Streams in node.js : Readable and Writable

27 / Jan / 2014 by Sahil Chitkara 0 comments

Within Node.js development, streams are used to transfer data including requests to an HTTP server, opening a file, stdin, stdout, stderr etc.

Streams can be readable, writable or both. We can connect the readable stream to a writable stream using the pipe method.

All streams are instances of event emitters in node.js. We can get different events for data, for example  data event, error event, end event, readable event, writable event etc.

Readable stream in node.js:

The Readable stream is a source of data which means that data comes out of Readable stream. It will not emit data until you indicate that you are ready to receive it.

Reading stream can be divided into two modes of stream based on their data flow mechanism:

  1. Flowing Mode: In Flowing Mode the data is read from the system and provided to your program.

  2. Non-Flowing Mode:  In non-flowing mode you have to call stream.read() explicitly to get chunk of data

General use cases of Readable streams in nodejs:

  • http response
  • http request
  • Reading a file
  • Child process
  • Sockets tcp, etc.

Below we have mentioned different methods and how they can be helpful:

1. read:

readable.read([size])

This method is used to pull out data from source explicitly. If there is no data to transfer, it will return null. Size specifies how much data you want to read.

var readable=fs.createReadStream(‘abc.txt’);
 readable.read();

2. setEncoding:

Readable.setEncoding(encoding)

You can specify the encoding scheme of the data you want to receive.Call this function if you want to get stream in strings of specified encoding instead of buffer.

var readable=fs.createReadStream(‘abc.txt’);
 readable.setEncoding('utf8');

3. Resume and  Pause:

readable.resume() and readable.pause()

readable.pause() will pause the emitting of data event and correspondingly readable.resume() will resume emitting.

Both (.pause() and .resume())can be used only in flowing mode.

4 .Pipe:

Readable.pipe(destination,[option])

Pipe() method is used to transfer data from Readable stream to Writable stream.

Destination is a writable stream.

5 .Unpipe:

Readable.unpipe(destination,[option])

This method will remove the hook set up on previous pipe call.

If destination is not specified, it will remove hook set up from all pipes.

Let take an example to see how pipe and unpipe works:


var fs= require(‘fs’)

var readable = fs.createReadStream(‘abc.txt’);

var writable=fs.createWriteStream(‘bcd.txt’);

readable.pipe(writable);

readable.on(‘end’,function(){

readable.unpipe(writable);

});

6. Events :

Data: Event ‘data’ will give you a chunk of data in event handler. It can also be used to buffer your string.

When data event is triggered, the stream will switch to flowing mode. As soon as data starts flowing we can get data in event handler.

End: Event ‘end’ is triggered when no more data is to be transferred.

Close: Event ’close’  is triggered when the resource is closed.

Error: Event ‘error’ is triggered when error occurs while flowing of data.

Example:

var readable = createReadStream(‘abc.txt’);
readable.on('data', function(chunk) {
   console.log('got %d bytes of data', chunk.length);
});
readable.on(‘error’, function(error) {
   console.log('Error while getting data', error);
});

readable.on(‘end’, function() {
   console.log(‘no more data to read’);
});
readable.on(‘close’, function() {
console.log('resource closed’);
});

We will be discussing more about Writable and Duplex Streams in our coming blogs.
Hope, it was helpful for you.

FOUND THIS USEFUL? SHARE IT

Leave a comment -