Skip to main content
Compatibility: Only available on Node.js.
Oracle Database supports document-centric AI workflows by combining semantic search over unstructured content with relational queries over business data in a single system. This makes it easier to build retrieval workflows (like RAG) without splitting data and vectors across multiple databases.
Why in-database document processing? You can apply Oracle Database capabilities—security, transactions, scalability, and high availability—to the same pipeline that loads, chunks, and stores content for AI search and retrieval.
If you are just starting with Oracle Database, consider exploring the free Oracle AI Database 26ai, which provides a simple way to get set up. While working with the database, it’s often advisable to avoid using the SYSTEM user for application workloads; instead, create a dedicated user with the minimum required privileges. For background on user administration, refer to the official Oracle Database Guide.

​
Overview

​
Integration details

ClassPackageCompatibilityLocalPY support
OracleDocLoader@oracle/langchain-oracledbNode-only✅✅
OracleTextSplitter@oracle/langchain-oracledbNode-only✅✅

​
Load documents

Users have the flexibility to load documents from either the Oracle Database, a file system, or both, by appropriately configuring the loader parameters. For comprehensive details on these parameters, please consult the Oracle AI Vector Search Guide. A significant advantage of utilizing OracleDocLoader is its capability to process over 150 distinct file formats, eliminating the need for multiple loaders for different document types. For a complete list of the supported formats, please refer to the Oracle Text Supported Document Formats.

​
Setup

To use OracleDocLoader install the @oracle/langchain-oracledb helpers (with @langchain/core) and make sure the Oracle Database driver prerequisites are met. Refer to the guide for comprehensive information about the Oracle Database driver.

​
Credentials

Set environment variables (or use another secrets manager) for the Oracle user that owns the source data. 
export ORACLE_USER=testuser
export ORACLE_PASSWORD=testuser
export ORACLE_DSN="localhost:1521/free"

​
Installation

npm install @oracle/langchain-oracledb @langchain/core

​
Instantiation

import oracledb from "oracledb";
import { OracleDocLoader } from "@oracle/langchain-oracledb";

const connection = await oracledb.getConnection({
  user: process.env.ORACLE_USER,
  password: process.env.ORACLE_PASSWORD,
  connectionString: process.env.ORACLE_DSN,
});

const loader = new OracleDocLoader(connection, {
  owner: "TESTUSER",
  tablename: "DEMO_TAB",
  colname: "DATA",
});

// Remember to close the connection (or pool) when your application shuts down.
// await connection.close();

​
Load from Oracle Database

const docs = await loader.load();
console.log(`Loaded ${docs.length} documents`);
console.log(docs[0]?.pageContent.slice(0, 120));

​
Load from files or directories

Switch the loader parameters to ingest local content. OracleDocLoader automatically handles more than 150 file formats; see the Oracle Text supported formats for the full list. 
const fileLoader = new OracleDocLoader(connection, {
  file: "/path/to/sample.pdf",
});

const directoryLoader = new OracleDocLoader(connection, {
  dir: "/path/to/documents",
});

​
Chunk documents

The documents may vary in size, ranging from small to very large. Users often prefer to chunk their documents into smaller sections to facilitate the generation of embeddings. A wide array of customization options is available for this splitting process. For comprehensive details regarding these parameters, please consult the Oracle AI Vector Search Guide. 
import { OracleTextSplitter } from "@oracle/langchain-oracledb";

const splitter = new OracleTextSplitter(connection, {
  split: "chars",
  max: 500,
  normalize: "all",
});

const chunks = await splitter.splitText(docs[0].pageContent);
console.log(`Generated ${chunks.length} chunks`);

​
Next steps

​
API reference

For detailed documentation of all OracleDocLoader and OracleTextSplitter features and configuration options head to the Oracle LangChain Oracle DB repository.
Edit this page on GitHub or file an issue.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.