#!/usr/bin/env python3
"""
思源笔记阅读器
支持：本地文件读取 / API 读取
"""

import json
import os
from pathlib import Path
from datetime import datetime

# 配置
SIYUAN_DIR = "/root/.openclaw/workspace/siyuan"
DATA_DIR = f"{SIYUAN_DIR}/data"

def get_all_notebooks():
    """获取所有笔记本"""
    notebooks = []
    notebooks_dir = f"{DATA_DIR}/notebooks"
    
    if not os.path.exists(notebooks_dir):
        return []
    
    for name in os.listdir(notebooks_dir):
        path = f"{notebooks_dir}/{name}"
        if os.path.isdir(path):
            notebooks.append({
                "id": name,
                "path": path
            })
    
    return notebooks

def read_sy_file(file_path):
    """读取 .sy 文件并解析"""
    try:
        with open(file_path, "r", encoding="utf-8") as f:
            data = json.load(f)
        
        # 提取标题
        title = data.get("title", "无标题")
        
        # 提取内容块
        content_blocks = []
        children = data.get("children", [])
        
        for child in children:
            block_type = child.get("type", "paragraph")
            content = ""
            
            if block_type == "paragraph":
                texts = child.get("paragraph", {}).get("rich_text", [])
                content = "".join([t.get("plain_text", "") for t in texts])
            elif block_type == "heading_1":
                texts = child.get("heading_1", {}).get("rich_text", [])
                content = "# " + "".join([t.get("plain_text", "") for t in texts])
            elif block_type == "heading_2":
                texts = child.get("heading_2", {}).get("rich_text", [])
                content = "## " + "".join([t.get("plain_text", "") for t in texts])
            elif block_type == "heading_3":
                texts = child.get("heading_3", {}).get("rich_text", [])
                content = "### " + "".join([t.get("plain_text", "") for t in texts])
            elif block_type == "bulleted_list_item":
                texts = child.get("bulleted_list_item", {}).get("rich_text", [])
                content = "• " + "".join([t.get("plain_text", "") for t in texts])
            elif block_type == "numbered_list_item":
                texts = child.get("numbered_list_item", {}).get("rich_text", [])
                content = "1. " + "".join([t.get("plain_text", "") for t in texts])
            elif block_type == "code_block":
                texts = child.get("code_block", {}).get("rich_text", [])
                content = "```\n" + "".join([t.get("plain_text", "") for t in texts]) + "\n```"
            elif block_type == "quote":
                texts = child.get("quote", {}).get("rich_text", [])
                content = "> " + "".join([t.get("plain_text", "") for t in texts])
            
            if content.strip():
                content_blocks.append(content)
        
        return {
            "title": title,
            "content": "\n".join(content_blocks),
            "blocks": len(content_blocks)
        }
    
    except Exception as e:
        return {"title": "读取错误", "content": str(e), "blocks": 0}

def get_notebook_content(notebook_id):
    """获取笔记本下所有文档"""
    notebook_path = f"{DATA_DIR}/notebooks/{notebook_id}"
    
    if not os.path.exists(notebook_path):
        return []
    
    docs = []
    for root, dirs, files in os.walk(notebook_path):
        for f in files:
            if f.endswith(".sy"):
                file_path = os.path.join(root, f)
                doc = read_sy_file(file_path)
                doc["file"] = f
                docs.append(doc)
    
    return docs

def search_content(keyword):
    """搜索内容"""
    results = []
    
    for notebook in get_all_notebooks():
        docs = get_notebook_content(notebook["id"])
        for doc in docs:
            if keyword.lower() in doc["content"].lower():
                results.append({
                    "notebook": notebook["id"],
                    "file": doc["file"],
                    "title": doc["title"],
                    "match": doc["content"][:200]
                })
    
    return results

def generate_report():
    """生成阅读报告"""
    print("\n" + "="*70)
    print("📚 思源笔记阅读报告")
    print(f"时间: {datetime.now().strftime('%Y-%m-%d %H:%M')}")
    print("="*70)
    
    notebooks = get_all_notebooks()
    print(f"\n📁 发现 {len(notebooks)} 个笔记本\n")
    
    total_docs = 0
    total_blocks = 0
    
    for notebook in notebooks:
        docs = get_notebook_content(notebook["id"])
        total_docs += len(docs)
        
        for doc in docs:
            total_blocks += doc["blocks"]
            print(f"📄 {doc['file']}: {doc['title']} ({doc['blocks']} 块内容)")
    
    print("\n" + "-"*70)
    print(f"📊 统计: {total_docs} 篇文档, {total_blocks} 个内容块")
    print("="*70)

def read_notebook(name):
    """读取指定笔记本"""
    notebooks = get_all_notebooks()
    
    # 模糊匹配笔记本名
    matched = [n for n in notebooks if name.lower() in n["id"].lower()]
    
    if not matched:
        print(f"❌ 未找到笔记本: {name}")
        return
    
    if len(matched) > 1:
        print(f"⚠️ 找到多个匹配的笔记本:")
        for m in matched:
            print(f"  - {m['id']}")
        return
    
    notebook = matched[0]
    docs = get_notebook_content(notebook["id"])
    
    print(f"\n📚 {notebook['id']}: {len(docs)} 篇文档\n")
    
    for doc in docs:
        print("="*60)
        print(f"📄 {doc['title']}")
        print("="*60)
        print(doc["content"])
        print()

def main():
    import sys
    
    if len(sys.argv) < 2:
        generate_report()
        return
    
    command = sys.argv[1]
    
    if command == "list":
        # 列出所有笔记本
        notebooks = get_all_notebooks()
        print(f"\n📚 笔记本列表 ({len(notebooks)} 个)\n")
        for n in notebooks:
            docs = get_notebook_content(n["id"])
            print(f"  📁 {n['id']}: {len(docs)} 篇文档")
    
    elif command == "read":
        # 读取指定笔记本
        if len(sys.argv) < 3:
            print("用法: python siyuan_reader.py read <笔记本名>")
            return
        name = sys.argv[2]
        read_notebook(name)
    
    elif command == "search":
        # 搜索内容
        if len(sys.argv) < 3:
            print("用法: python siyuan_reader.py search <关键词>")
            return
        keyword = sys.argv[2]
        results = search_content(keyword)
        print(f"\n🔍 搜索 '{keyword}': {len(results)} 条结果\n")
        for r in results:
            print(f"  📄 {r['file']}: {r['title']}")
            print(f"     {r['match'][:100]}...")
    
    else:
        print("用法:")
        print("  python siyuan_reader.py        # 生成报告")
        print("  python siyuan_reader.py list    # 列出笔记本")
        print("  python siyuan_reader.py read <名>  # 读取笔记本")
        print("  python siyuan_reader.py search <词>  # 搜索内容")

if __name__ == "__main__":
    main()