3

2026-09-01 13:00:36 +08:00
parent 1d8621717a
commit dfd38c99a0
35 changed files with 3141 additions and 21 deletions
@@ -0,0 +1,420 @@
+"""公共 AI 页面路由（/k/**）。
+
+规则：
+- 零 JS、零 Cookie、零登录、SSR 输出、标准 HTML
+- <meta name="robots" content="noindex,nofollow,noarchive">
+- <meta name="referrer" content="no-referrer">
+- 限流：内存 TokenBucket
+- 统一 404 防存在性探测
+"""
+
+import json
+
+from fastapi import APIRouter, Depends, Query, Request, Response
+from fastapi.responses import HTMLResponse, PlainTextResponse
+from sqlalchemy.orm import Session
+
+from app.api.deps import get_db
+from app.core.errors import NotFoundError, RateLimitedError
+from app.core.rate_limit import check_rate_limit
+from app.core.security import decrypt_token
+from app.services.kb_public_service import KbPublicService
+
+router = APIRouter(prefix="/k", tags=["public"])
+
+
+def _rate_limit(request: Request, token: str) -> None:
+    """限流检查。"""
+    ip = request.client.host if request.client else "unknown"
+    if not check_rate_limit(token_key=token[:16], ip_key=ip):
+        raise RateLimitedError()
+
+
+def _robots_meta() -> str:
+    return '<meta name="robots" content="noindex,nofollow,noarchive">'
+
+
+def _referrer_meta() -> str:
+    return '<meta name="referrer" content="no-referrer">'
+
+
+# --- 知识库入口（后缀路由必须先于无后缀路由注册）---
+
+
+@router.get("/{token}.md")
+def kb_index_markdown(
+    token: str,
+    request: Request,
+    db: Session = Depends(get_db),
+) -> PlainTextResponse:
+    """知识库首页（Markdown）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    docs, _ = svc.list_documents(kb)
+
+    lines = [f"# {kb.name}", ""]
+    if kb.description:
+        lines.append(kb.description)
+        lines.append("")
+
+    lines.append("## 文档列表")
+    lines.append("")
+
+    for doc in docs:
+        title = doc.title or doc.original_filename
+        lines.append(f"### {title}")
+        lines.append(f"- 类型：{doc.file_ext}")
+        if doc.description:
+            lines.append(f"- 描述：{doc.description}")
+        if doc.keywords:
+            lines.append(f"- 关键词：{doc.keywords}")
+        if doc.content_summary:
+            lines.append(f"- 摘要：{doc.content_summary}")
+        lines.append(f"- 更新时间：{doc.updated_at}")
+        lines.append("")
+
+    return PlainTextResponse(content="\n".join(lines), media_type="text/markdown")
+
+
+@router.get("/{token}.txt")
+def kb_index_text(
+    token: str,
+    request: Request,
+    db: Session = Depends(get_db),
+) -> PlainTextResponse:
+    """知识库首页（纯文本）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    docs, _ = svc.list_documents(kb)
+
+    lines = [kb.name, "=" * len(kb.name), ""]
+    if kb.description:
+        lines.append(kb.description)
+        lines.append("")
+
+    lines.append("文档列表：")
+    lines.append("")
+
+    for i, doc in enumerate(docs, 1):
+        title = doc.title or doc.original_filename
+        lines.append(f"{i}. {title}")
+        if doc.description:
+            lines.append(f"   描述：{doc.description}")
+        if doc.keywords:
+            lines.append(f"   关键词：{doc.keywords}")
+        lines.append("")
+
+    return PlainTextResponse(content="\n".join(lines), media_type="text/plain")
+
+
+@router.get("/{token}.json")
+def kb_index_json(
+    token: str,
+    request: Request,
+    db: Session = Depends(get_db),
+) -> Response:
+    """知识库首页（JSON）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    docs, _ = svc.list_documents(kb)
+
+    doc_list = []
+    for doc in docs:
+        doc_list.append({
+            "title": doc.title or doc.original_filename,
+            "file_type": doc.file_ext,
+            "description": doc.description,
+            "summary": doc.content_summary,
+            "keywords": doc.keywords.split(",") if doc.keywords else [],
+            "updated_at": doc.updated_at,
+        })
+
+    data = {
+        "name": kb.name,
+        "description": kb.description,
+        "document_count": len(doc_list),
+        "documents": doc_list,
+    }
+
+    return Response(
+        content=json.dumps(data, ensure_ascii=False, indent=2),
+        media_type="application/json",
+    )
+
+
+@router.get("/{token}")
+def kb_index_html(
+    token: str,
+    request: Request,
+    page: int = Query(1, ge=1),
+    db: Session = Depends(get_db),
+) -> HTMLResponse:
+    """知识库首页（HTML）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    docs, total = svc.list_documents(kb, page=page, page_size=50)
