Skip to main content



Bemi plugs into TypeORM and PostgreSQL to track database changes automatically. It unlocks robust context-aware audit trails and time travel querying inside your application.

This package is a recommended TypeORM integration, enabling you to pass application-specific context when performing database changes. This can include context such as the 'where' (API endpoint, worker, etc.), 'who' (user, cron job, etc.), and 'how' behind a change, thereby enriching the information captured by Bemi.

See this example repo as an Todo app example with TypeORM that automatically tracks and contextualizes all changes.


  • PostgreSQL 14+
  • TypeORM


  1. Install the NPM package
npm install @bemi-db/typeorm
  1. Generate a TypeORM migration file to add lightweight PostgreSQL triggers for passing application context with all data changes into PostgreSQL replication log
npx bemi migration:create ./path-to-migrations-dir
  1. Run pending TypeORM migrations
npx typeorm migration:run


Add an Express middleware to pass application context with all underlying data changes within an HTTP request:

import { setContext } from "@bemi-db/typeorm";
import express, { Request } from "express";
import { AppDataSource } from "./data-source";

const app = express();

// This is where you set any information that should be stored as context with all data changes
setContext(AppDataSource, (req: Request) => ({
endpoint: req.url,
params: req.body,
userId: req.user?.id,

AppDataSource.initialize() // initialize TypeORM connection as normal

Application context:

  • Is bound to the current asynchronous runtime execution context, for example, an HTTP request.
  • Is used only with INSERT, UPDATE, DELETE SQL queries performed via TypeORM. Otherwise, it is a no-op.
  • Is passed directly into PG Write-Ahead Log with data changes without affecting the structure of the database and SQL queries.

Database connection

Connect your PostgreSQL source database on to start ingesting and storing all data changes stitched together with application-specific context. The database connection details can be securely configured through the dashboard UI in a few seconds.


Once your destination PostgreSQL database has been fully provisioned, you'll see a "Connected" status. You can now test the connection after making database changes in your connected source database:

psql postgres://[USERNAME]:[PASSWORD]@[HOSTNAME]:5432/[DATABASE] -c 'SELECT "primary_key", "table", "operation", "before", "after", "context", "committed_at" FROM changes;'

primary_key | table | operation | before | after | context | committed_at
26 | todo | CREATE | {} | {"id": 26, "task": "Sleep", "isCompleted": false} | {"userId": 187234, "endpoint": "/todo", "params": {"task": "Sleep", "isCompleted": false}} | 2023-12-11 17:09:09+00
27 | todo | CREATE | {} | {"id": 27, "task": "Eat", "isCompleted": false} | {"userId": 187234, "endpoint": "/todo", "params": {"task": "Eat", "isCompleted": false}} | 2023-12-11 17:09:11+00
28 | todo | CREATE | {} | {"id": 28, "task": "Repeat", "isCompleted": false} | {"userId": 187234, "endpoint": "/todo", "params": {"task": "Repeat", "isCompleted": false}} | 2023-12-11 17:09:13+00
26 | todo | UPDATE | {"id": 26, "task": "Sleep", "isCompleted": false} | {"id": 26, "task": "Sleep", "isCompleted": true} | {"userId": 187234, "endpoint": "/todo/complete", "params": {"id": 26}} | 2023-12-11 17:09:15+00
27 | todo | DELETE | {"id": 27, "task": "Eat", "isCompleted": false} | {} | {"userId": 187234, "endpoint": "/todo/27", "params": {"id": 27}} | 2023-12-11 17:09:18+00

See Destination Database for more details.

Data change querying

Lastly, connect directly to the Bemi PostgreSQL database to easily query change data from your application.

To query the read-only historical data, add the Bemi destination database to TypeORM using multiple data source. Configuration setting are found directly on the Bemi dashboard:

import { DataSource } from "typeorm";
import { Change } from "@bemi-db/typeorm";

export const AppDataSource = new DataSource({
type: "postgres",
host: "localhost",
port: 3306,
username: "root",
password: "admin",
database: "db1",
entities: [__dirname + "/entity/*{.js,.ts}"],
synchronize: true,

export const BemiDataSource = new DataSource({
type: "postgres",
name: "bemiRead",
host: "",
port: 5432,
username: "u_9adb30103a55",
password: "password",
database: "db_9adb30103a55",
synchronize: false,
logging: true,
entities: [Change],
migrations: [],
ssl: { rejectUnauthorized: false },

Initialize the BemiDataSource the same place you would your main AppDataSource

.then(() => {
console.log("Connected to Bemi");
.catch((error) => {

Querying Changes:

import { Change } from "@bemi-db/typeorm";
import { BemiDataSource } from "./index";

const changeRepository = BemiDataSource.getRepository(Change);

const [changes, changesCount] = await changeRepository.findAndCount();
console.log("All changes: ", changes);
console.log("changes count: ", changesCount);


Distributed under the terms of the LGPL-3.0. If you need to modify and distribute the code, please release it to contribute back to the open-source community.