Upload 2 files

docker-compose.yaml

@@ -1,19 +1,185 @@
-```yaml
-version: "3.8"
+version: '3.8'
 
 services:
-  gradio:
+  # Persian Legal Scraper Main Application
+  persian-legal-scraper:
     build:
       context: .
       dockerfile: Dockerfile
-    container_name: legal_dashboard_gradio
+    container_name: persian-legal-scraper-app
     restart: unless-stopped
     ports:
       - "7860:7860"
+    environment:
+      # Application Settings
+      - PYTHONPATH=/app
+      - PYTHONUNBUFFERED=1
+      - LOG_LEVEL=INFO
+      - ENVIRONMENT=production
+      
+      # Gradio Configuration
+      - GRADIO_SERVER_NAME=0.0.0.0
+      - GRADIO_SERVER_PORT=7860
+      - GRADIO_SHARE=false
+      - GRADIO_THEME=default
+      
+      # AI/ML Model Settings
+      - HF_HOME=/app/cache
+      - TRANSFORMERS_CACHE=/app/cache/transformers
+      - TORCH_HOME=/app/cache/torch
+      - TOKENIZERS_PARALLELISM=false
+      
+      # Performance Optimization
+      - PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:128
+      - OMP_NUM_THREADS=2
+      - MKL_NUM_THREADS=2
+      
+      # Database Configuration
+      - DATABASE_PATH=/app/data/iranian_legal_archive_advanced.sqlite
+      - CACHE_DB_PATH=/app/data/cache_system.sqlite
+      - EMBEDDINGS_CACHE_PATH=/app/data/embeddings_cache.pkl
+      - VECTOR_INDEX_PATH=/app/data/faiss_index.bin
+      
+      # Security Settings
+      - ANTI_DDOS_ENABLED=true
+      - MAX_REQUESTS_PER_HOUR=100
+      - REQUEST_DELAY_MIN=1
+      - REQUEST_DELAY_MAX=5
+      
+      # Legal Sources Configuration  
+      - CRAWLER_ENABLED=true
+      - MAX_CRAWL_DEPTH=2
+      - CRAWL_DELAY_SECONDS=3
+      
+    volumes:
+      # Persistent data storage
+      - persian_legal_data:/app/data
+      - persian_legal_cache:/app/cache
+      - persian_legal_logs:/app/logs
+      - persian_legal_uploads:/app/uploads
+      
+      # Configuration files (optional)
+      - ./config:/app/config:ro
+      
+    networks:
+      - persian-legal-network
+      
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:7860"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 60s
+      
+    depends_on:
+      - redis-cache
+      
+    # Resource limits for production
+    deploy:
+      resources:
+        limits:
+          memory: 4G
+          cpus: '2.0'
+        reservations:
+          memory: 2G
+          cpus: '1.0'
+
+  # Redis for Advanced Caching (Optional)
+  redis-cache:
+    image: redis:7-alpine
+    container_name: persian-legal-redis
+    restart: unless-stopped
+    command: redis-server --appendonly yes --maxmemory 256mb --maxmemory-policy allkeys-lru
+    volumes:
+      - persian_legal_redis_data:/data
+    networks:
+      - persian-legal-network
+    ports:
+      - "6379:6379"
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 10s
+      timeout: 3s
+      retries: 3
+
+  # Nginx Reverse Proxy (Optional for production)
+  nginx-proxy:
+    image: nginx:alpine
+    container_name: persian-legal-nginx
+    restart: unless-stopped
+    ports:
+      - "80:80"
+      - "443:443"
+    volumes:
+      - ./nginx/nginx.conf:/etc/nginx/nginx.conf:ro
+      - ./nginx/ssl:/etc/nginx/ssl:ro
+      - persian_legal_logs:/var/log/nginx
     networks:
-      - app_network
+      - persian-legal-network
+    depends_on:
+      - persian-legal-scraper
+    profiles:
+      - production
+
+  # Database Backup Service (Optional)
+  db-backup:
+    image: alpine:latest
+    container_name: persian-legal-backup
+    restart: unless-stopped
+    command: |
+      sh -c '
+        apk add --no-cache sqlite curl
+        while true; do
+          echo "Starting backup at $$(date)"
+          sqlite3 /app/data/iranian_legal_archive_advanced.sqlite ".backup /app/backups/backup_$$(date +%Y%m%d_%H%M%S).sqlite"
+          find /app/backups -name "backup_*.sqlite" -mtime +7 -delete
+          echo "Backup completed at $$(date)"
+          sleep 86400
+        done
+      '
     volumes:
-      - ./data:/app/data:rw
-      - ./cache:/app/cache:rw
-      - ./logs:/app/logs:rw
-      -
+      - persian_legal_data:/app/data:ro
+      - persian_legal_backups:/app/backups
+    networks:
+      - persian-legal-network
+    profiles:
+      - production
+
+volumes:
+  # Persistent data volumes
+  persian_legal_data:
+    driver: local
+    driver_opts:
+      type: none
+      o: bind
+      device: ./data
+      
+  persian_legal_cache:
+    driver: local
+    driver_opts:
+      type: none  
+      o: bind
+      device: ./cache
+      
+  persian_legal_logs:
+    driver: local
+    driver_opts:
+      type: none
+      o: bind
+      device: ./logs
+      
+  persian_legal_uploads:
+    driver: local
+    
+  persian_legal_redis_data:
+    driver: local
+    
+  persian_legal_backups:
+    driver: local
+
+networks:
+  persian-legal-network:
+    driver: bridge
+    ipam:
+      config:
+        - subnet: 172.20.0.0/16

