static template)!"
    
    project_id = str(uuid.uuid4())
    projects = load_projects()
    projects[project_id] = {
        "name": project_name,
        "description": description,
        "features": features,
        "analysis": analysis,
        "created_at": datetime.now().isoformat(),
        "ai_generated": llm_analysis is not None
    }
    save_projects(projects)
    
    return analysis, status_msg, project_id

def generate_code(project_id: str):
    projects = load_projects()
    if not project_id or project_id not in projects:
        return "Please create a project first."
    
    project = projects[project_id]
    
    # Try LLM code generation
    prompt = f"Generate Python FastAPI code for {project['name']}: {project['description']}. Include basic endpoints:"
    llm_code = call_hf_api(prompt)
    
    if llm_code and "def " in llm_code:
        code = f"""# {project["name"]} - AI Generated Code

{llm_code}

# Generated by AgentAI with Hugging Face LLM
"""
    else:
        # Fallback to static template
        code = f"""# {project["name"]} - Generated by AgentAI

from fastapi import FastAPI
from pydantic import BaseModel

app = FastAPI(title="{project["name"]}")

@app.get("/")
async def root():
    return {{"message": "Welcome to {project["name"]} API"}}

@app.get("/health")
async def health():
    return {{"status": "healthy"}}

# Generated by AgentAI - Hugging Face Spaces
"""
    
    return code

with gr.Blocks(title="AgentAI - HF Space", theme=gr.themes.Soft()) as demo:
    gr.Markdown("# 🤖 AgentAI - Hugging Face Space")
    
    with gr.Tab("🚀 Create Project"):
        with gr.Row():
            with gr.Column():
                project_name = gr.Textbox(label="Project Name")
                description = gr.Textbox(label="Description", lines=3)
                features = gr.Textbox(label="Features", lines=3)
                create_btn = gr.Button("Analyze", variant="primary")
            with gr.Column():
                analysis_output = gr.Markdown()
                status_output = gr.Textbox(label="Status")
                project_id_output = gr.Textbox(label="Project ID")
    
    with gr.Tab("💻 Generate Code"):
        with gr.Row():
            with gr.Column():
                input_project_id = gr.Textbox(label="Project ID")
                generate_btn = gr.Button("Generate Code", variant="primary")
            with gr.Column():
                code_output = gr.Code(language="python")
    
    create_btn.click(
        fn=analyze_requirements,
        inputs=[project_name, description, features],
        outputs=[analysis_output, status_output, project_id_output]
    )
    
    generate_btn.click(
        fn=generate_code,
        inputs=[input_project_id],
        outputs=[code_output]
    )

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