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

# 文档加载器集成

> 使用 LangChain JavaScript 与文档加载器集成。

文档加载器提供了一个**标准接口**，用于从不同来源（如 Slack、Notion 或 Google Drive）读取数据，并将其转换为 LangChain 的 [Document](https://reference.langchain.com/javascript/langchain-core/documents/Document) 格式。
这确保了无论数据来源如何，都能以一致的方式进行处理。

所有文档加载器都实现了 [BaseLoader](https://reference.langchain.com/javascript/classes/_langchain_core.document_loaders_base.BaseDocumentLoader.html) 接口。

## 接口

每个文档加载器可以定义自己的参数，但它们共享一个通用的 API：

* `load()`: 一次性加载所有文档。
* `loadAndSplit()`: 一次性加载所有文档，并将其拆分为更小的文档。

```typescript theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
import { CSVLoader } from "@langchain/community/document_loaders/fs/csv";

const loader = new CSVLoader(
  ...  // <-- 此处为集成特定参数
);
const data = await loader.load();
```

## 按类别划分

LangChain.js 以两种不同的方式对文档加载器进行分类：

* [文件加载器](/oss/javascript/integrations/document_loaders/file_loaders/)，用于从本地文件系统将数据加载到 LangChain 格式。
* [网络加载器](/oss/javascript/integrations/document_loaders/web_loaders/)，用于从远程源加载数据。

### 文件加载器

<Info>
  如果您想贡献一个集成，请参阅[贡献集成](/oss/javascript/contributing#add-a-new-integration)。
</Info>

#### PDF

| 文档加载器                                                                         | 描述                        | 包/API |
| ----------------------------------------------------------------------------- | ------------------------- | ----- |
| [`PDFLoader`](/oss/javascript/integrations/document_loaders/file_loaders/pdf) | 使用 pdf-parse 加载和解析 PDF 文件 | 包     |

#### 常见文件类型

| 文档加载器                                                                               | 描述                                    | 包/API |
| ----------------------------------------------------------------------------------- | ------------------------------------- | ----- |
| [CSV](/oss/javascript/integrations/document_loaders/file_loaders/csv)               | 从 CSV 文件加载数据，支持可配置的列提取                | 包     |
| [JSON](/oss/javascript/integrations/document_loaders/file_loaders/json)             | 使用 JSON 指针加载 JSON 文件以定位特定键            | 包     |
| [`JSONLines`](/oss/javascript/integrations/document_loaders/file_loaders/jsonlines) | 从 JSONLines/JSONL 文件加载数据              | 包     |
| [`Text`](/oss/javascript/integrations/document_loaders/file_loaders/text)           | 加载纯文本文件                               | 包     |
| [`DOCX`](/oss/javascript/integrations/document_loaders/file_loaders/docx)           | 加载 Microsoft Word 文档（.docx 和 .doc 格式） | 包     |
| [`EPUB`](/oss/javascript/integrations/document_loaders/file_loaders/epub)           | 加载 EPUB 文件，支持可选的章节拆分                  | 包     |
| [`PPTX`](/oss/javascript/integrations/document_loaders/file_loaders/pptx)           | 加载 PowerPoint 演示文稿                    | 包     |
| [`Subtitles`](/oss/javascript/integrations/document_loaders/file_loaders/subtitles) | 加载字幕文件（.srt 格式）                       | 包     |

#### 专用文件加载器

| 文档加载器                                                                                                   | 描述                                              | 包/API |
| ------------------------------------------------------------------------------------------------------- | ----------------------------------------------- | ----- |
| [`DirectoryLoader`](/oss/javascript/integrations/document_loaders/file_loaders/directory)               | 从目录加载所有文件，支持自定义加载器映射                            | 包     |
| [`UnstructuredLoader`](/oss/javascript/integrations/document_loaders/file_loaders/unstructured)         | 使用 Unstructured API 加载多种文件类型                    | API   |
| [`MultiFileLoader`](/oss/javascript/integrations/document_loaders/file_loaders/multi_file)              | 从多个独立文件路径加载数据                                   | 包     |
| [`ChatGPT`](/oss/javascript/integrations/document_loaders/file_loaders/chatgpt)                         | 加载 ChatGPT 对话导出文件                               | 包     |
| [Notion Markdown](/oss/javascript/integrations/document_loaders/file_loaders/notion_markdown)           | 加载导出为 Markdown 的 Notion 页面                      | 包     |
| [`OracleDocLoader`](/oss/javascript/integrations/document_loaders/file_loaders/oracleai)                | 摄取 Oracle AI Vector Search 表或 Oracle Text 支持的文件 | 包     |
| [OpenAI Whisper Audio](/oss/javascript/integrations/document_loaders/file_loaders/openai_whisper_audio) | 使用 OpenAI Whisper API 转录音频文件                    | API   |