+
+    doc_rows = ""
+    for doc in docs:
+        doc_url = f"/k/{token}/doc/{decrypt_token(doc.doc_token_encrypted) if doc.doc_token_encrypted else ''}"
+        keywords_html = f'<span class="keywords">关键词：{doc.keywords}</span>' if doc.keywords else ""
+        summary_html = f'<p class="summary">{doc.content_summary or ""}</p>' if doc.content_summary else ""
+        doc_rows += f"""
+        <div class="doc-item">
+          <h3><a href="{doc_url}">{doc.title or doc.original_filename}</a></h3>
+          <p class="meta">类型：{doc.file_ext} | 更新：{doc.updated_at}</p>
+          {summary_html}
+          {keywords_html}
+        </div>
+        """
+
+    total_pages = (total + 49) // 50
+    pagination = ""
+    if total_pages > 1:
+        pagination = f'<p class="pagination">第 {page} / {total_pages} 页</p>'
+
+    html = f"""<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+    <meta charset="UTF-8">
+    <title>{kb.name}</title>
+    {_robots_meta()}
+    {_referrer_meta()}
+    <style>
+        body {{ font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, sans-serif; max-width: 800px; margin: 0 auto; padding: 20px; line-height: 1.6; }}
+        h1 {{ color: #333; }}
+        .doc-item {{ border-bottom: 1px solid #eee; padding: 15px 0; }}
+        .doc-item h3 {{ margin: 0 0 5px 0; }}
+        .doc-item a {{ color: #0066cc; text-decoration: none; }}
+        .doc-item a:hover {{ text-decoration: underline; }}
+        .meta {{ color: #666; font-size: 0.9em; margin: 5px 0; }}
+        .summary {{ color: #444; font-size: 0.95em; margin: 5px 0; }}
+        .keywords {{ color: #888; font-size: 0.85em; }}
+        .pagination {{ color: #666; font-size: 0.9em; text-align: center; }}
+        .footer {{ margin-top: 30px; padding-top: 15px; border-top: 1px solid #eee; color: #999; font-size: 0.85em; }}
+    </style>
+</head>
+<body>
+    <h1>{kb.name}</h1>
+    {"<p>" + kb.description + "</p>" if kb.description else ""}
+    <h2>文档列表</h2>
+    {doc_rows if doc_rows else "<p>暂无文档。</p>"}
+    {pagination}
+    <div class="footer">
+        <p>This page is an AI-readable knowledge base index. Use the document links above to retrieve specific documents.</p>
+        <p>本页为 AI 可读知识库目录，请通过上述文档链接获取具体内容。</p>
+    </div>
+</body>
+</html>"""
+    return HTMLResponse(content=html)
+
+
+# --- 单文档访问 ---
+
+
+@router.get("/{token}/doc/{doc_token}")
+def doc_page_html(
+    token: str,
+    doc_token: str,
+    request: Request,
+    db: Session = Depends(get_db),
+) -> HTMLResponse:
+    """文档页面（HTML）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    doc = svc.get_document_by_token(kb, doc_token)
+    markdown_content = svc.get_document_markdown(doc)
+
+    # Markdown → HTML（简单转换）
+    html_content = _markdown_to_html(markdown_content)
+
+    keywords_html = f"<p><strong>关键词：</strong>{doc.keywords}</p>" if doc.keywords else ""
+
+    html = f"""<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+    <meta charset="UTF-8">
+    <title>{doc.title or doc.original_filename}</title>
+    {_robots_meta()}
+    {_referrer_meta()}
+    <style>
+        body {{ font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, sans-serif; max-width: 800px; margin: 0 auto; padding: 20px; line-height: 1.8; }}
+        h1 {{ color: #333; }}
+        .meta {{ color: #666; font-size: 0.9em; margin-bottom: 20px; }}
+        .content {{ margin-top: 20px; }}
+        .content h1, .content h2, .content h3 {{ color: #333; }}
+        .content pre {{ background: #f5f5f5; padding: 15px; border-radius: 5px; overflow-x: auto; }}
+        .content code {{ background: #f0f0f0; padding: 2px 5px; border-radius: 3px; }}
+        .content table {{ border-collapse: collapse; width: 100%; }}
+        .content th, .content td {{ border: 1px solid #ddd; padding: 8px; text-align: left; }}
+        .content th {{ background: #f5f5f5; }}
+        .back {{ margin-top: 30px; }}
+        .back a {{ color: #0066cc; }}
+    </style>
+</head>
+<body>
+    <h1>{doc.title or doc.original_filename}</h1>
+    <div class="meta">
+        <p>类型：{doc.file_ext} | 更新：{doc.updated_at}</p>
+        {"<p><strong>描述：</strong>" + doc.description + "</p>" if doc.description else ""}
+        {keywords_html}
+    </div>
+    <div class="content">
+        {html_content}
+    </div>
