app.py

import gradio as gr
import json
import time
import os
from datetime import datetime
from llm_utils import LLM_PROVIDERS, get_llm_models, validate_api_key, call_llm, parse_llm_output, create_zip_from_files
from project_types import PROJECT_TYPES
from advanced_prompts import (
    get_advanced_context_prompt,
    get_advanced_commands_prompt,
    get_advanced_prompts_prompt,
    calculate_project_complexity,
    get_industry_specific_requirements
)

# Global state
generated_files = {'context': {}, 'commands': {}, 'prompts': {}}
stats = {'context': {'tokens': 0, 'time': 0}, 'commands': {'tokens': 0, 'time': 0}, 'prompts': {'tokens': 0, 'time': 0}}
project_analysis = {}

# Sektörler
INDUSTRIES = [
    "Teknoloji/Yazılım", "Finans/Fintech", "Sağlık/Healthcare", 
    "E-ticaret", "Eğitim/EdTech", "Oyun/Gaming", "Sosyal Medya",
    "IoT/Akıllı Cihazlar", "Blockchain/Web3", "AI/ML Servisleri",
    "SaaS B2B", "SaaS B2C", "Marketplace", "Diğer"
]

# Proje özellikleri
PROJECT_FEATURES = [
    "Kullanıcı Yönetimi", "Ödeme İşlemleri", "Gerçek Zamanlı Özellikler",
    "AI/ML Entegrasyonu", "Çoklu Platform", "3. Parti Entegrasyonlar",
    "Offline Çalışma", "Çoklu Dil Desteği", "Bildirim Sistemi",
    "Analytics/Raporlama", "Sosyal Özellikler", "Güvenlik/Compliance"
]

def analyze_project(project_type, project_idea, industry, features):
    """Proje analizi yapar ve öneriler üretir"""
    analysis = {
        "complexity": calculate_project_complexity(project_type, project_idea, features),
        "industry_requirements": get_industry_specific_requirements(industry),
        "estimated_files": {},
        "tech_recommendations": [],
        "architecture_suggestion": ""
    }
    
    # Teknoloji önerileri
    if "AI/ML" in project_type:
        analysis["tech_recommendations"] = ["Python", "TensorFlow/PyTorch", "MLflow", "Docker", "Kubernetes"]
        analysis["architecture_suggestion"] = "Microservices with ML Pipeline"
    elif "Web" in project_type:
        analysis["tech_recommendations"] = ["React/Next.js", "Node.js/Python", "PostgreSQL", "Redis", "Docker"]
        analysis["architecture_suggestion"] = "Modern Jamstack Architecture"
    elif "Mobil" in project_type:
        analysis["tech_recommendations"] = ["React Native/Flutter", "Firebase", "GraphQL", "Redux/MobX"]
        analysis["architecture_suggestion"] = "Clean Architecture with BLoC/MVVM"
    
    return analysis

def generate_section_advanced(provider, model, api_key, project_type, project_idea, 

                            section, industry, features, tech_details, progress=gr.Progress()):
    """Gelişmiş section oluşturma"""
    try:
        start_time = time.time()
        
        # Proje analizi
        progress(0.05, desc="Proje analiz ediliyor...")
        analysis = analyze_project(project_type, project_idea, industry, features)
        global project_analysis
        project_analysis = analysis
        
        # Progress
        progress(0.1, desc=f"{section.capitalize()} için akıllı prompt hazırlanıyor...")
        
        # Section'a göre advanced prompt seç
        if section == 'context':
            prompt = get_advanced_context_prompt(project_type, project_idea, tech_details)
        elif section == 'commands':
            context_summary = f"Proje {len(generated_files['context'])} context dosyası içeriyor"
            prompt = get_advanced_commands_prompt(project_type, project_idea, context_summary)
        elif section == 'prompts':
            prompt = get_advanced_prompts_prompt(project_type, project_idea, tech_details)
        
        # Sektör gereksinimlerini ekle
        if industry != "Diğer":
            prompt += f"\n\nSEKTÖR GEREKSİNİMLERİ ({industry}):\n"
            prompt += "\n".join(f"- {req}" for req in analysis["industry_requirements"])
        
        # Token tahmini
        prompt_tokens = len(prompt.split()) * 1.3
        
        progress(0.3, desc="AI ile iletişim kuruluyor...")
        
        # LLM çağrısı
        llm_response = call_llm(provider, model, api_key, prompt)
        
        if not llm_response or "hata" in llm_response.lower():
            return f"<p style='color:red;'>❌ Hata: {llm_response}</p>", None, None
        
        progress(0.6, desc="Yanıt işleniyor ve optimize ediliyor...")
        
