This library utilize ReadableStream
that provide in fetch
, to let you process structured data while downloading it.
This library is most usable when you want to process data like the following:
Data stream contains data units with fixed length
bace138be6953bbff4c0a97057d55691c124dc949ad38ae4d9dc31b35d4cef4cfa7416d63d56b3d17b49555aab14c5c1ac466ecec1c74c6db6dbe81518cee1c1
The above data is the combine of 4 md5 hashes, and you want to process it as soon as every 32 bytes of data became available.
Fixed length
const url = 'http://localhost:3000/data';
const streamer = new FetchStreamer(url, 32, 'utf-8');
streamer.onData((data) => {
// callback will be called every time 32 bytes of data became available
})
streamer.start();
Dynamic length
const url = 'http://localhost:3000/data';
// This time, the actual data length will be presented in every chunk's first byte.
const streamer = new FetchStreamer(url, 1, 'utf-8');
streamer.onData((data, info) => {
if (info.targetSize === 1) {
console.log(`It's header, next chunk's length should be`, data);
return +data;
} else {
console.log(`Consumable data`, data.length, data);
return 1;
}
});
streamer.start();
Checkout example folder for more details.
url: Reading url initialSize: Default chunk size textDecoderEncoding: If supplied, the data you receive in onData callback will be decoded first. See also https://developer.mozilla.org/en-US/docs/Web/API/TextDecoder
This event will be fired every time when buffer reaches targeted size.
data: Chunk data info:
{
"targetSize": 0, // this chunk's target size
"fulfilled": true, // indicates this chunk matches the length you are expect, it will be false when it's final chunk and the legnth is shorter than the setting
}
This event will be fired when there is no more data to read
info:
{
"bytesProcessed": 39,
"bytesReceived": 39,
"elapsed": 1107
}
Pause processing, this will not pause fetching.
Resume paused processing process.
MIT