# Clone del repository principale
git clone https://github.com/risik01/stock-ai.git
cd stock-ai

# Verifica struttura files
ls -la

# Crea virtual environment
python3 -m venv stock-ai-env

# Attiva environment
# Su Linux/macOS:
source stock-ai-env/bin/activate
# Su Windows:
# stock-ai-env\Scripts\activate

# Verifica attivazione
which python
# Output: /path/to/stock-ai-env/bin/python

# Crea conda environment
conda create -n stock-ai python=3.10
conda activate stock-ai

# Verifica installazione
python --version
# Output: Python 3.10.x

# Aggiorna pip
pip install --upgrade pip

# Installa requirements principali
pip install -r requirements.txt

# Verifica installazioni chiave
python -c "import torch; print(f'PyTorch: {torch.__version__}')"
python -c "import pandas; print(f'Pandas: {pandas.__version__}')"
python -c "import yfinance; print('YFinance: OK')"

# Dependencies specifiche per News Trading AI
pip install feedparser==6.0.10
pip install textblob==0.18.0
pip install vaderSentiment==3.3.2
pip install nltk==3.8.1
pip install beautifulsoup4==4.12.2
pip install lxml==4.9.3

# Download NLTK data
python -c "import nltk; nltk.download('vader_lexicon'); nltk.download('punkt')"

# Test importazioni principali
python -c "
import torch
import pandas as pd
import numpy as np
import yfinance as yf
import feedparser
import textblob
import vaderSentiment
print('✅ Tutte le dependencies installate correttamente!')
"

# Crea directories necessarie
mkdir -p data/models
mkdir -p data/cache
mkdir -p logs
mkdir -p config/backup

# Imposta permessi
chmod 755 data/ logs/ config/

# Copia template configurazione
cp config/settings.json.example config/settings.json

# Edita configurazione base
nano config/settings.json

# Test 1: Data Collector
python src/data_collector.py --test

# Test 2: RL Agent (quick test)
python src/test_rl.py

# Test 3: Portfolio System
python src/portfolio.py --test

# Test 4: Strategy Engine
python src/strategy_engine.py --test

# Test News RSS Collector
python trading-new/news_rss_collector.py --test

# Test Sentiment Analysis
python trading-new/news_sentiment_analyzer.py --test

# Test News Trading AI
python trading-new/news_based_trading_ai.py --test

# Avvia dashboard principale
python src/web_dashboard.py

# Test accesso
curl http://localhost:5000
# O apri browser: http://localhost:5000

# Avvia news dashboard
python trading-new/news_web_dashboard.py

# Test accesso
curl http://localhost:5001
# O apri browser: http://localhost:5001

# Test CLI principale
python src/cli_monitor.py --help

# Test CLI news trading  
python trading-new/news_trading_cli.py --help

# Crea database SQLite
python -c "
import sqlite3
conn = sqlite3.connect('data/trading.db')
conn.execute('''
    CREATE TABLE IF NOT EXISTS trades (
        id INTEGER PRIMARY KEY,
        symbol TEXT,
        action TEXT,
        quantity INTEGER,
        price REAL,
        timestamp DATETIME
    )
''')
conn.commit()
conn.close()
print('✅ Database SQLite creato')
"

# Installa PostgreSQL (Ubuntu)
sudo apt update
sudo apt install postgresql postgresql-contrib

# Crea database
sudo -u postgres createdb stockai
sudo -u postgres psql -c "CREATE USER stockai WITH PASSWORD 'your_password';"
sudo -u postgres psql -c "GRANT ALL PRIVILEGES ON DATABASE stockai TO stockai;"

# Aggiorna config/settings.json
# "database": {
#     "type": "postgresql",
#     "host": "localhost",
#     "port": 5432,
#     "name": "stockai",
#     "user": "stockai",
#     "password": "your_password"
# }

# Crea file per API keys
touch config/.env

# Edita file .env
nano config/.env

# Alpha Vantage (per dati estesi)
ALPHA_VANTAGE_API_KEY=your_alpha_vantage_key

# News APIs
NEWS_API_KEY=your_news_api_key
POLYGON_API_KEY=your_polygon_key

# Discord/Slack (per notifiche)
DISCORD_WEBHOOK_URL=your_discord_webhook
SLACK_WEBHOOK_URL=your_slack_webhook

# Imposta permessi restrittivi
chmod 600 config/.env

# Aggiungi al .gitignore
echo "config/.env" >> .gitignore
echo "config/secrets/" >> .gitignore

# 1. Raccolta dati iniziale
python src/data_collector.py --symbols AAPL,GOOGL --period 1y

# 2. Training rapido RL Agent
python src/train_rl.py --episodes 100 --quick

# 3. Test ciclo trading completo
python src/main.py --mode test --duration 60

# 4. Test News Trading
python trading-new/news_trading_cli.py cycle

# Log sistema principale
tail -f logs/trading.log

# Log news trading
tail -f data/aggressive_trader.log

# Log performance
tail -f data/performance_history.json

# Errore: No module named 'torch'
pip install torch torchvision torchaudio

# Errore: SSL certificate error
pip install --trusted-host pypi.org --trusted-host files.pythonhosted.org <package>

# Errore: Permission denied
sudo chown -R $USER:$USER /path/to/stock-ai

# Test connettività Yahoo Finance
python -c "import yfinance as yf; print(yf.download('AAPL', period='1d'))"

# Test RSS feeds
python -c "import feedparser; print(len(feedparser.parse('https://feeds.finance.yahoo.com/rss/2.0/headline').entries))"

# Verifica sintassi JSON
python -c "import json; json.load(open('config/settings.json'))"

# Reset configurazione
cp config/settings.json.example config/settings.json

# Pulizia cache
rm -rf data/cache/*
rm -rf logs/*.log

# Verifica spazio
df -h
du -sh data/

# Build Docker image
docker build -t stock-ai .

# Run container
docker run -d -p 5000:5000 -p 5001:5001 --name stock-ai-container stock-ai

# Verifica
docker ps
docker logs stock-ai-container

# Crea service systemd
sudo nano /etc/systemd/system/stock-ai.service

# Abilita e avvia service
sudo systemctl enable stock-ai.service
sudo systemctl start stock-ai.service
sudo systemctl status stock-ai.service

# Installa monitoring tools
pip install psutil prometheus_client

# Setup log rotation
sudo nano /etc/logrotate.d/stock-ai