### 网络加载器

#### 网页

| 文档加载器                                                                                                  | 描述                                     | 网络支持 | 包/API |
| ------------------------------------------------------------------------------------------------------ | -------------------------------------- | :--: | ----- |
| [`Cheerio`](/oss/javascript/integrations/document_loaders/web_loaders/web_cheerio)                     | 使用 Cheerio 加载网页（轻量级，不执行 JavaScript）    |   ✅  | 包     |
| [`Playwright`](/oss/javascript/integrations/document_loaders/web_loaders/web_playwright)               | 使用 Playwright 加载动态网页（支持 JavaScript 渲染） |   ❌  | 包     |
| [`Puppeteer`](/oss/javascript/integrations/document_loaders/web_loaders/web_puppeteer)                 | 使用 Puppeteer 加载动态网页（无头 Chrome）         |   ❌  | 包     |
| [`FireCrawl`](/oss/javascript/integrations/document_loaders/web_loaders/firecrawl)                     | 爬取网站并将其转换为适合 LLM 的 Markdown            |   ✅  | API   |
| [`Spider`](/oss/javascript/integrations/document_loaders/web_loaders/spider)                           | 快速爬虫，将网站转换为 HTML、Markdown 或文本          |   ✅  | API   |
| [`RecursiveUrlLoader`](/oss/javascript/integrations/document_loaders/web_loaders/recursive_url_loader) | 递归加载网页，跟随链接                            |   ❌  | 包     |
| [`Sitemap`](/oss/javascript/integrations/document_loaders/web_loaders/sitemap)                         | 从 sitemap.xml 加载所有页面                   |   ✅  | 包     |
| [`Browserbase`](/oss/javascript/integrations/document_loaders/web_loaders/browserbase)                 | 使用托管的无头浏览器加载网页，支持隐身模式                  |   ✅  | API   |
| [`WebPDFLoader`](/oss/javascript/integrations/document_loaders/web_loaders/pdf)                        | 在网络环境中加载 PDF 文件                        |   ✅  | 包     |

#### 云提供商

| 文档加载器                                                                                                                  | 描述                             | 网络支持 | 包/API |
| ---------------------------------------------------------------------------------------------------------------------- | ------------------------------ | :--: | ----- |
| [`S3`](/oss/javascript/integrations/document_loaders/web_loaders/s3)                                                   | 从 AWS S3 存储桶加载文件               |   ❌  | 包     |
| [Azure Blob Storage Container](/oss/javascript/integrations/document_loaders/web_loaders/azure_blob_storage_container) | 从 Azure Blob Storage 容器加载所有文件  |   ❌  | 包     |
| [Azure Blob Storage File](/oss/javascript/integrations/document_loaders/web_loaders/azure_blob_storage_file)           | 从 Azure Blob Storage 加载单个文件    |   ❌  | 包     |
| [Google Cloud Storage](/oss/javascript/integrations/document_loaders/web_loaders/google_cloud_storage)                 | 从 Google Cloud Storage 存储桶加载文件 |   ❌  | 包     |
| [Google Cloud SQL for PostgreSQL](/oss/javascript/integrations/document_loaders/web_loaders/google_cloudsql_pg)        | 从 Cloud SQL PostgreSQL 数据库加载文档 |   ✅  | 包     |

#### 生产力工具

| 文档加载器                                                                                | 描述                      | 网络支持 | 包/API |
| ------------------------------------------------------------------------------------ | ----------------------- | :--: | ----- |
| [Notion API](/oss/javascript/integrations/document_loaders/web_loaders/notionapi)    | 通过 API 加载 Notion 页面和数据库 |   ✅  | API   |
| [`Figma`](/oss/javascript/integrations/document_loaders/web_loaders/figma)           | 加载 Figma 文件数据           |   ✅  | API   |
| [`Confluence`](/oss/javascript/integrations/document_loaders/web_loaders/confluence) | 从 Confluence 空间加载页面     |   ❌  | API   |
| [`GitHub`](/oss/javascript/integrations/document_loaders/web_loaders/github)         | 从 GitHub 仓库加载文件         |   ✅  | API   |
| [`GitBook`](/oss/javascript/integrations/document_loaders/web_loaders/gitbook)       | 加载 GitBook 文档页面         |   ✅  | 包     |
| [`Jira`](/oss/javascript/integrations/document_loaders/web_loaders/jira)             | 从 Jira 项目加载问题           |   ❌  | API   |
| [`Airtable`](/oss/javascript/integrations/document_loaders/web_loaders/airtable)     | 从 Airtable 基础加载记录       |   ✅  | API   |
| [`Taskade`](/oss/javascript/integrations/document_loaders/web_loaders/taskade)       | 加载 Taskade 项目数据         |   ✅  | API   |