        # Response analizi
        response_tokens = len(llm_response.split()) * 1.3
        total_tokens = prompt_tokens + response_tokens
        
        # Dosyaları parse et
        parsed_files = parse_llm_output(llm_response)
        
        if not parsed_files:
            return f"<p style='color:red;'>❌ Dosyalar ayrıştırılamadı.</p>", None, None
        
        # Kaydet
        generated_files[section] = parsed_files
        
        # İstatistikleri güncelle
        elapsed_time = time.time() - start_time
        stats[section] = {'tokens': int(total_tokens), 'time': round(elapsed_time, 2)}
        
        progress(0.9, desc="Kalite kontrolü yapılıyor...")
        
        # Dosya listesi ve kalite skoru
        file_count = len(parsed_files)
        quality_score = min(100, 60 + (file_count * 4))  # Basit kalite skoru
        
        # Başarı mesajı
        file_list_html = "<div style='max-height: 200px; overflow-y: auto;'>"
        for i, (file, content) in enumerate(parsed_files.items()):
            size_kb = len(content.encode('utf-8')) / 1024
            file_list_html += f"""

            <div style='padding: 5px; border-bottom: 1px solid #e5e7eb;'>

                <span style='color: #10b981;'>✅</span> 

                <strong>{file}</strong> 

                <span style='color: #6b7280; font-size: 0.9em;'>({size_kb:.1f} KB)</span>

            </div>

            """
        file_list_html += "</div>"
        
        success_msg = f"""

<div style='background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); padding: 20px; border-radius: 12px; color: white;'>

    <h3 style='margin-top: 0;'>✨ {section.capitalize()} Başarıyla Oluşturuldu!</h3>

    <div style='display: grid; grid-template-columns: repeat(3, 1fr); gap: 15px; margin: 15px 0;'>

        <div style='background: rgba(255,255,255,0.2); padding: 10px; border-radius: 8px;'>

            <div style='font-size: 24px; font-weight: bold;'>{file_count}</div>

            <div style='font-size: 12px; opacity: 0.9;'>Dosya Oluşturuldu</div>

        </div>

        <div style='background: rgba(255,255,255,0.2); padding: 10px; border-radius: 8px;'>

            <div style='font-size: 24px; font-weight: bold;'>~{int(total_tokens)}</div>

            <div style='font-size: 12px; opacity: 0.9;'>Token Kullanıldı</div>

        </div>

        <div style='background: rgba(255,255,255,0.2); padding: 10px; border-radius: 8px;'>

            <div style='font-size: 24px; font-weight: bold;'>{elapsed_time:.1f}s</div>

            <div style='font-size: 12px; opacity: 0.9;'>İşlem Süresi</div>

        </div>

    </div>

    <div style='margin-top: 15px;'>

        <div style='font-weight: bold; margin-bottom: 10px;'>📁 Oluşturulan Dosyalar:</div>

        {file_list_html}

    </div>

    <div style='margin-top: 15px; padding: 10px; background: rgba(255,255,255,0.1); border-radius: 8px;'>

        <span style='font-weight: bold;'>Kalite Skoru:</span> 

        <span style='font-size: 20px;'>{quality_score}/100</span>

    </div>

</div>

"""
        
        # Analiz raporu
        analysis_report = f"""

<div style='margin-top: 20px; padding: 15px; background: #f3f4f6; border-radius: 8px;'>

    <h4 style='margin-top: 0;'>📊 Proje Analizi</h4>

    <ul style='margin: 10px 0;'>

        <li><strong>Karmaşıklık:</strong> {analysis['complexity']['context'] + analysis['complexity']['commands'] + analysis['complexity']['prompts']} puan</li>

        <li><strong>Sektör:</strong> {industry}</li>

        <li><strong>Önerilen Teknolojiler:</strong> {', '.join(analysis['tech_recommendations'][:3])}</li>

        <li><strong>Mimari:</strong> {analysis['architecture_suggestion']}</li>

    </ul>

</div>

"""
        
        return success_msg + analysis_report, None, analysis
        
    except Exception as e:
        import traceback
        error_detail = traceback.format_exc()
        return f"<p style='color:red;'>❌ Hata: {str(e)}</p><pre>{error_detail}</pre>", None, None

def download_all_professional():
    """Profesyonel ZIP paketi oluşturur"""
    all_files = {}
    
    # Dosyaları topla
    for section in ['context', 'commands', 'prompts']:
        if generated_files[section]:
            all_files.update(generated_files[section])
    
    if not all_files:
        return None, "<p style='color:red;'>❌ İndirilecek dosya yok.</p>"
    