dockerfile

@@ -1,34 +1,129 @@
-# Dockerfile
-FROM python:3.9-slim
+# Dockerfile for Advanced Iranian Legal Archive System
+# Optimized for Hugging Face Spaces deployment
 
-WORKDIR /app
+# Stage 1: Builder
+FROM python:3.10-slim AS builder
 
-# نصب dependencies سیستمی
+# Install build dependencies and system packages
 RUN apt-get update && apt-get install -y \
-    git \
+    build-essential \
+    gcc \
+    g++ \
+    libffi-dev \
+    libssl-dev \
+    wget \
     curl \
-    && rm -rf /var/lib/apt/lists/*
+    && rm -rf /var/lib/apt/lists/* \
+    && apt-get clean
 
-# کپی فایل requirements
-COPY requirements.txt .
+# Upgrade pip and install build tools
+RUN pip install --no-cache-dir --upgrade pip setuptools wheel
+
+# Create virtual environment
+RUN python -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
 
-# نصب پکیج های پایتون
+# Copy requirements and install Python dependencies
+WORKDIR /build
+COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# کپی فایل اصلی
-COPY persian_legal_scraper.py .
+# Create cache directory for models
+RUN mkdir -p /app/cache/transformers /app/cache/sentence-transformers
+
+# Pre-download Persian BERT models (Primary classification model)
+RUN python -c "from transformers import AutoModel, AutoTokenizer; \
+               print('Downloading ParsBERT...'); \
+               AutoModel.from_pretrained('HooshvareLab/bert-base-parsbert-uncased', cache_dir='/app/cache/transformers'); \
+               AutoTokenizer.from_pretrained('HooshvareLab/bert-base-parsbert-uncased', cache_dir='/app/cache/transformers')" || true
+
+# Pre-download NER model for entity recognition
+RUN python -c "from transformers import AutoModel, AutoTokenizer; \
+               print('Downloading Persian NER model...'); \
+               AutoModel.from_pretrained('HooshvareLab/bert-fa-base-uncased-ner', cache_dir='/app/cache/transformers'); \
+               AutoTokenizer.from_pretrained('HooshvareLab/bert-fa-base-uncased-ner', cache_dir='/app/cache/transformers')" || true
+
+# Pre-download Persian embedding model for semantic search
+RUN python -c "from sentence_transformers import SentenceTransformer; \
+               print('Downloading Persian embedding model...'); \
+               model = SentenceTransformer('xmanii/maux-gte-persian'); \
+               model.save('/app/cache/sentence-transformers/maux-gte-persian')" || true
+
+# Pre-download multilingual sentence transformer as fallback
+RUN python -c "from sentence_transformers import SentenceTransformer; \
+               print('Downloading multilingual model...'); \
+               model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2'); \
