Math.imul

[matthewleon]

No description provided.

[paf31]

This doesn't seem to be used.

[matthewleon]

Doesn't line 27 ensure that it will be used in environments like Internet Explorer that don't have native imul?

[matthewleon]

I've fixed up some of the jshint complaints on this... But the ones regarding the bitwise operations seem strange to me. I'll try add a comment in to disable those.

[matthewleon]

Doesn't line 27 ensure that it will be used in environments like Internet Explorer that don't have native imul?

[paf31]

Sorry, somehow I missed where your emulated function was used. I don't know how well it will play with DCE though, we might need to combine them.

I'm not sure if we should bake in a polyfill or let the user choose one though, especially since I don't know the license of that code.

[hdgarrood]

I think it would be best to include this polyfill; if a better polyfill does exist we can always update this later. Not dealing with this here will result in a poor UX for devs who need to support IE, I think. The code is public domain; see https://developer.mozilla.org/en-US/docs/MDN/About#Copyrights_and_licenses:

Code samples added on or after August 20, 2010 are in the public domain. No licensing notice is necessary, but if you need one, you can use: "Any copyright is dedicated to the Public Domain. http://creativecommons.org/publicdomain/zero/1.0/".

The Math.imul page was created in 2012.

[paf31]

Okay, thanks Harry. Sounds good to me then, assuming we can check that DCE works, or modify if necessary.

[matthewleon]

Okay, I've convinced jshint that I'm not a war criminal. I compiled with -Oand the output still includes the polyfill. Is there a further check that should be performed for DCE? Including this in a project with a Main and bundling?

[paf31]

Yes, you would need to reference it from some separate entry point, I think.

[matthewleon]

I've created a branch with a Main: https://github.com/matthewleon/purescript-math/tree/imul-test-dce

main = logShow $ 3 `imul` 5

I've pasted the output of pulp browserify -O below. DCE preserves the polyfill.

(function e(t,n,r){function s(o,u){if(!n[o]){if(!t[o]){var a=typeof require=="function"&&require;if(!u&&a)return a(o,!0);if(i)return i(o,!0);var f=new Error("Cannot find module '"+o+"'");throw f.code="MODULE_NOT_FOUND",f}var l=n[o]={exports:{}};t[o][0].call(l.exports,function(e){var n=t[o][1][e];return s(n?n:e)},l,l.exports,e,t,n,r)}return n[o].exports}var i=typeof require=="function"&&require;for(var o=0;o<r.length;o++)s(r[o]);return s})({1:[function(require,module,exports){
// Generated by purs bundle 0.11.7
var PS = {};
(function(exports) {
    "use strict";

  exports.log = function (s) {
    return function () {
      console.log(s);
      return {};
    };
  };
})(PS["Control.Monad.Eff.Console"] = PS["Control.Monad.Eff.Console"] || {});
(function(exports) {
    "use strict";

  exports.showIntImpl = function (n) {
    return n.toString();
  };
})(PS["Data.Show"] = PS["Data.Show"] || {});
(function(exports) {
  // Generated by purs version 0.11.7
  "use strict";
  var $foreign = PS["Data.Show"];     
  var Show = function (show) {
      this.show = show;
  };                                                 
  var showInt = new Show($foreign.showIntImpl);
  var show = function (dict) {
      return dict.show;
  };
  exports["Show"] = Show;
  exports["show"] = show;
  exports["showInt"] = showInt;
})(PS["Data.Show"] = PS["Data.Show"] || {});
(function(exports) {
  // Generated by purs version 0.11.7
  "use strict";
  var $foreign = PS["Control.Monad.Eff.Console"];
  var Control_Monad_Eff = PS["Control.Monad.Eff"];
  var Data_Show = PS["Data.Show"];
  var Data_Unit = PS["Data.Unit"];
  var logShow = function (dictShow) {
      return function (a) {
          return $foreign.log(Data_Show.show(dictShow)(a));
      };
  };
  exports["logShow"] = logShow;
})(PS["Control.Monad.Eff.Console"] = PS["Control.Monad.Eff.Console"] || {});
(function(exports) {
    "use strict";            

  function nativeImul(a) {
    return function (b) {
      return Math.imul(a, b);
    };
  }

  // https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Math/imul
  function emulatedImul(a) {
    /*jshint bitwise: false*/
    return function (b) {
      var ah = a >>> 16 & 0xffff;
      var al = a & 0xffff;
      var bh = b >>> 16 & 0xffff;
      var bl = b & 0xffff;
      // the shift by 0 fixes the sign on the high part
      // the final |0 converts the unsigned value into a signed value
      return al * bl + (ah * bl + al * bh << 16 >>> 0) | 0;
    };
  }

  exports.imul = Math.imul ? nativeImul : emulatedImul;
})(PS["Math"] = PS["Math"] || {});
(function(exports) {
  // Generated by purs version 0.11.7
  "use strict";
  var $foreign = PS["Math"];
  exports["imul"] = $foreign.imul;
})(PS["Math"] = PS["Math"] || {});
(function(exports) {
  // Generated by purs version 0.11.7
  "use strict";
  var Control_Monad_Eff = PS["Control.Monad.Eff"];
  var Control_Monad_Eff_Console = PS["Control.Monad.Eff.Console"];
  var Data_Function = PS["Data.Function"];
  var Data_Show = PS["Data.Show"];
  var $$Math = PS["Math"];
  var Prelude = PS["Prelude"];        
  var main = Control_Monad_Eff_Console.logShow(Data_Show.showInt)($$Math.imul(3)(5));
  exports["main"] = main;
})(PS["Main"] = PS["Main"] || {});
PS["Main"].main();

},{}]},{},[1]);

[paf31]

I'm fine with this then. Anyone else have comments?

[hdgarrood]

Perhaps it should say in the docs that the imul that is exported from the Math module will use the native Math.imul if it is available or a polyfill otherwise? Looks great though 👍