    # README.md oluştur
    readme_content = f"""# {project_analysis.get('project_name', 'Project')} - Context Engineering



## 📋 Proje Özeti

- **Tür:** {project_analysis.get('project_type', 'N/A')}

- **Sektör:** {project_analysis.get('industry', 'N/A')}

- **Karmaşıklık:** {project_analysis.get('complexity', {}).get('context', 0) + project_analysis.get('complexity', {}).get('commands', 0) + project_analysis.get('complexity', {}).get('prompts', 0)} puan



## 📁 Dosya Yapısı

- `context/` - Proje bağlamı ve dokümantasyon ({len(generated_files['context'])} dosya)

- `commands/` - Otomasyon ve yönetim komutları ({len(generated_files['commands'])} dosya)  

- `prompts/` - AI asistan prompt şablonları ({len(generated_files['prompts'])} dosya)



## 🚀 Başlangıç

1. Context dosyalarını okuyarak projeyi anlayın

2. Commands klasöründeki setup komutlarını çalıştırın

3. Prompts şablonlarını AI asistanınızla kullanın



## 📊 İstatistikler

- Toplam Dosya: {len(all_files)}

- Toplam Token: ~{sum(s['tokens'] for s in stats.values())}

- Oluşturma Süresi: {sum(s['time'] for s in stats.values()):.1f} saniye



---

Generated by Context Engineer v2.0

"""
    
    all_files['README.md'] = readme_content
    
    # Manifest dosyası
    manifest = {
        "version": "2.0",
        "generated_at": datetime.now().isoformat(),
        "statistics": stats,
        "project_analysis": project_analysis,
        "file_count": {
            "context": len(generated_files['context']),
            "commands": len(generated_files['commands']),
            "prompts": len(generated_files['prompts'])
        }
    }
    all_files['manifest.json'] = json.dumps(manifest, indent=2)
    
    # ZIP oluştur
    zip_data = create_zip_from_files(all_files)
    timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
    zip_file_path = f"./context_engineering_pro_{timestamp}.zip"
    
    with open(zip_file_path, "wb") as f:
        f.write(zip_data)
    
    # Özet rapor
    total_size_kb = sum(len(content.encode('utf-8')) for content in all_files.values()) / 1024
    
    summary = f"""

<div style='background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); padding: 30px; border-radius: 16px; color: white;'>

    <h2 style='margin-top: 0;'>🎉 Proje Paketi Hazır!</h2>

    

    <div style='display: grid; grid-template-columns: repeat(2, 1fr); gap: 20px; margin: 20px 0;'>

        <div style='background: rgba(255,255,255,0.2); padding: 20px; border-radius: 12px;'>

            <h3 style='margin-top: 0;'>📊 Genel İstatistikler</h3>

            <ul style='list-style: none; padding: 0;'>

                <li>📁 <strong>Toplam Dosya:</strong> {len(all_files)}</li>

                <li>💾 <strong>Paket Boyutu:</strong> {total_size_kb:.1f} KB</li>

                <li>🎯 <strong>Token Kullanımı:</strong> ~{sum(s['tokens'] for s in stats.values())}</li>

                <li>⏱️ <strong>Toplam Süre:</strong> {sum(s['time'] for s in stats.values()):.1f}s</li>

            </ul>

        </div>

        

        <div style='background: rgba(255,255,255,0.2); padding: 20px; border-radius: 12px;'>

            <h3 style='margin-top: 0;'>📂 Klasör Detayları</h3>

            <ul style='list-style: none; padding: 0;'>

                <li>📘 <strong>Context:</strong> {len(generated_files['context'])} dosya</li>

                <li>⚡ <strong>Commands:</strong> {len(generated_files['commands'])} dosya</li>

                <li>💡 <strong>Prompts:</strong> {len(generated_files['prompts'])} dosya</li>

                <li>📄 <strong>Ekstra:</strong> README.md, manifest.json</li>

            </ul>

        </div>

    </div>

    

    <div style='background: rgba(255,255,255,0.1); padding: 15px; border-radius: 8px; margin-top: 20px;'>

        <p style='margin: 0; text-align: center; font-size: 18px;'>

            ✨ Profesyonel context engineering paketiniz başarıyla oluşturuldu!

