node-data-transform

Transforms data in a stream. Supports append, prepend, replace, erase, and compare. In the future it will support split.

Install

npm install @voliware/node-data-transform

Why do I need it?

If you have any stream in Node.js - a file, a TCP stream, a native Node request or response object - and you need to append data, prepend data, erase data, compare data, or replace data in the stream, you need this.

What is it?

An enhanced Transform object, which is a native Nodejs stream object. Transforms are passed to the pipe() function when dealing with readable streams. When data flows through the readable stream, it also flows through the transform. The transform can actually change the data as it goes down the pipe. DataTransform is able to process the chunks perfectly, whether you receive massive chunks at a time, or literally one byte at a time.

How do I use it?

Create one or many DataTransform objects that do as many of the following actions

1. append(match, content)
2. prepend(match, content)
3. replace(match, content)
4. erase(match)
5. compare(match)

Here, match means what are we looking for in the stream, and content is what we will append, prepend, or replace it with. For erase and compare, we don't need any content. erase simply removes the data while compare emits an event.

Example

In this example, we will create just one DataTransform that will erase data, append some data with a file, prepend some data with a file, and replace some data with text. Note that the matches are named the same as the functions for clarity.

Input HTML file

<div><!-- append --><!-- prepend --></div><div><!-- replace --><!-- erase --></div>

Setup

letreadable=Fs.createReadStream("base.html");letwritable=Fs.createWriteStream("index.html");letdatatransform=newDataTransform().erase('<!-- erase -->')// erase <!-- erase --> .append('<!-- append -->',{file: "append.html"})// append with contents of a file.prepend('<!-- prepend -->',{file: "prepend.html"})// prepend with contents of a file.replace('<!-- replace -->',{string: "Replaced!"});// replace with text

The result will be

<div><!-- append --><divid="append"></div><divid="prepend"></div><!-- prepend --></div><div>
Replaced!
</div>

What can I use it for?

A minifier

letdatatransform=newDataTransform().erase(' ')// erase spaces.erase(Buffer.from([0x0d,0x0a]))// erase new lines 

An HTML injector

letdatatransform=newDataTransform().append('<!-- templates -->',{file: "templates.html"})// append with contents of a file

Search and replace in a file

letdatatransform=newDataTransform().replace('youre',{string: "you're"})// search and replace 

Options

There are two options when creating a DataTransform.

1. concat
   • If true [default], all chunks are concatenated before processing. This is usually what you want.
   • If false, chunks are processed and sent downstream as they arrive.
2. modifiers
   • Instead of using the API you can pass along an array of modifiers to the constructor.
   • Each modifier object has the properties of
     • action - "append", "prepend", "compare", "replace", or "erase"
     • match - the string or buffer to match against
     • content - the string or buffer or filepath to append, prepend, or replace with

Example 1

Append " Senior" each time we find "Joe".

Prepend "Mr. " each time we find "Joe".

Replace "dog" with "cat".

Erase all "bad words"!

letdatatransform=newDataTransform({concat: false,modifiers: [newModifier("append","Joe",{string: " Senior"}),newModifier("prepend","Joe",{string: " Mr."}),newModifier("replace","dog",{string: "cat"}),newModifier("erase",{string: "bad words"})]});

Example 2

Even better, you can use an array to, for example, append many files or strings after one match.

letdatatransform=newDataTransform({concat: false,modifiers: [newModifier("append","<!--templates-->",[{file: './src/html/template1.html'},{file: './src/html/template2.html'},{string: '<template id="abc">ABC</template>'}])]});