Body parsing

Defaults

JSON (application/json and +json types) and urlencoded bodies are parsed automatically, with a 100kb limit, exactly like @nestjs/platform-express. Requests without a body, or with an unrecognized Content-Type, leave req.body undefined.

• Invalid JSON → 400 Bad Request with the parser message.
• Top-level primitives ("str", 42) are rejected in strict mode, like body-parser.
• Bodies over the limit → 413 Payload Too Large.
• Unsupported charsets → 415.

Raise the limit for the whole app:

new BunAdapter({ bodyLimit: '5mb' });

Raw body

Webhook signatures need the exact bytes:

const app = await NestFactory.create<NestBunApplication>(AppModule, new BunAdapter(), { rawBody: true });

@Post('stripe')
webhook(@RawBody() raw: Buffer, @Headers('stripe-signature') sig: string) { ... }

More parsers

app.useBodyParser('text');                       // text/plain → string
app.useBodyParser('raw', { type: 'image/*' });   // → Buffer
app.useBodyParser('json', { limit: '10mb', type: 'application/vnd.custom+json' });

Parsers run in registration order and each one only handles requests whose Content-Type matches its type.

Turning parsing off

NestFactory.create(AppModule, new BunAdapter(), { bodyParser: false });

Then read the body yourself from the Web Request:

@Post('upload')
async upload(@Req() req: BunRequest) {
  const form = await req.native.formData();
  const file = form.get('file') as File;
}

req.native is the underlying Web Request. Its json(), text(), arrayBuffer() and formData() are all available as long as no parser consumed the body first.