No, this is not possible. ES6 modules need to be able to statically resolve their dependencies, without executing module code, so that `import` statements do work reliably. The module specifier must be a string literal.

However, the module loader of your choice should support dynamic loading of modules with variable names. You wouldn&#39;t be able to get a bound `app` identifier in your module scope however (and cannot reexport it), it typically would only be available in a callback or so.

ES6 imports are declarative and meant for static analysis. They cannot be dynamic.

Generally the expectation would be that if you had some conditional imports, it would be the responsibility of the underlying path resolution logic to decide what module to load based on some static path.

That or you&#39;d use the async module loader rather than declarative imports.

I wrote a program that has the trait `Animal` and the struct `Dog` implementing the trait. It also has a struct `AnimalHouse` storing an animal as a trait object `Box&lt;Animal&gt;`.

    
    trait Animal {
        fn speak(&amp;self);
    }
    
    struct Dog {
        name: String,
    }
    
    impl Dog {
        fn new(name: &amp;str) -&gt; Dog {
            return Dog {
                name: name.to_string(),
            };
        }
    }
    
    impl Animal for Dog {
        fn speak(&amp;self) {
            println!{&quot;{}: ruff, ruff!&quot;, self.name};
        }
    }
    
    struct AnimalHouse {
        animal: Box&lt;Animal&gt;,
    }
    
    fn main() {
        let house = AnimalHouse {
            animal: Box::new(Dog::new(&quot;Bobby&quot;)),
        };
        house.animal.speak();
    }

It returns &quot;Bobby: ruff, ruff!&quot; as expected, but if I try to clone `house` the compiler returns errors:

    fn main() {
        let house = AnimalHouse {
            animal: Box::new(Dog::new(&quot;Bobby&quot;)),
        };
        let house2 = house.clone();
        house2.animal.speak();
    }

&lt;!-- language: none --&gt;

    error[E0599]: no method named `clone` found for type `AnimalHouse` in the current scope
      --&gt; src/main.rs:31:24
       |
    23 | struct AnimalHouse {
       | ------------------ method `clone` not found for this
    ...
    31 |     let house2 = house.clone();
       |                        ^^^^^
       |
       = help: items from traits can only be used if the trait is implemented and in scope
       = note: the following trait defines an item `clone`, perhaps you need to implement it:
               candidate #1: `std::clone::Clone`

I tried to add `#[derive(Clone)]` before `struct AnimalHouse` and got another error:

&lt;!-- language: none --&gt;

    error[E0277]: the trait bound `Animal: std::clone::Clone` is not satisfied
      --&gt; src/main.rs:25:5
       |
    25 |     animal: Box&lt;Animal&gt;,
       |     ^^^^^^^^^^^^^^^^^^^ the trait `std::clone::Clone` is not implemented for `Animal`
       |
       = note: required because of the requirements on the impl of `std::clone::Clone` for `std::boxed::Box&lt;Animal&gt;`
       = note: required by `std::clone::Clone::clone`

How do I make the struct `AnimalHouse` cloneable? Is it idiomatic Rust to use a trait object actively, in general?

How to clone a struct storing a boxed trait object?

My Today extension needs to have a dynamic height based on the contents the widget is displaying. I was able to achieve this by adding a constraint on my bottom-most element: bottom layout guide&#39;s top is less than or equal to the bottom most element&#39;s bottom, with a constant of 0, priority 999, multiplier 1.

This works exactly as expected on iPhones - the widget height fits all content plus the default bottom margin is applied before the next widget is shown.

But on iPad it seems to be setting my widget&#39;s height equal to the max height Notification Center will allow a widget to be - there&#39;s a lot of space underneath my widget, it&#39;s nearly full screen.

