#!/bin/bash
# ============================================================================
# SEO 检查脚本 - 检查文章中的链接和SEO问题
# ============================================================================

# 配置
WORKSPACE="/root/.openclaw/workspace"
HQBSH_DIR="${WORKSPACE}/hqbsh"
SEO_LOG="/var/log/openclaw/seo-check.log"

mkdir -p "$(dirname "$SEO_LOG")"

log() {
    echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1" >> "$SEO_LOG"
}

# ============================================================================
# 检查问题链接
# ============================================================================
check_broken_links() {
    local file="$1"
    local issues=()
    
    # 1. 检查伪链接格式 [标题]()
    while IFS= read -r line; do
        if echo "$line" | grep -qE '\[.+\]\(\s*\)'; then
            issues+=("空括号链接: $line")
        fi
    done < "$file"
    
    # 2. 检查锚点伪链接 [标题](#)
    while IFS= read -r line; do
        if echo "$line" | grep -qE '\[.+\]\(#\)'; then
            issues+=("空锚点链接: $line")
        fi
    done < "$file"
    
    # 3. 检查纯文本"原贴"字样但无链接
    if grep -qE '原贴.*：$' "$file"; then
        issues+=("原贴后面没有链接")
    fi
    
    # 4. 检查链接格式不完整 [标题](链接不完整
    if grep -qE '\]\([^\)]+$' "$file"; then
        issues+=("未闭合的链接括号")
    fi
    
    printf '%s\n' "${issues[@]}"
}

# ============================================================================
# 检查SEO元素
# ============================================================================
check_seo_elements() {
    local file="$1"
    local issues=()
    
    # 检查是否缺少 H1/H2 标题
    if ! grep -qE '^# ' "$file"; then
        issues+=("缺少 H1 标题 (# 标题)")
    fi
    
    # 检查图片是否缺少 alt 文本
    while IFS= read -r line; do
        if echo "$line" | grep -qE '!\[\]\('; then
            issues+=("图片缺少 alt 文本: $line")
        fi
    done < "$file"
    
    # 检查段落是否太长（超过300字）
    local content=$(cat "$file" | sed 's/```.*//g' | sed 's/```//g')
    local paragraphs=$(echo "$content" | awk '/^[^#*>]/ {if(length > 200) print NR": "$0}' | head -5)
    if [ -n "$paragraphs" ]; then
        issues+=("长段落可能影响阅读性")
    fi
    
    printf '%s\n' "${issues[@]}"
}

# ============================================================================
# 生成SEO优化建议
# ============================================================================
generate_seo_suggestions() {
    local file="$1"
    local suggestions=()
    
    # 建议1: 外部链接应该添加真实URL
    if grep -qE '\[.+\]\(\s*\)' "$file"; then
        suggestions+=("需要修复空链接")
    fi
    
    # 建议2: 内部链接应该添加锚点ID
    if grep -qE '\[.+\]\(#.*\)' "$file"; then
        suggestions+=("内部锚点链接格式正确 ✓")
    else
        if grep -qE '^## ' "$file"; then
            suggestions+=("建议添加目录和锚点链接")
        fi
    fi
    
    # 建议3: 元数据
    if ! grep -qE '^> ' "$file"; then
        suggestions+=("建议添加作者/时间等元数据")
    fi
    
    printf '%s\n' "${suggestions[@]}"
}

# ============================================================================
# 修复问题链接
# ============================================================================
fix_links() {
    local file="$1"
    local fixed=0
    
    # 修复空括号链接 [标题]()
    if grep -qE '\[.+\]\(\s*\)' "$file"; then
        echo "发现空链接，正在修复..."
        # 将 [标题]() 转换为 [标题](真实链接占位)
        sed -i 's/\]\(\s*\)/](请添加真实链接)/g' "$file"
        fixed=$((fixed + 1))
    fi
    
    # 修复原贴链接
    if grep -qE '原贴.*：$' "$file"; then
        echo "发现原贴占位符，正在修复..."
        sed -i 's/原贴.*：$/原贴链接: 请添加真实URL/g' "$file"
        fixed=$((fixed + 1))
    fi
    
    if [ $fixed -gt 0 ]; then
        echo "✅ 已修复 $fixed 个问题"
        return 0
    else
        echo "✅ 未发现问题链接"
        return 1
    fi
}

# ============================================================================
# 主程序 - 检查最近文章
# ============================================================================
check_recent_articles() {
    local count=${1:-5}
    
    echo "╔════════════════════════════════════════════════════╗"
    echo "║        SEO 检查 - 最近 $count 篇文章              ║"
    echo "╚════════════════════════════════════════════════════╝"
    echo ""
    
    # 查找最近修改的markdown文件
    local files=$(find "${HQBSH_DIR}" -name "*.md" -mmin -180 2>/dev/null | head "$count")
    
    if [ -z "$files" ]; then
        echo "⚠️  最近没有找到hqbsh文章"
        echo ""
        echo "检查 workspace 下的其他文章："
        files=$(find "${WORKSPACE}" -name "*.md" -mmin -180 2>/dev/null | grep -v sleep-learning | head "$count")
    fi
    
    local total_issues=0
    
    for file in $files; do
        local filename=$(basename "$file")
        echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
        echo "📄 检查: $filename"
        echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
        
        # 检查问题链接
        local issues=$(check_broken_links "$file")
        if [ -n "$issues" ]; then
            echo "❌ 问题链接："
            echo "$issues" | head -5
            total_issues=$((total_issues + $(echo "$issues" | wc -l)))
        else
            echo "✅ 链接检查：正常"
        fi
        
        # 检查SEO元素
        local seo_issues=$(check_seo_elements "$file")
        if [ -n "$seo_issues" ]; then
            echo "⚠️  SEO建议："
            echo "$seo_issues" | head -3
        fi
        
        # 生成优化建议
        local suggestions=$(generate_seo_suggestions "$file")
        if [ -n "$suggestions" ]; then
            echo ""
            echo "💡 优化建议："
            echo "$suggestions"
        fi
        
        echo ""
    done
    
    echo "╔════════════════════════════════════════════════════╗"
    echo "║  共发现 $total_issues 个问题                      ║"
    echo "╚════════════════════════════════════════════════════╝"
}

# ============================================================================
# 主程序
# ============================================================================
case "$1" in
    check)
        check_recent_articles "${2:-5}"
        ;;
    fix)
        fix_links "$2"
        ;;
    *)
        echo "用法: $0 {check [数量]|fix <文件>}"
        echo ""
        echo "示例："
        echo "  $0 check 10     # 检查最近10篇文章"
        echo "  $0 fix file.md  # 修复指定文件的链接问题"
        ;;
esac