You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Each blob is identified by its unique bounds within the Hypercore, e.g. { byteOffset: 0, blockOffset: 0, blockLength: 5, byteLength: 327680 }, which makes them easy to save and retrieve:
constblobs=newHyperblobs(core)// ID is { byteOffset: 0, blockOffset: 0, blockLength: 1, byteLength: 11 }constid=awaitblobs.put(Buffer.from('hello world','utf-8'))awaitblobs.get(id)// Buffer.from('hello world', 'utf-8')
You can also get from start/end bounds within a single blob:
constblobs=newHyperblobs(core)// ID is { byteOffset: 0, blockOffset: 0, blockLength: 1, byteLength: 11 }constid=awaitblobs.put(Buffer.from('hello world','utf-8'))awaitblobs.get(id,{start: 1,length: 2})// Buffer.from('el', 'utf-8')
If the blob is large, there's a Streams interface (createReadStream and createWriteStream) too.
Installation
npm i hyperblobs
API
constHyperblobs=require('hyperblobs')
const blobs = new Hyperblobs(core, opts)
Create a new blob store wrapping a single Hypercore.
Options can include:
{
blockSize: 64KB// The block size that will be used when storing large blobs.}
const id = await blobs.put(blob, opts)
Store a new blob. If the blob is large, it will be chunked according to opts.blockSize (default 64KB).
Options can include:
{blockSize: 64KB,// The block size that will be used when storing large blobs.start: 0,// Relative offset to start within the blobend: blob.length-1,// End offset within the blob (inclusive)length: blob.length,// Number of bytes to read.
core // A custom core to write (overrides the default core)}
const content = await blobs.get(id, opts)
Return a complete blob as a Buffer.
id is the value returned by put
Options can include:
{core,// A custom core to read from (overrides the default core)wait: true,// Wait for block to be downloadedtimeout: 0// Wait at max some milliseconds (0 means no timeout)}
await blobs.clear(id, opts)
Remove a blob from the core.
opts are the same as Hypercore.clear method.
const stream = blobs.createReadStream(id, opts)
Create a Readable stream that will yield the id blob.
Options match the get options.
const stream = blobs.createWriteStream(opts)
Create a Writable stream that will save a blob.
The corresponding ID will be set on the stream at stream.id.