***How can I remove that extra space?***

I know exactly what the problem is but am not sure how to solve it - see the &quot;The Problem&quot; section. First let me explain the setup:

**The Setup:**  
I set this extension&#39;s view up in a storyboard, nothing is done programmatically. The view consists of 5 elements stacked vertically, and some others horizontally. These are the auto layout constraints for that vertical line from top to bottom - where not stated priority is 1000, multiplier 1:

    UILabel: height = 35, top space to top layout guide with constant of 10
    UIButton: equal height and width to a different button (whose aspect ratio is 1:1, there is no fixed width/height), top space to label 10
    UIButton: equal height and width to same button, top space to above button 8
    UIButton: equal height and width to same button, top space to above button 8
    UIButton: equal height and width to same button, top space to above button 8, bottom space to bottom layout guide &lt;= 0 with priority 999

**The Necessary Behavior:**  

 - I need a grid of 4x4 buttons located underneath a single full-width label
 - Each button should be the exact same width and height - all perfect squares
 - No extra space underneath the last row of buttons

**The Result:**  
![enter image description here][1]

**The Expected Result:**  
![enter image description here][2]

**The Problem:**  
The aspect ratio constraints on all of the buttons end up indirectly imposing an &#39;implicit&#39; aspect ratio constraint on the height of the widget view when it is sent `systemLayoutSizeFittingSize:withHorizontalFittingPriority:verticalFittingPriority:`, where it’s passed the necessary width (724) for layout at the required priority and a height of 0 (to compress the view) at the fitting level priority. This results in a tall widget view on iPad where the view is wider to begin with. But there is no actual aspect-ratio constraint that can be removed. Essentially, because I&#39;ve applied aspect-ratio constraints to all the button, the height of the today extension is dependent on its width (considering all the constraints together, aspect-ratio and otherwise). Thus the height of the extension is ungainly in a wide area, manifesting itself on a wide device like the iPad. The constraints on the buttons need to be reconsidered, or somehow adjusted.

**The Sample Project:**  
A [sample project that demonstrates the issue][3] is available from CloudApp so you can download and play with it.

**What I&#39;ve Tried:**  
I tried removing the default margin insets by overriding `widgetMarginInsetsForProposedMarginInsets` and returning `0` for the bottom. This did remove the default padding, therefore decreased the height a bit, but there&#39;s still a lot of extra space underneath it.

The `UILabel` had a constraint on it: label&#39;s leading equal to superview&#39;s leading - constant 0, priority 1000, multiplier 1. If I simply change that to the superview&#39;s leading **margin**, the extra bottom spacing magically disappears. I wondered if it was because the elements were becoming too large, so increasing the amount of left spacing would decrease their size, but I tried keeping it set to the regular leading and increasing the constant and that didn&#39;t resolve the issue. But this only resolves the issue for iPad in portrait. And it doesn&#39;t even fully resolve it, every time you pull down Notification Center it starts at the large height then shrinks down to the proper size. In landscape it never does shrink to the proper size.

**Attempted Solutions:**

 - @Lefteris proposed hardcoding the size, which won&#39;t work in this case
   as the height is dynamic and the widget is available for many screen
   sizes/orientations.
 - @Yuyutsu attempted to solve it, but it unfortunately doesn&#39;t meet the requirements and exhibits conflicting constraints and a modified layout.


  [1]: https://i.stack.imgur.com/De0Pd.png
  [2]: https://i.stack.imgur.com/KooVk.png
  [3]: https://dzwonsemrish7.cloudfront.net/items/1X2O3b053x0a2D2L1c2S/TestWidgetHeight.zip

Today Extension height for iPad much larger than specified

This works:

    import app from &#39;./../app.js&#39;;

But this doesn&#39;t:

    import app from path.join(process.cwd(), &#39;app&#39;);

I&#39;m Getting:

    SyntaxError: /path/file.js: Unexpected token (5:16)
    &gt; 5 | import app from path.join(process.cwd(), &#39;app&#39;);
        |                 ^


It is possible (and/or how) to use &quot;dynamic&quot; paths? (not hardcoding the path or rely in relative paths).

Importing modules using ES6 syntax and dynamic path

This works:
<pre><code class="hljs language-javascript">import app from './../app.js';
</code></pre>
But this doesn't:
<pre><code class="hljs language-arduino">import app from path.join(process.cwd(), 'app');
</code></pre>
I'm Getting:
<pre><code class="hljs language-javascript">SyntaxError: /path/file.js: Unexpected token (5:16)
> 5 | import app from path.join(process.cwd(), 'app');
 | ^
</code></pre>
It is possible (and/or how) to use "dynamic" paths? (not hardcoding the path or rely in relative paths).

