DirectoryLoader integration | LangChain 한국어

Tip

Compatibility: Only available on Node.js.

This notebook provides a quick overview for getting started with DirectoryLoader document loaders. For detailed documentation of all DirectoryLoader features and configurations head to the API reference.

This example goes over how to load data from folders with multiple files. The second argument is a map of file extensions to loader factories. Each file will be passed to the matching loader, and the resulting documents will be concatenated together.

Example folder:

src/document_loaders/example_data/example/
├── example.json
├── example.jsonl
├── example.txt
└── example.csv

Overview

Integration details

Class	Package	Compatibility	Local	PY support
`DirectoryLoader`	langchain	Node-only	✅	✅

Setup

To access DirectoryLoader document loader you'll need to install the langchain package.

Installation

The LangChain DirectoryLoader integration lives in the langchain package:

npm

yarn

pnpm

npm install langchain @langchain/core

yarn add langchain @langchain/core

pnpm add langchain @langchain/core

Instantiation

Now we can instantiate our model object and load documents:

Warning

The @langchain/community package is no longer maintained. Examples that import from @langchain/community may be outdated or broken. Use with caution.

import { DirectoryLoader } from "@langchain/classic/document_loaders/fs/directory";
import {
  JSONLoader,
  JSONLinesLoader,
} from "@langchain/classic/document_loaders/fs/json";
import { TextLoader } from "@langchain/classic/document_loaders/fs/text";
import { CSVLoader } from "@langchain/community/document_loaders/fs/csv";

const loader = new DirectoryLoader(
  "../../../../../../examples/src/document_loaders/example_data",
  {
    ".json": (path) => new JSONLoader(path, "/texts"),
    ".jsonl": (path) => new JSONLinesLoader(path, "/html"),
    ".txt": (path) => new TextLoader(path),
    ".csv": (path) => new CSVLoader(path, "text"),
  }
);

Load

const docs = await loader.load()
// disable console.warn calls
console.warn = () => {}
docs[0]

Document {
  pageContent: 'Foo\nBar\nBaz\n\n',
  metadata: {
    source: '/Users/bracesproul/code/lang-chain-ai/langchainjs/examples/src/document_loaders/example_data/example.txt'
  },
  id: undefined
}

console.log(docs[0].metadata)

{
  source: '/Users/bracesproul/code/lang-chain-ai/langchainjs/examples/src/document_loaders/example_data/example.txt'
}

API reference

For detailed documentation of all DirectoryLoader features and configurations head to the API reference.