        </p>

    </div>

</div>

"""
    
    return gr.File(value=zip_file_path, visible=True), summary

# Ana arayüz
with gr.Blocks(theme=gr.themes.Soft(), css="""

    .gradio-container {

        font-family: 'Inter', sans-serif;

    }

    .gr-button-primary {

        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);

        border: none;

    }

    .gr-button-secondary {

        background: linear-gradient(135deg, #f093fb 0%, #f5576c 100%);

        border: none;

    }

""") as demo:
    
    gr.Markdown("""

    # 🚀 Context Engineer Pro v2.0

    ### Enterprise-Grade AI-Powered Project Structure Generator

    

    Projeleriniz için **akıllı**, **dinamik** ve **sektöre özgü** context engineering yapıları oluşturun.

    """)
    
    with gr.Tab("🎯 Proje Bilgileri"):
        with gr.Row():
            with gr.Column(scale=1):
                gr.Markdown("### 🏢 Proje Detayları")
                
                project_type = gr.Dropdown(
                    choices=PROJECT_TYPES,
                    label="Proje Türü",
                    value=PROJECT_TYPES[0]
                )
                
                industry = gr.Dropdown(
                    choices=INDUSTRIES,
                    label="Sektör",
                    value="Teknoloji/Yazılım"
                )
                
                features = gr.CheckboxGroup(
                    choices=PROJECT_FEATURES,
                    label="Proje Özellikleri",
                    value=["Kullanıcı Yönetimi"]
                )
                
            with gr.Column(scale=2):
                gr.Markdown("### 📝 Proje Açıklaması")
                
                project_idea_input = gr.Textbox(
                    label="Proje Fikri",
                    lines=6,
                    value="Detaylı proje açıklamanızı buraya yazın..."
                )
                
                tech_details_input = gr.Textbox(
                    label="Teknik Detaylar (Opsiyonel)",
                    lines=4,
                    value="Kullanmayı düşündüğünüz teknolojiler, özel gereksinimler..."
                )
    
    with gr.Tab("🔧 LLM Konfigürasyonu"):
        with gr.Row():
            with gr.Column():
                llm_provider = gr.Dropdown(
                    choices=list(LLM_PROVIDERS.keys()),
                    label="AI Sağlayıcı",
                    value="Gemini"
                )
                
                llm_model = gr.Dropdown(
                    choices=[],
                    label="Model",
                    interactive=False
                )
                
            with gr.Column():
                api_key_input = gr.Textbox(
                    label="API Anahtarı",
                    type="password"
                )
                
                validate_button = gr.Button("🔐 API Doğrula", variant="secondary")
                validation_status = gr.Markdown()
    
    with gr.Tab("🏗️ Yapı Oluşturma"):
        gr.Markdown("""

        ### 🎨 Akıllı Dosya Üretimi

        Her bölüm, projenizin özelliklerine göre **dinamik olarak** oluşturulur.

        """)
        
        with gr.Row():
            context_btn = gr.Button("📘 Context Oluştur", variant="primary", size="lg")
            commands_btn = gr.Button("⚡ Commands Oluştur", variant="primary", size="lg")
            prompts_btn = gr.Button("💡 Prompts Oluştur", variant="primary", size="lg")
        
        with gr.Row():
            generate_all_btn = gr.Button("🎯 Hepsini Oluştur (Sırayla)", variant="secondary")
            analyze_btn = gr.Button("🔍 Proje Analizi", variant="secondary")
        
        output_display = gr.Markdown()
        analysis_display = gr.Markdown()
    
    with gr.Tab("📦 İndirme & İstatistikler"):
        with gr.Row():
            download_btn = gr.Button("📥 Profesyonel Paket İndir", variant="primary", size="lg")
            clear_btn = gr.Button("🗑️ Temizle", variant="secondary")
        
        download_file = gr.File(label="İndirilebilir Paket", visible=False)
        stats_display = gr.Markdown()
    
    # Event handlers
    def update_models(provider):
        if provider in LLM_PROVIDERS:
            return gr.Dropdown(
                choices=LLM_PROVIDERS[provider].get("models", []),
                interactive=True
            )
        return gr.Dropdown(choices=[], interactive=False)
    
    def validate_api(provider, api_key):
        if not provider or not api_key:
            return gr.Markdown("<p style='color:orange;'>⚠️ Sağlayıcı ve API anahtarı gerekli.</p>"), gr.Dropdown()
        
        is_valid, message = validate_api_key(provider, api_key)
        if is_valid:
            models = get_llm_models(provider, api_key)
            return (
                gr.Markdown(f"<p style='color:green;'>✅ {message}</p>"),
                gr.Dropdown(choices=models, value=models[0] if models else None, interactive=True)
            )
        return (
            gr.Markdown(f"<p style='color:red;'>❌ {message}</p>"),
            gr.Dropdown(choices=[], interactive=False)
        )
    
    def analyze_project_details(project_type, project_idea, industry, features):
        analysis = analyze_project(project_type, project_idea, industry, features)
        
        report = f"""

