Reference — /s3

import { S3ObjectStore } from 'git-fs-s3/s3';

Runtime support

Node.jsCloudflare WorkersDenoBrowser

Constructor

classS3ObjectStore

new S3ObjectStore(options: S3ObjectStoreOptions)

Parameter	Type	Description
options.client	S3Client	A configured client — works with AWS S3, Cloudflare R2, MinIO, Backblaze B2.
options.bucket	string
options.prefix?	string	Key prefix for every object.
options.contentType?	(key: string) => string | undefined	Derives a Content-Type for uploads. Cosmetic only — git never reads it back.

Implements ObjectStore (see the guide) over any S3-compatible API, including the optional getRange.

Range reads

S3ObjectStore implements the optional ObjectStore.getRange with S3’s Range header, inclusive at both ends, returning null when the key doesn’t exist. This is what lets getObjectSize answer an object’s size without fetching the pack that holds it; a store without getRange makes that function report null, and the caller reads the object instead.

Example

import { S3Client } from '@aws-sdk/client-s3';
import { S3ObjectStore } from 'git-fs-s3/s3';

const store = new S3ObjectStore({
  client: new S3Client({
    region: 'auto',
    endpoint: process.env.R2_ENDPOINT, // omit for AWS S3
    credentials: {
      accessKeyId: process.env.R2_ACCESS_KEY_ID!,
      secretAccessKey: process.env.R2_SECRET_ACCESS_KEY!,
    },
  }),
  bucket: 'my-git-repos',
  contentType: (key) =>
    /\/(HEAD|config)$|\/refs\//.test(key) ? 'text/plain' : undefined,
});

Notes

Tip

One instance per bucket/prefix — but reuse a single S3Client across every store you create. A busy repo page can mean hundreds of object reads; a shared client pools HTTP connections instead of reconnecting per store.

@aws-sdk/client-s3 is an optional peer dependency, loaded only when you import this subpath — the root package and /http//ops never pull it in.