compressStream

Compress files as an async generator — process each result before the next starts.

Signature

function compressStream(
  files: AsyncIterable<File | Blob> | Iterable<File | Blob>,
  options?: StreamCompressionOptions
): AsyncGenerator<CompressionResult, void, unknown>

Parameters

files

Type	AsyncIterable<File | Blob> | Iterable<File | Blob>

Sync arrays, generators, or async iterables (e.g. streaming uploads).

options

Type	StreamCompressionOptions

Extends CompressionOptions with:

Option	Type	Description
chunkSize	number	Reserved for chunked pipelines
onChunk	(chunk: Blob, index: number) => void	Called after each successful compression

Returns

AsyncGenerator<CompressionResult> — for await each result in order.

Examples

Array of files

import { compressStream } from '@pantanal/pixu'

for await (const result of compressStream(files, {
  quality: 0.8,
  maxWidth: 1600,
  onChunk: (blob, index) => {
    console.log('done', index, blob.size)
  },
})) {
  await upload(result.file)
}

Async generator source

async function* fromUrls(urls: string[]) {
  for (const url of urls) {
    const res = await fetch(url)
    yield await res.blob()
  }
}

for await (const result of compressStream(fromUrls(urls), {
  format: 'image/webp',
  quality: 0.75,
})) {
  console.log(result.compressionRatio)
}

Backpressure-friendly pipeline

async function pipeline(files: File[]) {
  const gen = compressStream(files, { quality: 0.8 })
  for await (const result of gen) {
    await slowNetworkUpload(result.file) // waits before next compress
  }
}

When to use

API	Best for
compressBatch	Parallel throughput, known file list
compressStream	Sequential processing, memory-sensitive, async sources
AdvancedBatchProcessor	Retries, pause/cancel, priority queues

Related

• compressBatch
• Memory helpers — shouldUseStreaming