#### 搜索与数据 API

| 文档加载器                                                                                    | 描述                                     | 网络支持 | 包/API |
| ---------------------------------------------------------------------------------------- | -------------------------------------- | :--: | ----- |
| [`SearchAPI`](/oss/javascript/integrations/document_loaders/web_loaders/searchapi)       | 从 SearchAPI 加载网络搜索结果（Google、YouTube 等） |   ✅  | API   |
| [`SerpApi`](/oss/javascript/integrations/document_loaders/web_loaders/serpapi)           | 从 SerpApi 加载网络搜索结果                     |   ✅  | API   |
| [Apify Dataset](/oss/javascript/integrations/document_loaders/web_loaders/apify_dataset) | 从 Apify 平台加载抓取的数据                      |   ✅  | API   |

#### 音频与视频

| 文档加载器                                                                                                    | 描述                             | 网络支持 | 包/API |
| -------------------------------------------------------------------------------------------------------- | ------------------------------ | :--: | ----- |
| [`YouTube`](/oss/javascript/integrations/document_loaders/web_loaders/youtube)                           | 加载 YouTube 视频字幕                |   ✅  | 包     |
| [`AssemblyAI`](/oss/javascript/integrations/document_loaders/web_loaders/assemblyai_audio_transcription) | 使用 AssemblyAI API 转录音频和视频文件    |   ✅  | API   |
| [`Soniox`](/oss/javascript/integrations/document_loaders/web_loaders/soniox)                             | 使用 Soniox API 转录多语言音频文件，支持可选翻译 |   ✅  | API   |
| [`Sonix`](/oss/javascript/integrations/document_loaders/web_loaders/sonix_audio_transcription)           | 使用 Sonix API 转录音频文件            |   ❌  | API   |

#### 其他

| 文档加载器                                                                                                  | 描述                                      | 网络支持 | 包/API |
| ------------------------------------------------------------------------------------------------------ | --------------------------------------- | :--: | ----- |
| [`Couchbase`](/oss/javascript/integrations/document_loaders/web_loaders/couchbase)                     | 使用 SQL++ 查询从 Couchbase 数据库加载文档          |   ✅  | 包     |
| [`LangSmith`](/oss/javascript/integrations/document_loaders/web_loaders/langsmith)                     | 从 LangSmith 加载数据集和跟踪记录                  |   ✅  | API   |
| [Hacker News](/oss/javascript/integrations/document_loaders/web_loaders/hn)                            | 加载 Hacker News 主题和评论                    |   ✅  | 包     |
| [`IMSDB`](/oss/javascript/integrations/document_loaders/web_loaders/imsdb)                             | 从 Internet Movie Script Database 加载电影剧本 |   ✅  | 包     |
| [College Confidential](/oss/javascript/integrations/document_loaders/web_loaders/college_confidential) | 从 College Confidential 加载大学信息           |   ✅  | 包     |
| [Blockchain Data](/oss/javascript/integrations/document_loaders/web_loaders/sort_xyz_blockchain)       | 通过 Sort.xyz API 加载区块链数据（NFT、交易）         |   ✅  | API   |

## 所有文档加载器

<Columns cols={3}>
  <Card title="Airtable" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/airtable" arrow="true" cta="查看指南" />

  <Card title="Apify Dataset" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/apify_dataset" arrow="true" cta="查看指南" />

  <Card title="AssemblyAI Audio Transcript" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/assemblyai_audio_transcription" arrow="true" cta="查看指南" />

  <Card title="Azure Blob Storage Container" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/azure_blob_storage_container" arrow="true" cta="查看指南" />

  <Card title="Azure Blob Storage File" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/azure_blob_storage_file" arrow="true" cta="查看指南" />

