npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@dotdo/pg-protocol

v0.1.1

Published

The postgres client/server binary protocol, implemented in TypeScript

Readme

@dotdo/pg-protocol

PostgreSQL wire protocol for JavaScript.

import { serialize, Parser, messages } from '@dotdo/pg-protocol'

// Build protocol messages
const startup = serialize.startup({ user: 'postgres', database: 'mydb' })
const query = serialize.query('SELECT * FROM users')

// Parse responses
const parser = new Parser()
parser.parse(buffer, (msg) => {
  if (msg.name === 'dataRow') {
    console.log('Row:', msg.fields)
  }
})

Why @dotdo/pg-protocol?

You're building database tooling. You need:

  • Direct PostgreSQL communication (no client library overhead)
  • Custom connection pooling (your way, not theirs)
  • Protocol-level features (COPY, replication, SASL)
  • Portable code (Node.js, Bun, Cloudflare Workers, browser)

The old way: Wrap an existing client. Fight its abstractions. Hope it works in your runtime.

The pg-protocol way: Raw protocol access. Build exactly what you need. Works everywhere JavaScript runs.

What You Get

  • Full protocol support - Startup, query, extended query, COPY, auth
  • Zero dependencies - Just TypeScript
  • Streaming parser - Process messages as they arrive
  • Platform agnostic - ArrayBuffer-based, works anywhere
  • Type-safe - Full TypeScript definitions

Installation

npm install @dotdo/pg-protocol

Quick Start

Connect and Query

import { serialize, Parser } from '@dotdo/pg-protocol'

// Create TCP connection (platform-specific)
const socket = await connect('localhost', 5432)

// Send startup message
await socket.write(serialize.startup({
  user: 'postgres',
  database: 'mydb'
}))

// Handle authentication
const parser = new Parser()

socket.on('data', (buffer) => {
  parser.parse(buffer, (msg) => {
    switch (msg.name) {
      case 'authenticationOk':
        // Authenticated - send query
        socket.write(serialize.query('SELECT 1'))
        break

      case 'dataRow':
        console.log('Row:', msg.fields)
        break

      case 'readyForQuery':
        console.log('Ready for next query')
        break

      case 'error':
        console.error('Error:', msg.message)
        break
    }
  })
})

Extended Query Protocol

Prepared statements with parameters:

// Parse (prepare) the statement
await socket.write(serialize.parse({
  name: 'get_user',
  text: 'SELECT * FROM users WHERE id = $1',
  types: []  // Let PostgreSQL infer types
}))

// Bind parameters
await socket.write(serialize.bind({
  portal: '',
  statement: 'get_user',
  values: ['user-123']
}))

// Execute
await socket.write(serialize.execute({ rows: 0 }))

// Sync to get results
await socket.write(serialize.sync())

COPY Protocol

Bulk data transfer:

// Start COPY
await socket.write(serialize.query('COPY users FROM STDIN'))

// Wait for copyInResponse
parser.parse(buffer, (msg) => {
  if (msg.name === 'copyInResponse') {
    // Send data chunks
    const chunk = Buffer.from('1\tAlice\n2\tBob\n')
    socket.write(serialize.copyData(chunk.buffer))

    // Signal completion
    socket.write(serialize.copyDone())
  }
})

Message Types

Client Messages (serialize.*)

| Method | Description | |--------|-------------| | startup(opts) | Connection startup with parameters | | password(pwd) | Password authentication | | query(sql) | Simple query | | parse(opts) | Prepare statement | | bind(opts) | Bind parameters | | execute(opts) | Execute portal | | describe(opts) | Describe statement/portal | | close(opts) | Close statement/portal | | sync() | Sync after extended query | | flush() | Flush output | | end() | Terminate connection | | copyData(data) | COPY data chunk | | copyDone() | COPY complete | | copyFail(msg) | COPY failed | | cancel(pid, key) | Cancel running query |

