@bluealba/object-transform

Overview

The main purpose of this module is to perform a series of declarative transformations over an input object returning a modified output object (this library do not have side effect over its input).

Install

You can install it with npm using

npm install @bluealba/object-transform

Usage

The transform fuction exposed by this module expects an array of transformations and the object to transform. It returns a new transformed object.

import transform from "@bluealba/object-transform";

const input = {
	name: "Claudio's travel collection",
	collection: [
		{ 
			name: "The End of Eternity", 
			code: "9780345318329", 
			rating: "4.75",
			authors: [
				{ name: "Isaac Asimov" }
			]
		},
		{ 
			name: "Dune", 
			code: "9783641173081",
			authors: [
				{ name: "Frank Herbert" }
			]
		}
	]		
};

const output = transform([
	{ type: "rename", path: "collection", toPath: "books" },
	{ type: "rename", path: "books.[].code", toPath: "books.$1.codes.isbn" },
	{ type: "exclude", path: "books.[].rating" },
], input);

As expected output will now look like this

{
	name: "Claudio's travel collection",
	books: [
		{ 
			name: "The End of Eternity", 
			code: { 
				isbn: "9780345318329"
			},
			authors: [
				{ name: "Isaac Asimov" }
			]
		},
		{ 
			name: "Dune", 
			code: {
				isbn: "9783641173081"
			},
			authors: [
				{ name: "Frank Herbert" }
			]
		}
	]		
};

Transformations

This library bases on declarative transformations that can be applied to the input object. A given transformation is tested over every property of the input object. Transformations are configured to match a certain path.

A path can be expressed using an extended dot notation. Square brackets [] can be used as a wildcard to match all items inside an array. An star * can be used to match any subpath.

In the example above:

• name - will match the name property of the root object.
• *.name - will match all name nodes of both book and author.
• collection.[].name - will match only the names of the books.
• collection.[].author.[].name - will match only the author names (*.author.[].name would work as well)

Even though we plan to expand this currently only three basic types of transformations are supported

Exclude transformation

Formerly known as remove. The simplest transformation. It is configured only with a path. Any given node whose path matches this rule will be excluded from the final output. Its children won't be included as well, unless a rename transformation alters that behavior.

Example:

{ type: "exclude", path: "books.[].rating" }

Rename transformation

The rename takes a path and new path (toPath). It will simple "move" any node that matches to the newer destination.

Captured group placeholders ($1, $2...) can be used to match the values of the path wildcards ([] and *)

In the example above:

• name -> collectionName will rename the root's object name property to collectionName
• collection.[].name -> collection.$1.bookName will rename every book's name property to bookName. Notice the $1 is a placeholder for whatever value the wildcard [] matches with.
• *.name -> $1.identifier will rename both book's name property and author's name property to identifier. Here $1 is a placeholder for whatever the wildcard * matches (which might be very different subpaths!)

Example:

{ type: "rename", path: "books.[].author", toPath: "books.$1.authors" }

Map transformation

The map doesn't affect the matched node path but its value. It's configured with a transformation function fn that receives both the value and the full path and returns the new value.

{ type: "map", path: "books.[].author", fn: (value, path) => value.toUppercase() }

Known limitations

Currently we don't support cicles. We are aiming this to perform transformation in JSON-like tree structures. We might added in the future but it's not in the inmediate roadmap.