olivernn / augment.js Goto Github PK

As described here Mozila doc, there's a method string.includes to check whether a string contains searchString or not.

if (!String.prototype.includes) {
  String.prototype.includes = function(search, start) {
    'use strict';
    if (typeof start !== 'number') {
      start = 0;
    }

    if (start + search.length > this.length) {
      return false;
    } else {
      return this.indexOf(search, start) !== -1;
    }
  };
}

We should also add this in augment.js

https://developer.mozilla.org/en/ECMAScript_DontEnum_attribute

Documented there. See es5-shim for fix.

I love the intention of the library but was wondering if there was a good reason not to include the Object.create method as implemented here:

https://developer.mozilla.org/en/JavaScript/Reference/Global_Objects/Object/create

Only adds 10 LOC: https://developer.mozilla.org/en/JavaScript/Reference/Global_Objects/Object/create

Or was there a reason to exclude it? There are Object extensions already, so...

bound = function () {
    return self.apply( nop.prototype && this instanceof nop ? this : ( obj || {} ),
        args.concat( slice.call(arguments) ) );
};

http://es5.github.com/#x15.3.4.5

Set the [[Construct]] internal property of F as described in 15.3.4.5.2.

http://es5.github.com/#x15.3.4.5.2

Return the result of calling the [[Construct]] internal method oftarget providing args as the arguments.

http://es5.github.com/#x13.2.2

Return obj.

Basically your code always returns the return value of the bound function and doesn't handle cases like

new ( (function () {
  return undefined;
}).bind(lulz, args) ) !== undefined;

Here the return value of invoking new (and [[Construct]]) on the bound function should be that the new instance is returned, not the return value of the bound function.

Again see es5-shim for implementation.

See shim :

http://forrst.com/posts/Object_getPrototypeOf_shim-eNB

It might be worthwhile to add this to augment.

The docs site is down.

https://gist.github.com/1120592

Lists a bunch of bugs that are in various implementations of ES5 shims. You may want to thoroughly check yours does not fail any.

For example

String#trim

using buggy character class \s instead of explicitly defining whitespace/line terminators

Note that \s fails in IE sometimes (because IE, rage.)

Following the argumentation and the benchmarks of http://blog.stevenlevithan.com/archives/faster-trim-javascript I suggest to use the trim implementation from there.

Extended Years not supported by Date.prototype.toISOString

AKA do we need both projects?

just curious as I don't know which project to recommend.

Thank you :)

link: https://github.com/kriskowal/es5-shim/

Right now augment.string.trim.js removes all the whitespace, instead of just the beginning and the end. I think the proper implementation should be more along these lines: https://gist.github.com/973478

Object.create is that much used now but certainly it is important to have i think - checkout: http://javascript.crockford.com/prototypal.html . Also in this bundle may be provided JSON.parse and JSON.stringify for completeness.

Regards,
Daniel

P.S. i'm not a ruby dev so it's hard for me to just send you pull request for those

It should be easier to build a stripped down subset of augment, including only the methods that you want to polyfill.