I have some code:

**baseTypes.ts**

    export namespace Living.Things {
      export class Animal {
        move() { /* ... */ }
      }
      export class Plant {
        photosynthesize() { /* ... */ }
      }
    }

**dog.ts**

    import b = require(&#39;./baseTypes&#39;);
    
    export namespace Living.Things {
      // Error, can&#39;t find name &#39;Animal&#39;, ??
      export class Dog extends Animal {
        woof() { }
      }
    }

**tree.ts**

    // Error, can&#39;t use the same name twice, ??
    import b = require(&#39;./baseTypes&#39;);
    import b = require(&#39;./dogs&#39;);
    
    namespace Living.Things {
      // Why do I have to write b.Living.Things.Plant instead of b.Plant??
      class Tree extends b.Living.Things.Plant {
    
      }
    }

This is all very confusing. I want to have a bunch of external modules all contribute types to the same namespace, `Living.Things`. It seems that this doesn&#39;t work at all -- I can&#39;t see `Animal` in `dogs.ts`. I have to write the full namespace name `b.Living.Things.Plant` in `tree.ts`. It doesn&#39;t work to combine multiple objects in the same namespace across file. How do I do this?

How do I use namespaces with TypeScript external modules?

I have an array of Promises that I&#39;m resolving with `Promise.all(arrayOfPromises);`

I go on to continue the promise chain. Looks something like this

    existingPromiseChain = existingPromiseChain.then(function() {
      var arrayOfPromises = state.routes.map(function(route){
        return route.handler.promiseHandler();
      });
      return Promise.all(arrayOfPromises)
    });

    existingPromiseChain = existingPromiseChain.then(function(arrayResolved) {
      // do stuff with my array of resolved promises, eventually ending with a res.send();
    });

I want to add a catch statement to handle an individual promise in case it errors, but when I try, `Promise.all` returns the first error it finds (disregards the rest), and then I can&#39;t get the data from the rest of the promises in the array (that didn&#39;t error).

I&#39;ve tried doing something like ..

    existingPromiseChain = existingPromiseChain.then(function() {
          var arrayOfPromises = state.routes.map(function(route){
            return route.handler.promiseHandler()
              .then(function(data) {
                 return data;
              })
              .catch(function(err) {
                 return err
              });
          });
          return Promise.all(arrayOfPromises)
        });
    
    existingPromiseChain = existingPromiseChain.then(function(arrayResolved) {
          // do stuff with my array of resolved promises, eventually ending with a res.send();
    });

But that doesn&#39;t resolve.

Thanks!

--

Edit:

What the answers below said were completely true, the code was breaking due to other reasons. In case anyone is interested, this is the solution I ended up with ...

Node Express Server Chain

    serverSidePromiseChain
        .then(function(AppRouter) {
            var arrayOfPromises = state.routes.map(function(route) {
                return route.async();
            });
            Promise.all(arrayOfPromises)
                .catch(function(err) {
                    // log that I have an error, return the entire array;
                    console.log(&#39;A promise failed to resolve&#39;, err);
                    return arrayOfPromises;
                })
                .then(function(arrayOfPromises) {
                    // full array of resolved promises;
                })
        };


---
API Call (route.async call)

    return async()
        .then(function(result) {
            // dispatch a success
            return result;
        })
        .catch(function(err) {
            // dispatch a failure and throw error
            throw err;
        });


Putting the `.catch` for `Promise.all` before the `.then` seems to have served the purpose of catching any errors from the original promises, but then returning the entire array to the next `.then`

Thanks! 

Handling errors in Promise.all

This is weird.  This is what happens at the JavaScript console in Chrome (version 42.0.2311.135, 64-bit).

