> ## Documentation Index
> Fetch the complete documentation index at: https://cndoc-langchain.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Undatasio 集成

> 使用 LangChain Python 与 Undatasio 文档加载器集成。

本笔记本提供了快速入门 **UnDatasIO 文档加载器** 的概览。UnDatasIO 支持高效加载和解析多种文档格式，包括 PDF、PNG、JPG、JPEG 和 JFIF，具备文档懒加载和原生异步支持等功能，所有操作均通过 UnDatasIO 安全的云 API 完成。这些能力使得处理后的数据可直接用于 RAG 等生成式 AI 工作流。

有关所有功能和配置的详细文档，请参阅官方 API 参考。

## 概览

### 加载器功能

|         来源        | 文档懒加载 | 原生异步支持 |
| :---------------: | :---: | :----: |
| `UnDatasIOLoader` |   ✅   |    ✅   |

## 设置

### 凭证

UnDatasIO 需要一个 API 令牌。
在 [undatas.io](https://undatas.io) 生成一个免费令牌，并在下方单元格中设置：

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
import getpass
import os

if "UNDATASIO_TOKEN" not in os.environ:
    os.environ["UNDATASIO_TOKEN"] = getpass.getpass(
        "Enter your UnDatasIO API token: "
    )
```

### 安装

#### 常规安装

运行本笔记本其余部分需要以下软件包。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
# 安装软件包，兼容 API 分区
pip install langchain-undatasio
```

### 初始化

**UnDatasIOLoader** 支持通过 UnDatasIO 云 API 进行单文件上传和解析。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
from langchain_undatasio import UnDatasIOLoader

loader = UnDatasIOLoader(
    token=os.environ["UNDATASIO_TOKEN"],
    file_path="demo.pdf"
)
```

### 加载

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
docs = loader.load()
docs[0]
```

```text theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
Document(
    metadata={'source': 'demo.pdf', 'task_id': 't1', 'file_id': 'f1'},
    page_content='Growing a Tail: Increasing Output Diversity in Large Language Models\n\nAuthors: Michal Shur-Ofry1, Bar Horowitz-Amsalem1†, Adir Rahamim2, Yonatan Belinkov2*\n\nAffiliations:\n\n1Law Faculty, Hebrew University of Jerusalem; Jerusalem, Israel.\n\n2Faculty of Computer Science, Technion – I'
)
```

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
print(docs[0].page_content[:300])
```

```text theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
Growing a Tail: Increasing Output Diversity in Large Language Models

Authors: Michal Shur-Ofry1, Bar Horowitz-Amsalem1†, Adir Rahamim2, Yonatan Belinkov2*

Affiliations:

1Law Faculty, Hebrew University of Jerusalem; Jerusalem, Israel.

2Faculty of Computer Science, Technion – I
```

### 懒加载

**UnDatasIOLoader** 支持懒加载，以实现内存高效的迭代。

```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
pages = []
for doc in loader.lazy_load():
    pages.append(doc)

pages[0]
```

```text theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
Document(
    metadata={'source': 'demo.pdf', 'task_id': 't1', 'file_id': 'f1'},
    page_content='Growing a Tail: Increasing Output Diversity in Large Language Models\n\nAuthors: Michal Shur-Ofry1, Bar Horowitz-Amsalem1†, Adir Rahamim2, Yonatan Belinkov2*\n\nAffiliations:\n\n1Law Faculty, Hebrew University of Jerusalem; Jerusalem, Israel.\n\n2Faculty of Computer Science, Technion – I'
)
```

## 另请参阅

* [UnDatasIO](https://undatas.io)
* [langchain-undatasio](https://pypi.org/project/langchain-undatasio/)

***

<div className="source-links">
  <Callout icon="terminal-2">
    [连接这些文档](/use-these-docs) 到 Claude、VSCode 等，通过 MCP 获取实时答案。
  </Callout>

  <Callout icon="edit">
    [在 GitHub 上编辑此页面](https://github.com/langchain-ai/docs/edit/main/src/oss/python/integrations/document_loaders/undatasio.mdx) 或 [提交问题](https://github.com/langchain-ai/docs/issues/new/choose)。
  </Callout>
</div>