auto-js-ipfs

Automatically get some IPFS download/upload/pinning functionality based on your environment.

How

• Detect IPFS support in the global fetch() API (e.g. for Agregore)
• Or attempt to use the local Kubo Daemon URL that was provided
• Or Detect a locally running go-ipfs daemon
  • Either default Kubo RPC port (9090)
  • Or detect the Brave browser and it's embedded daemon
• Or use web3.storage (if an auth token is provided)
• Or use estuary (if an auth token is provided)
• Or use readonly mode (if enabled)
• Or throw an error

API

opts => {daemonURL, web3StorageToken, estuaryToken, publicGatewayURL}

These options can be used to either detect what the best supported options are, or to create an instance of this API.

All options are optional, and by default just the local node/agregore support will be used.

daemonURL is the URL of an IPFS daemon that should be used.

web3StorageToken is the Authorization token for use in the web3.storage API.

web3StorageURL is the web3.storage API server URL that should be used.

estuaryToken is the Authorization token for use in the Estuary API.

estuaryURL is the Estuary API server URL that should be used.

publicGatewayURL is the public IPFS gateway to use for loading content without a local node.

readonly is an option for whether there should be a fallback to "readonly" mode where only api.get and api.getSize will be supported. Enabled by default.

async detect(opts) => Promise<[{type, url?}]>

Detect what's supported in the current environment.

type can be one of daemon, fetch, web3.storage, estuary

url is an optional value that will point to ipfs daemon URL being used or the public gateway being used, or the backend API URL to be used.

async choose(choice) => Promise<API>

Initialize the API based on the choice selected from detect()

async create(opts) => Promise<API>

Create an API instance by auto-detecting the "best" option available.

api.get(url, {start, end, format, signal}) => AsyncIterator<ArrayBuffer>

Get data from an ipfs:// URL, can optionally chose a start and end offset for loading data.

You can also optionally supply the format you want to load the data in. Valid formats are raw for getting raw Blocks from IPFS, and car for downloading a whole tree as a CAR (Content ARchive) file.

api.getSize(url, signal) => Promise<Number>

Get info about the size of a file at a URL.

api.uploadCar(carFileAsyncIterator, signal) => Promise<[url]>

Upload a CAR file. Returns an array of root ipfs:// URLs

api.uploadFile(fileAsyncIterator, fileName?, signal) => Promise<url>

Upload a file to the backend and get back a URL. Optionally specify a file name so that it will be wrapped in a folder.

api.clear(url, signal?) => Promise

Clear a URL that was previously downloaded and it's child nodes. This works by unpinning data when in kubo, removing it from the list in web3.storage, or deleting it from cache in Agregore.