&lt;!-- language: lang-none --&gt;

    &gt; 0
    &lt; 0
    &gt; 00
    &lt; 0
    &gt; 0.0
    &lt; 0
    &gt; 00.0
    X Uncaught &gt; SyntaxError: Unexpected number

Firefox 37.0.2 does the same, although its error message is:

&lt;!-- language: lang-none --&gt;

    SyntaxError: missing ; before statement

There&#39;s probably some technical explanation regarding the way JavaScript parses numbers, and perhaps it can only happen when tinkering at the console prompt, but it still seems wrong.

Why does it do that?


Why does 00.0 cause a syntax error?

I am playing around with the ReactJS framework on [JSBin](https://jsbin.com/).

I have noticed that if my component name starts with a lowercase letter it does not work.

For instance the following does not render:


    var fml = React.createClass({
      render: function () {
        return &lt;a href=&#39;google.com&#39;&gt;Go&lt;/a&gt;
      }
    });
    
    React.render(&lt;fml /&gt;, document.body);



But as soon as I replace the `fml` with `Fml` it does render. 

Is there a reason I cannot begin tags with small letters?

ReactJS component names must begin with capital letters?

In AngularJs `ng-enabled` directive is not provided. Is there any proper reason to not providing that directive in the framework, because we are having both `ng-show` and `ng-hide` when you can just use `ng-hide` to achieve our goal.

It wouldn&#39;t be nice just to check 
    **`ng-enabled=&quot;attribute.value === true&quot;`**

instead of
    **`ng-disabled=&quot;!(attribute.value === true)&quot;`** 

it will increase the readability of the code.

In angularjs we are having ng-disabled directive, why ng-enabled directive is not provided by the framework as we are having ng-show and ng-hide

I&#39;m attempting to load an RSA private key into my nodejs application using environment variables, but the newlines seem to be being auto-escaped.

For the following, assume that the ```PRIVATE_KEY``` env var is set to the following (not my actual key):

    PRIVATE_KEY=&quot;-----BEGIN RSA PRIVATE KEY-----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\n37sJ5QsW+sJyoNde3xH8vdXhzU7eT82D6X/scw9RZz+/6rCJ4p0=\n-----END RSA PRIVATE KEY-----&quot;

If I call ```console.log``` directly with the preceding string, I get the following as output:

    -----BEGIN RSA PRIVATE KEY-----
    MIICXAIBAAKBgQCqGKukO1De7zhZj6+H0qtjTkVxwTCpvKe4eCZ0FPqri0cb2JZfXJ/DgYSF6vUp
    wmJG8wVQZKjeGcjDOL5UlsuusFncCzWBQ7RKNUSesmQRMSGkVb1/3j+skZ6UtW+5u09lHNsj6tQ5
    1s1SPrCBkedbNf0Tp0GbMJDyR4e9T04ZZwIDAQABAoGAFijko56+qGyN8M0RVyaRAXz++xTqHBLh
    3tx4VgMtrQ+WEgCjhoTwo23KMBAuJGSYnRmoBZM3lMfTKevIkAidPExvYCdm5dYq3XToLkkLv5L2
    pIIVOFMDG+KESnAFV7l2c+cnzRMW0+b6f8mR1CJzZuxVLL6Q02fvLi55/mbSYxECQQDeAw6fiIQX
    GukBI4eMZZt4nscy2o12KyYner3VpoeE+Np2q+Z3pvAMd/aNzQ/W9WaI+NRfcxUJrmfPwIGm63il
    AkEAxCL5HQb2bQr4ByorcMWm/hEP2MZzROV73yF41hPsRC9m66KrheO9HPTJuo3/9s5p+sqGxOlF
    L0NDt4SkosjgGwJAFklyR1uZ/wPJjj611cdBcztlPdqoxssQGnh85BzCj/u3WqBpE2vjvyyvyI5k
    X6zk7S0ljKtt2jny2+00VsBerQJBAJGC1Mg5Oydo5NwD6BiROrPxGo2bpTbu/fhrT8ebHkTz2epl
    U9VQQSQzY1oZMVX8i1m5WUTLPz2yLJIBQVdXqhMCQBGoiuSoSjafUhV7i1cEGpb88h5NBYZzWXGZ
    37sJ5QsW+sJyoNde3xH8vdXhzU7eT82D6X/scw9RZz+/6rCJ4p0=
    -----END RSA PRIVATE KEY-----