+               model.save('/app/cache/sentence-transformers/paraphrase-multilingual')" || true
+
+# Try to download FaBERT (latest SOTA Persian model)
+RUN python -c "from transformers import AutoModel, AutoTokenizer; \
+               print('Downloading FaBERT...'); \
+               AutoModel.from_pretrained('sbunlp/fabert', cache_dir='/app/cache/transformers'); \
+               AutoTokenizer.from_pretrained('sbunlp/fabert', cache_dir='/app/cache/transformers')" || echo "FaBERT download failed, will fallback to ParsBERT"
+
+# Stage 2: Production
+FROM python:3.10-slim
+
+# Install runtime dependencies
+RUN apt-get update && apt-get install -y \
+    sqlite3 \
+    libsqlite3-dev \
+    curl \
+    && rm -rf /var/lib/apt/lists/* \
+    && apt-get clean
+
+# Create non-root user for security
+RUN groupadd -g 1000 appuser && useradd -r -u 1000 -g appuser appuser
+
+# Copy virtual environment from builder stage
+COPY --from=builder /opt/venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Copy pre-downloaded models
+COPY --from=builder /app/cache /app/cache
+
+# Create application directories
+RUN mkdir -p /app/data /app/logs /app/uploads /app/tmp && \
+    chown -R appuser:appuser /app
+
+# Set working directory
+WORKDIR /app
 
-# ایجاد دایرکتوری های کش
-RUN mkdir -p /tmp/hf_cache /tmp/torch_cache
+# Copy application files
+COPY --chown=appuser:appuser . .
 
-# تنظیم محیط
-ENV TRANSFORMERS_CACHE=/tmp/hf_cache
-ENV HF_HOME=/tmp/hf_cache
-ENV TORCH_HOME=/tmp/torch_cache
+# Environment variables for Iranian Legal Archive System
+ENV PYTHONPATH=/app
+ENV PYTHONUNBUFFERED=1
+ENV HF_HOME=/app/cache
+ENV TRANSFORMERS_CACHE=/app/cache/transformers
+ENV TORCH_HOME=/app/cache/torch
 ENV TOKENIZERS_PARALLELISM=false
+ENV LOG_LEVEL=INFO
+ENV ENVIRONMENT=production
+
+# Gradio specific settings
+ENV GRADIO_SERVER_NAME=0.0.0.0
+ENV GRADIO_SERVER_PORT=7860
+ENV GRADIO_SHARE=false
+
+# Memory optimization for HF Spaces
+ENV PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:128
+ENV OMP_NUM_THREADS=2
+ENV MKL_NUM_THREADS=2
+
+# Switch to non-root user
+USER appuser
+
+# Create data directories with proper permissions
+RUN mkdir -p data/cache_system.sqlite data/iranian_legal_archive_advanced.sqlite data/embeddings_cache.pkl data/faiss_index.bin
+
+# Health check
+HEALTHCHECK --interval=30s --timeout=30s --start-period=60s --retries=3 \
+    CMD curl -f http://localhost:7860 || exit 1
 
-# پورت اکسپوز
+# Expose Gradio port
 EXPOSE 7860
 
-# اجرای برنامه
-CMD ["python", "persian_legal_scraper.py"]
+# Command to run the application (matching README.md app_file)
+CMD ["python", "persian_legal_scraper.py"]