  <Card title="Blockchain Data" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/sort_xyz_blockchain" arrow="true" cta="查看指南" />

  <Card title="Browserbase" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/browserbase" arrow="true" cta="查看指南" />

  <Card title="ChatGPT" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/chatgpt" arrow="true" cta="查看指南" />

  <Card title="Cheerio" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/web_cheerio" arrow="true" cta="查看指南" />

  <Card title="College Confidential" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/college_confidential" arrow="true" cta="查看指南" />

  <Card title="Confluence" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/confluence" arrow="true" cta="查看指南" />

  <Card title="Couchbase" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/couchbase" arrow="true" cta="查看指南" />

  <Card title="CSV" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/csv" arrow="true" cta="查看指南" />

  <Card title="DirectoryLoader" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/directory" arrow="true" cta="查看指南" />

  <Card title="DOCX" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/docx" arrow="true" cta="查看指南" />

  <Card title="EPUB" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/epub" arrow="true" cta="查看指南" />

  <Card title="Figma" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/figma" arrow="true" cta="查看指南" />

  <Card title="FireCrawl" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/firecrawl" arrow="true" cta="查看指南" />

  <Card title="GitHub" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/github" arrow="true" cta="查看指南" />

  <Card title="GitBook" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/gitbook" arrow="true" cta="查看指南" />

  <Card title="Google Cloud SQL for PostgreSQL" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/google_cloudsql_pg" arrow="true" cta="查看指南" />

  <Card title="Google Cloud Storage" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/google_cloud_storage" arrow="true" cta="查看指南" />

  <Card title="Hacker News" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/hn" arrow="true" cta="查看指南" />

  <Card title="IMSDB" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/imsdb" arrow="true" cta="查看指南" />

  <Card title="Jira" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/jira" arrow="true" cta="查看指南" />

  <Card title="JSON" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/json" arrow="true" cta="查看指南" />

  <Card title="JSONLines" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/jsonlines" arrow="true" cta="查看指南" />

  <Card title="LangSmith" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/langsmith" arrow="true" cta="查看指南" />

  <Card title="MultiFileLoader" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/multi_file" arrow="true" cta="查看指南" />

  <Card title="Notion API" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/notionapi" arrow="true" cta="查看指南" />

  <Card title="Notion Markdown" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/notion_markdown" arrow="true" cta="查看指南" />

  <Card title="OpenAI Whisper Audio" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/openai_whisper_audio" arrow="true" cta="查看指南" />

  <Card title="OracleDocLoader" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/oracleai" arrow="true" cta="查看指南" />

  <Card title="PDFLoader" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/pdf" arrow="true" cta="查看指南" />

  <Card title="Playwright" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/web_playwright" arrow="true" cta="查看指南" />

  <Card title="PPTX" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/pptx" arrow="true" cta="查看指南" />

  <Card title="Puppeteer" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/web_puppeteer" arrow="true" cta="查看指南" />

  <Card title="RecursiveUrlLoader" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/recursive_url_loader" arrow="true" cta="查看指南" />

  <Card title="S3" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/s3" arrow="true" cta="查看指南" />

  <Card title="SearchAPI" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/searchapi" arrow="true" cta="查看指南" />

  <Card title="SerpApi" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/serpapi" arrow="true" cta="查看指南" />

  <Card title="Sitemap" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/sitemap" arrow="true" cta="查看指南" />

  <Card title="Soniox" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/soniox" arrow="true" cta="查看指南" />

  <Card title="Sonix Audio" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/sonix_audio_transcription" arrow="true" cta="查看指南" />

  <Card title="Spider" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/spider" arrow="true" cta="查看指南" />

  <Card title="Subtitles" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/subtitles" arrow="true" cta="查看指南" />

  <Card title="Taskade" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/taskade" arrow="true" cta="查看指南" />

  <Card title="Text" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/text" arrow="true" cta="查看指南" />

  <Card title="UnstructuredLoader" icon="link" href="/oss/javascript/integrations/document_loaders/file_loaders/unstructured" arrow="true" cta="查看指南" />

  <Card title="WebPDFLoader" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/pdf" arrow="true" cta="查看指南" />

  <Card title="YouTube" icon="link" href="/oss/javascript/integrations/document_loaders/web_loaders/youtube" arrow="true" cta="查看指南" />
</Columns>

***

<div className="source-links">
  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/i18n\zh-CN\oss\javascript\integrations\document_loaders\index.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>

  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>
</div>