Server Messages (via Parser)

| Message | Description | |---------|-------------| | authenticationOk | Authentication successful | | authenticationCleartextPassword | Cleartext password requested | | authenticationMD5Password | MD5 password requested | | authenticationSASL | SASL auth requested | | parameterStatus | Server parameter | | backendKeyData | Process ID and secret key | | readyForQuery | Ready for queries | | rowDescription | Column metadata | | dataRow | Row data | | commandComplete | Command finished | | error | Error response | | notice | Notice/warning | | copyInResponse | Ready for COPY data | | copyOutResponse | Sending COPY data | | copyData | COPY data chunk | | copyDone | COPY complete |

Authentication

Cleartext Password

parser.parse(buffer, (msg) => {
  if (msg.name === 'authenticationCleartextPassword') {
    socket.write(serialize.password('mysecretpassword'))
  }
})

MD5 Password

import { createHash } from 'crypto'

parser.parse(buffer, (msg) => {
  if (msg.name === 'authenticationMD5Password') {
    const salt = msg.salt  // 4 bytes
    const hash = md5(md5(password + user) + salt)
    socket.write(serialize.password('md5' + hash))
  }
})

SASL (SCRAM-SHA-256)

parser.parse(buffer, (msg) => {
  if (msg.name === 'authenticationSASL') {
    // msg.mechanisms = ['SCRAM-SHA-256']
    const clientFirst = buildClientFirst(nonce)
    socket.write(serialize.sendSASLInitialResponseMessage(
      'SCRAM-SHA-256',
      clientFirst
    ))
  }

  if (msg.name === 'authenticationSASLContinue') {
    const serverFirst = msg.data
    const clientFinal = buildClientFinal(serverFirst, password)
    socket.write(serialize.sendSCRAMClientFinalMessage(clientFinal))
  }

  if (msg.name === 'authenticationSASLFinal') {
    // Verify server signature
    const serverFinal = msg.data
    verifyServerSignature(serverFinal)
  }
})

Streaming Parser

Process large result sets efficiently:

const parser = new Parser()

// Parser handles partial messages automatically
socket.on('data', (chunk) => {
  // Chunk may contain partial messages
  // Parser buffers and emits complete messages
  parser.parse(chunk, (msg) => {
    if (msg.name === 'dataRow') {
      processRow(msg.fields)  // Process immediately, don't buffer
    }
  })
})

Platform Usage

Cloudflare Workers

import { serialize, Parser } from '@dotdo/pg-protocol'

export default {
  async fetch(request, env) {
    const socket = await env.POSTGRES.connect()

    await socket.write(serialize.startup({
      user: env.PG_USER,
      database: env.PG_DATABASE
    }))

    // ... handle auth and queries
  }
}

Node.js

import { serialize, Parser } from '@dotdo/pg-protocol'
import { Socket } from 'net'

const socket = new Socket()
socket.connect(5432, 'localhost')

socket.on('connect', () => {
  socket.write(serialize.startup({ user: 'postgres', database: 'mydb' }))
})

Browser (WebSocket proxy)

import { serialize, Parser } from '@dotdo/pg-protocol'

const ws = new WebSocket('wss://proxy.example.com/pg')
const parser = new Parser()

ws.onopen = () => {
  ws.send(serialize.startup({ user: 'postgres', database: 'mydb' }))
}

ws.onmessage = (event) => {
  parser.parse(event.data, (msg) => {
    // Handle messages
  })
}

Integration with postgres.do

pg-protocol powers the postgres.do internals:

// This is how postgres.do Durable Objects communicate

import { serialize, Parser } from '@dotdo/pg-protocol'

// Internal DO-to-DO protocol communication
const response = await doStub.fetch('https://internal/query', {
  method: 'POST',
  body: serialize.query('SELECT * FROM users')
})

const parser = new Parser()
parser.parse(await response.arrayBuffer(), handleMessage)

Performance

| Feature | Benefit | |---------|---------| | Zero-copy parsing | Minimal memory allocations | | Streaming | Process as data arrives | | ArrayBuffer | No Buffer/Uint8Array conversion | | No dependencies | Small bundle size |

Links

Related Packages

Credits

Adapted from Brian C's pg-protocol with modifications for universal JavaScript runtime support.

License

Apache-2.0