nanocode/libraryofshortstories-scraper/start.sh

#!/bin/bash
# Library of Short Stories 下载器启动脚本

SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
LOG_DIR="$SCRIPT_DIR/logs"
mkdir -p "$LOG_DIR"

show_help() {
    echo "📚 Library of Short Stories 下载器"
    echo ""
    echo "用法:"
    echo "  ./start.sh fetch              # Phase 1: 获取故事索引"
    echo "  ./start.sh download           # Phase 2: 开始下载（前台）"
    echo "  ./start.sh download-bg        # Phase 2: 后台下载"
    echo "  ./start.sh status             # 查看下载状态"
    echo "  ./start.sh stop               # 停止后台下载"
    echo "  ./start.sh reset              # 重置索引（重新获取链接）"
    echo ""
    echo "目录:"
    echo "  索引: config/stories_index.json"
    echo "  故事: /home/ubuntu/media/libraryofshortstories/"
    echo "  日志: logs/"
}

case "${1:-help}" in
    fetch|index)
        echo "🔍 Phase 1: 获取故事索引..."
        python3 "$SCRIPT_DIR/scripts/fetch_index.py"
        ;;
    
    download)
        echo "📥 Phase 2: 开始下载（前台模式）..."
        echo "按 Ctrl+C 可以暂停，下次会继续从断点下载"
        echo ""
        python3 "$SCRIPT_DIR/scripts/download_stories.py"
        ;;
    
    download-bg|bg)
        echo "📥 Phase 2: 开始后台下载..."
        
        # 检查是否已在运行
        if pgrep -f "download_stories.py" > /dev/null; then
            echo "⚠️  下载进程已在运行"
            echo "查看日志: tail -f $LOG_DIR/download.log"
            exit 1
        fi
        
        # 后台运行
        nohup python3 "$SCRIPT_DIR/scripts/download_stories.py" \
            > "$LOG_DIR/download.log" 2>&1 &
        
        PID=$!
        echo "✅ 后台进程已启动 (PID: $PID)"
        echo ""
        echo "查看状态:"
        echo "  实时日志: tail -f $LOG_DIR/download.log"
        echo "  查看进程: ps aux | grep download_stories"
        echo "  停止下载: ./start.sh stop"
        ;;
    
    status)
        echo "📊 下载状态"
        echo ""
        
        # 检查进程
        if pgrep -f "download_stories.py" > /dev/null; then
            echo "✅ 后台下载正在运行"
            pgrep -f "download_stories.py" | xargs ps -o pid,etime,%cpu,%mem,command -p
        else
            echo "⏹️  后台下载未运行"
        fi
        
        echo ""
        
        # 显示索引统计
        if [ -f "$SCRIPT_DIR/config/stories_index.json" ]; then
            python3 -c "
import json
import os
from datetime import datetime

with open('$SCRIPT_DIR/config/stories_index.json') as f:
    data = json.load(f)
    stories = data.get('stories', [])
    pending = len([s for s in stories if s['status'] == 'pending'])
    downloaded = len([s for s in stories if s['status'] == 'downloaded'])
    failed = len([s for s in stories if s['status'] == 'failed'])
    
    # 实际目录数量
    media_dir = '/home/ubuntu/media/libraryofshortstories'
    if os.path.exists(media_dir):
        actual_count = len([d for d in os.listdir(media_dir) if os.path.isdir(os.path.join(media_dir, d))])
    else:
        actual_count = 0
    
    # 计算进度和预计时间
    total = len(stories)
    progress = (downloaded / total) * 100 if total > 0 else 0
    eta_minutes = (pending * 7) // 60
    eta_hours = eta_minutes // 60
    eta_remain = eta_minutes % 60
    
    print(f'📚 下载统计:')
    print(f'   总计: {total}')
    print(f'   已完成: {downloaded} ({progress:.1f}%)')
    print(f'   待下载: {pending}')
    print(f'   失败: {failed}')
    print(f'   实际目录: {actual_count}')
    print(f'')
    if pending > 0:
        if eta_hours > 0:
            print(f'   预计剩余: ~{eta_hours}小时{eta_remain}分钟')
        else:
            print(f'   预计剩余: ~{eta_remain}分钟')
    print(f'   上次索引更新: {data.get(\"fetch_date\", \"unknown\")[:19]}')
"
        else
            echo "❌ 索引文件不存在，请先运行: ./start.sh fetch"
        fi
        
        echo ""
        echo "📁 存储位置: /home/ubuntu/media/libraryofshortstories/"
        ;;
    
    stop)
        echo "⏹️  停止后台下载..."
        
        if pgrep -f "download_stories.py" > /dev/null; then
            pkill -f "download_stories.py"
            echo "✅ 已停止"
        else
            echo "⚠️  没有运行的下载进程"
        fi
        ;;
    
    reset)
        echo "⚠️  这将重置索引，重新获取链接"
        read -p "确定继续? (y/N): " confirm
        if [ "$confirm" = "y" ] || [ "$confirm" = "Y" ]; then
            rm -f "$SCRIPT_DIR/config/stories_index.json"
            echo "✅ 索引已重置"
            echo "现在可以重新运行: ./start.sh fetch"
        else
            echo "👋 已取消"
        fi
        ;;
    
    log|logs)
        if [ -f "$LOG_DIR/download.log" ]; then
            tail -50 "$LOG_DIR/download.log"
        else
            echo "暂无日志文件"
        fi
        ;;
    
    help|--help|-h)
        show_help
        ;;
    
    *)
        echo "❌ 未知命令: $1"
        show_help
        exit 1
        ;;
esac