A highly-efficient, simple (no frills) object pool.

deePool (aka "deep-pool") is an object pool with efficiency (speed, low memory, little-to-no GC) as its primary design goal.

As such, there are no configuration options to tweak behavior. It does one thing, and one thing well. If you want to re-configure the behavior, modify the code. :)

Also, this library doesn't really try to do much in the way of graceful handling of errors or mistakes, as that stuff just slows it down. You should be careful in how you use deePool.

There's only one method on the main module API:

deePool.create( objectFactory )

• create(..): produces a new pool instance. You can create as many separate pools as you need. You can even create a pool whose objects are themselves pools.

  objectFactory must be a function that produces a single new empty object (or array, etc) that you want available in the pool. Examples:

  var myArrays = deePool.create( function makeArray(){
      return [
          [ "foo", "bar", "baz" ],
          [ 1, 2, 3 ]
      ];
  } );
  
  var myWidgets = deePool.create( function makeWidgets(){
      return new SomeCoolWidget(1,2,3);
  } );

Each pool instance has four simple methods on its API:

pool.use()

pool.recycle( obj )

pool.grow( additionalSize [ = currentSize ] )

pool.size()

• pool.use(): retrieves an available object instance from the pool. Example:

  var arr = myArrays.use();

  Note: If the pool doesn't have any free instances, it will automatically grow (double in size, or set initially to 5 if currently empty) and then return one of the new instances.

• pool.recycle(..): inserts an object instance back into the pool for later reuse. Example:

  myArrays.recycle( arr );

  Tips:

  • Don't forget to recycle(..) object instances after you're done using them. They are not automatically recycled; if you don't recycle, the pool will run out of available instances and keep growing unboundedly.
  • Don't recycle(..) an object instance until you're fully done with it. As objects are held by reference in JS, if you hold onto a reference and modify an already recycled object, you will cause difficult to track down bugs in your application! To avoid this pitfall, unset your reference(s) to a pooled object immediately after recycle(..)ing it.
  • Don't recycle(..) objects that weren't created for the pool and extracted by a previous use().
  • Don't recycle(..) an object more than once. This will end up creating multiple references in the pool to the same object, which will cause difficult to track down bugs in your application! To avoid this pitfall, unset your reference(s) to a pooled object immediately after recycle(..).
  • Don't create references between object instances in the pool, or you will cause difficult to track down bugs in your application!
  • If you need to "condition" or "reset" an object instance for its later reuse, do so before passing it into recycle(..). If a pooled object holds references to other objects, and you want that memory freed up, make sure to unset those references.

  Note: If you insert an object instance into a pool that has no empty slots (this is always a mistake, but is not disallowed!), the pool will grow in size by 1 to make room for the inserted object.

• pool.grow(..): An optional number passed will specify how many instances to add to the pool (each created by the objectFactory() function specified in the deePool.create(..) call).

  If no number is passed, the default is the current size of the pool (doubling it in size). grow(..) returns the new size of the pool. Examples:

  var myPool = deePool.create(makeArray);
  
  var arr = myPool.use();  // pool size now `5`
  
  myPool.grow( 3 );        // pool size now `8`
  myPool.grow();           // pool size now `16`
  
  var size = myPool.grow( 5 );
  size;                    // 21

  Tips:

  • A new pool starts out empty (size: 0). Always call grow(..) with a valid positive (non-zero) number to initialize the pool before using it. Otherwise, the call will have no effect; on an empty pool, this will confusingly leave the pool empty.

  • An appropriate initial size for the pool will depend on the tasks you are performing; essentially, how many objects will you need to use concurrently?

    You should profile for this with your use-case(s) to determine what the most likely maximum pool size is. You'll want a pool size that's big enough so it doesn't have to grow very often, but not so big that it wastes memory.

  • Don't grow the pool manually unless you are sure you know what you're doing. It's usually better to set the pool size initially and let it grow automatically (via use()) only as it needs to.

• pool.size(): Returns the number of overall slots in the pool (both used and unused). Example:

  var myPool = deePool.create(makeArray);
  
  myPool.size();     // 0
  
  myPool.grow(5);
  myPool.grow(10);
  myPool.grow(5);
  
  var item1 = myPool.use();
  var item2 = myPool.use();
  
  myPool.size();     // 20

To install this package from npm:

npm install deepool

And to require it in a node script:

var deePool = require("deepool");
// e.g.
//    x = deePool.create(..);

// or:
var { create } = require("deepool");
// e.g.
//    x = create(..);

As of version 3.0.0, the package is also available as an ES Module, and can be imported as so:

import deePool from "deepool";
// e.g.
//    x = deePool.create(..);

// or:
import { create } from "deePool";
// e.g.
//    x = create(..);

The distribution files come pre-built with the npm package distribution, so you shouldn't need to rebuild it under normal circumstances.

However, if you download this repository via Git:

1. The included build utility (scripts/build-core.js) builds (and minifies) the dist/* files.

2. To install the build and test dependencies, run npm install from the project root directory.

3. To manually run the build utility with npm:

   npm run build
   

4. To run the build utility directly without npm:

   node scripts/build-core.js
   

To run the performance benchmarks, you must first build the core library.

With npm, run:

npm run perfs

Or, manually:

node perfs/node-perfs.js

You can also run the performance benchmarks in your browser by opening up perfs/index.html.

Note:: For UI responsiveness reasons, the browser hosting of the test suite runs in a web worker. Due to browser security restrictions around web workers, you will need to access perfs/index.html via a web context (like http://localhost:8080/perfs/index.html, via a local web-server instance in the main project root) instead of file context (like file:///perfs/index.html).

A test suite is included in this repository, as well as the npm package distribution. The default test behavior runs the test suite using the files in src/.

1. The tests are run with QUnit.

2. You can run the tests in a browser by opening up tests/index.html.

3. To run the test utility:

   npm test
   

   Other npm test scripts:

   • npm run test:package will run the test suite as if the package had just been installed via npm. This ensures package.json:main and exports entry points are properly configured.

   • npm run test:umd will run the test suite against the dist/umd/* files instead of the src/* files.

   • npm run test:esm will run the test suite against the dist/esm/* files instead of the src/* files.

   • npm run test:all will run all four modes of the test suite.

All code and documentation are (c) 2021 Kyle Simpson and released under the MIT License. A copy of the MIT License is also included.