You can see that the newline chars are being respected. However, if I call ```console.log(process.env[&quot;PRIVATE_KEY&quot;])```, the output contains the `\n` literal instead of actual newlines:

    -----BEGIN RSA PRIVATE KEY-----\nMIICXAIBAAKBgQCqGKukO1De7zhZj6+H0qtjTkVxwTCpvKe4eCZ0FPqri0cb2JZfXJ/DgYSF6vUp\nwmJG8wVQZKjeGcjDOL5UlsuusFncCzWBQ7RKNUSesmQRMSGkVb1/3j+skZ6UtW+5u09lHNsj6tQ5\n1s1SPrCBkedbNf0Tp0GbMJDyR4e9T04ZZwIDAQABAoGAFijko56+qGyN8M0RVyaRAXz++xTqHBLh\n3tx4VgMtrQ+WEgCjhoTwo23KMBAuJGSYnRmoBZM3lMfTKevIkAidPExvYCdm5dYq3XToLkkLv5L2\npIIVOFMDG+KESnAFV7l2c+cnzRMW0+b6f8mR1CJzZuxVLL6Q02fvLi55/mbSYxECQQDeAw6fiIQX\nGukBI4eMZZt4nscy2o12KyYner3VpoeE+Np2q+Z3pvAMd/aNzQ/W9WaI+NRfcxUJrmfPwIGm63il\nAkEAxCL5HQb2bQr4ByorcMWm/hEP2MZzROV73yF41hPsRC9m66KrheO9HPTJuo3/9s5p+sqGxOlF\nL0NDt4SkosjgGwJAFklyR1uZ/wPJjj611cdBcztlPdqoxssQGnh85BzCj/u3WqBpE2vjvyyvyI5k\nX6zk7S0ljKtt2jny2+00VsBerQJBAJGC1Mg5Oydo5NwD6BiROrPxGo2bpTbu/fhrT8ebHkTz2epl\nU9VQQSQzY1oZMVX8i1m5WUTLPz2yLJIBQVdXqhMCQBGoiuSoSjafUhV7i1cEGpb88h5NBYZzWXGZ\n37sJ5QsW+sJyoNde3xH8vdXhzU7eT82D6X/scw9RZz+/6rCJ4p0=\n-----END RSA PRIVATE KEY-----

I&#39;ve been trying to find more information online about how node processes environment variables with no luck. How do I load this key via env vars while maintaining newline characters? If that&#39;s not possible, how can I restore them?

Environment variables containing newlines in Node?

I have an issue I&#39;ve not seen before with the Mongoose findByIdAndUpdate not returning the correct model in the callback.

Here&#39;s the code:

        var id = args._id;
        var updateObj = {updatedDate: Date.now()};
        _.extend(updateObj, args);

        Model.findByIdAndUpdate(id, updateObj, function(err, model) {
            if (err) {
                logger.error(modelString +&#39;:edit&#39; + modelString +&#39; - &#39; + err.message);
                self.emit(&#39;item:failure&#39;, &#39;Failed to edit &#39; + modelString);
                return;
            }
            self.emit(&#39;item:success&#39;, model);
        });

The original document in the db looks like this:

    {
        _id: 1234
        descriptors: Array[2],
        name: &#39;Test Name 1&#39;
    }

The updateObj going in looks like this:

    {
        _id: 1234
        descriptors: Array[2],
        name: &#39;Test Name 2&#39;
    }  

The model returned from the callback is identical to the original model, not the updatedObj.
If I query the db, it has been updated correctly.  It&#39;s just not being returned from the database.

This feels like a &#39;stupid-user&#39; error, but I can&#39;t see it.  Any ideas greatly appreciated. 

Mongoose findByIdAndUpdate not returning correct model

After learning about [Observables](https://github.com/jhusain/asyncgenerator), I find them quite similar to [Node.js streams][1]. Both have a mechanism of notifying the consumer whenever new data arrives, an error occurs or there is no more data (EOF).

