@atlaskit/chunkinator
v6.1.0
Published
Upload large files from the browser with ease
Downloads
135,206
Readme
chunkinator
Upload large files from the browser with ease
Pipeline pieces description
See examples/integrator.ts
for good example of usage.
import {chunkinator} from '@atlaskit/chunkinator';
chunkinator
is a main customizable high level component that gives you a way to upload large files.
It takes a file as an input and pushes it through Observable pipe that consist of the following parts:
slicenator
hashinator
uploadinator
processinator
As a second argument chunkinator
takes configuration options with following interface:
interface Options {
chunkSize: number;
hashingConcurrency: number;
hashingFunction?: HashingFunction;
uploadingConcurrency: number;
uploadingFunction: UploadingFunction;
processingBatchSize: number;
processingFunction?: ProcessingFunction;
}
As a third argument chunkinator
takes list of handy callbacks. Currently there is only one: onProgress
.
slicenator
This component takes a blob, chunk size and returns Observable sequential stream of blob chunks.
Options' chunkSize
is what defines chunk size
hashinator
This component takes stream of blobs and hashes each chunk. Returning stream contains blob and it's hash.
Options' hashingConcurrency
defines how many
chunks are hashed at the same time. hashingFunction
is optional hashing function. If hashingFunction
is not
provided SHA-1
will be used instead.
uploadinator
This component takes hashed blobs and uploads them if blob doesn't exists on a backend.
Options' uploadingFunction
is a callback that takes a blob and uploads it to the server.
uploadingConcurrency
controls how many of these calls happens at the same time.
processinator
With this component consumer has an opportunity to finalize already uploaded chunks in some way.
Options' processingFunction
defines a function that is called with sequential batches of uploaded chunks.
processingBatchSize
controls how many uploaded chunks are given to processing function.
onProgress callback
This callback is called with every uploaded chunk. The only argument it is called with a portion of already uploaded chunks to it's total number.