+    <div class="back">
+        <a href="/k/{token}">← 返回知识库目录</a>
+    </div>
+</body>
+</html>"""
+    return HTMLResponse(content=html)
+
+
+@router.get("/{token}/doc/{doc_token}.md")
+def doc_page_markdown(
+    token: str,
+    doc_token: str,
+    request: Request,
+    db: Session = Depends(get_db),
+) -> PlainTextResponse:
+    """文档（Markdown）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    doc = svc.get_document_by_token(kb, doc_token)
+    markdown_content = svc.get_document_markdown(doc)
+    return PlainTextResponse(content=markdown_content, media_type="text/markdown")
+
+
+@router.get("/{token}/doc/{doc_token}.txt")
+def doc_page_text(
+    token: str,
+    doc_token: str,
+    request: Request,
+    db: Session = Depends(get_db),
+) -> PlainTextResponse:
+    """文档（纯文本）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    doc = svc.get_document_by_token(kb, doc_token)
+    markdown_content = svc.get_document_markdown(doc)
+
+    # 去掉 Markdown 标记
+    import re
+
+    text = re.sub(r"[#*_`\[\]()>]", "", markdown_content)
+    text = re.sub(r"\n{3,}", "\n\n", text)
+    return PlainTextResponse(content=text.strip(), media_type="text/plain")
+
+
+# --- 搜索 ---
+
+
+@router.get("/{token}/search")
+def search_html(
+    token: str,
+    q: str = Query(..., min_length=1),
+    page: int = Query(1, ge=1),
+    request: Request = None,
+    db: Session = Depends(get_db),
+) -> HTMLResponse:
+    """搜索文档（HTML）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    results, total = svc.search_documents(kb, q, page=page)
+
+    result_items = ""
+    for item in results:
+        doc_url = f"/k/{token}/doc/{item['url_hint'] or ''}"
+        result_items += f"""
+        <div class="result-item">
+          <h3><a href="{doc_url}">{item['title']}</a></h3>
+          <p class="meta">类型：{item['file_type']} | 更新：{item['updated_at']}</p>
+          {"<p>" + (item.get('description') or '') + "</p>"}
+        </div>
+        """
+
+    html = f"""<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+    <meta charset="UTF-8">
+    <title>搜索：{q} - {kb.name}</title>
+    {_robots_meta()}
+    {_referrer_meta()}
+    <style>
+        body {{ font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, sans-serif; max-width: 800px; margin: 0 auto; padding: 20px; line-height: 1.6; }}
+        .result-item {{ border-bottom: 1px solid #eee; padding: 15px 0; }}
+        .result-item a {{ color: #0066cc; text-decoration: none; }}
+        .meta {{ color: #666; font-size: 0.9em; }}
+    </style>
+</head>
+<body>
+    <h1>搜索：{q}</h1>
+    <p>共找到 {total} 个结果</p>
+    {result_items if result_items else "<p>未找到相关文档。</p>"}
+    <p><a href="/k/{token}">← 返回知识库目录</a></p>
+</body>
+</html>"""
+    return HTMLResponse(content=html)
+
+
+@router.get("/{token}/search.json")
+def search_json(
+    token: str,
+    q: str = Query(..., min_length=1),
+    page: int = Query(1, ge=1),
+    request: Request = None,
+    db: Session = Depends(get_db),
+) -> Response:
+    """搜索文档（JSON）。"""
+    _rate_limit(request, token)
+    svc = KbPublicService(db)
+    kb = svc.get_kb_by_token(token)
+    results, total = svc.search_documents(kb, q, page=page)
+
+    data = {
+        "query": q,
+        "total": total,
+        "results": results,
+    }
+    return Response(
+        content=json.dumps(data, ensure_ascii=False, indent=2),
+        media_type="application/json",
+    )
+
+
+def _markdown_to_html(markdown: str) -> str:
+    """简单 Markdown → HTML 转换（安全处理）。"""
+    import re
+
+    # 转义 HTML 特殊字符
+    html = markdown.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;")
+
+    # 标题
+    html = re.sub(r"^#### (.+)$", r"<h4>\1</h4>", html, flags=re.MULTILINE)
+    html = re.sub(r"^### (.+)$", r"<h3>\1</h3>", html, flags=re.MULTILINE)
+    html = re.sub(r"^## (.+)$", r"<h2>\1</h2>", html, flags=re.MULTILINE)
+    html = re.sub(r"^# (.+)$", r"<h1>\1</h1>", html, flags=re.MULTILINE)
+
+    # 粗体/斜体
+    html = re.sub(r"\*\*(.+?)\*\*", r"<strong>\1</strong>", html)
+    html = re.sub(r"\*(.+?)\*", r"<em>\1</em>", html)
+
+    # 代码块
+    html = re.sub(r"```[\s\S]*?```", lambda m: f"<pre><code>{m.group(0)[3:-3]}</code></pre>", html)
+
+    # 行内代码
+    html = re.sub(r"`([^`]+)`", r"<code>\1</code>", html)
+
+    # 段落（双换行 → <p>）
+    html = re.sub(r"\n\n+", "</p><p>", html)
+    html = f"<p>{html}</p>"
+
+    return html