<div style='background: #f9fafb; padding: 20px; border-radius: 12px; border: 1px solid #e5e7eb;'>

    <h3>🔍 Proje Analiz Raporu</h3>

    

    <div style='display: grid; grid-template-columns: repeat(2, 1fr); gap: 20px;'>

        <div>

            <h4>📊 Karmaşıklık Analizi</h4>

            <ul>

                <li>Context dosyaları: ~{analysis['complexity']['context']} adet</li>

                <li>Command dosyaları: ~{analysis['complexity']['commands']} adet</li>

                <li>Prompt şablonları: ~{analysis['complexity']['prompts']} adet</li>

            </ul>

        </div>

        

        <div>

            <h4>🛠️ Teknoloji Önerileri</h4>

            <ul>

                {"".join(f"<li>{tech}</li>" for tech in analysis['tech_recommendations'][:5])}

            </ul>

        </div>

    </div>

    

    <div style='margin-top: 20px;'>

        <h4>🏗️ Önerilen Mimari</h4>

        <p>{analysis['architecture_suggestion']}</p>

    </div>

    

    {f'''

    <div style='margin-top: 20px;'>

        <h4>🏢 Sektör Gereksinimleri ({industry})</h4>

        <ul>

            {"".join(f"<li>{req}</li>" for req in analysis['industry_requirements'])}

        </ul>

    </div>

    ''' if analysis['industry_requirements'] else ''}

</div>

"""
        return report
    
    def generate_all_sequential(provider, model, api_key, project_type, project_idea, 

                               industry, features, tech_details):
        results = []
        for section in ['context', 'commands', 'prompts']:
            result, _, _ = generate_section_advanced(
                provider, model, api_key, project_type, project_idea,
                section, industry, features, tech_details
            )
            results.append(result)
            time.sleep(2)  # Rate limiting
        
        return "<br><br>".join(results)
    
    def clear_all():
        global generated_files, stats, project_analysis
        generated_files = {'context': {}, 'commands': {}, 'prompts': {}}
        stats = {'context': {'tokens': 0, 'time': 0}, 'commands': {'tokens': 0, 'time': 0}, 'prompts': {'tokens': 0, 'time': 0}}
        project_analysis = {}
        return gr.Markdown("✅ Tüm veriler temizlendi."), gr.File(visible=False), gr.Markdown("")
    
    # Wire up events
    llm_provider.change(fn=update_models, inputs=llm_provider, outputs=llm_model)
    validate_button.click(fn=validate_api, inputs=[llm_provider, api_key_input], outputs=[validation_status, llm_model])
    
    # Analyze button
    analyze_btn.click(
        fn=analyze_project_details,
        inputs=[project_type, project_idea_input, industry, features],
        outputs=analysis_display
    )
    
    # Section buttons
    context_btn.click(
        fn=lambda p, m, a, pt, pi, i, f, t: generate_section_advanced(p, m, a, pt, pi, 'context', i, f, t),
        inputs=[llm_provider, llm_model, api_key_input, project_type, project_idea_input, industry, features, tech_details_input],
        outputs=[output_display, download_file, analysis_display]
    )
    
    commands_btn.click(
        fn=lambda p, m, a, pt, pi, i, f, t: generate_section_advanced(p, m, a, pt, pi, 'commands', i, f, t),
        inputs=[llm_provider, llm_model, api_key_input, project_type, project_idea_input, industry, features, tech_details_input],
        outputs=[output_display, download_file, analysis_display]
    )
    
    prompts_btn.click(
        fn=lambda p, m, a, pt, pi, i, f, t: generate_section_advanced(p, m, a, pt, pi, 'prompts', i, f, t),
        inputs=[llm_provider, llm_model, api_key_input, project_type, project_idea_input, industry, features, tech_details_input],
        outputs=[output_display, download_file, analysis_display]
    )
    
    generate_all_btn.click(
        fn=generate_all_sequential,
        inputs=[llm_provider, llm_model, api_key_input, project_type, project_idea_input, industry, features, tech_details_input],
        outputs=output_display
    )
    
    # Download and clear
    download_btn.click(fn=download_all_professional, outputs=[download_file, stats_display])
    clear_btn.click(fn=clear_all, outputs=[output_display, download_file, stats_display])

if __name__ == "__main__":
    demo.launch(debug=True, share=False)