I would love to learn about the conceptual/functional differences between the two. Thanks!


  [1]: https://nodejs.org/api/stream.html

Node.js Streams vs. Observables

I&#39;m writing a program in Node.js that (in some situations) wants to act as a simple filter: read everything from stdin (up to end of file), do some processing, write the result to stdout.

How do you do the &#39;read everything from stdin&#39; part? The closest solutions I&#39;ve found so far, seem to work either for one line at a time from the console, or else only work when stdin is a file not a pipe.

Read all text from stdin to a string

In JS it doesn&#39;t seem possible to check if an argument passed to a function is actually of the type &#39;error&#39; or an instance of Error.

For example, this is not valid:

`typeof err === &#39;error&#39;`

since there are only 6 possible types (in the form of strings):

The typeof operator returns type information as a string. There are six possible values that `typeof` returns:

&gt;  &quot;number&quot;, &quot;string&quot;, &quot;boolean&quot;, &quot;object&quot;, &quot;function&quot; and &quot;undefined&quot;.

[MSDN][1]

But what if I have a simple use case like this:

    function errorHandler(err) {
    
        if (typeof err === &#39;error&#39;) {
            throw err;
        }
        else {
            console.error(&#39;Unexpectedly, no error was passed to error handler. But here is the message:&#39;,err);
        }
    }

so what is the best way to determine if an argument is an instance of Error?

is the `instanceof` operator of any help?


  [1]: https://msdn.microsoft.com/library/259s7zc1%28v=vs.94%29.aspx &quot;MSDN JS typeof info&quot;

checking for typeof error in JS

I&#39;m trying to write a script to check a website. It&#39;s the first time I&#39;m using selenium. I&#39;m trying to run the script on a OSX system. Although I checked in /Library/Python/2.7/site-packages and selenium-2.46.0-py2.7.egg is present, when I run the script it keeps telling me that there is no selenium module to import.


This is the log that I get when I run my code:

&gt;     Traceback (most recent call last):
&gt;       File &quot;/Users/GiulioColleluori/Desktop/Class_Checker.py&quot;, line 10, in &lt;module&gt;
&gt;         from selenium import webdriver
&gt;     ImportError: No module named &#39;selenium&#39;

 

ImportError: No module named &#39;selenium&#39;

I am fairly new to angular and using it with JSON api files. TO test, I am trying to use the free github api (my names for functions are for a different json api that i will be working with later). I just wanted to see if my functions were working with `console.log()`, but i receive this error in the console.

&gt; Uncaught Error: [$injector:modulerr] Failed to instantiate module
&gt; MesaViewer due to: Error: [$injector:nomod] Module &#39;MesaViewer&#39; is not
&gt; available! You either misspelled the module name or forgot to load it.
&gt; If registering a module ensure that you specify the dependencies as
&gt; the second argument.


I have spelled `MesaViewer` the exact same in both, and dependencies are seen in the second line!

    var app = angular.module(&quot;MesaViewer&quot;);
    var MainController = function($scope, $location, $http, $routeParams) {

What did I do wrong?
Here is my plunk:  http://plnkr.co/edit/sZPaFbzbOB6AmVCLL1vq

Module is not available, misspelled or forgot to load (but I didn&#39;t)

I defined a class in a module:

    &quot;use strict&quot;;
    
    var AspectTypeModule = function() {};
    module.exports = AspectTypeModule;
    
    var AspectType = class AspectType {
        // ...    
    };
    
    module.export.AspectType = AspectType;

But I get the following error message:

    TypeError: Cannot set property &#39;AspectType&#39; of undefined
        at Object.&lt;anonymous&gt; (...\AspectType.js:30:26)
        at Module._compile (module.js:434:26)
        ....

How should I export this class and use it in another module? I have seen other SO questions, but I get other error messages when I try to implement their solutions.

How to properly export an ES6 class in Node 4?

I need to do something like:

    if (condition) {
        import something from &#39;something&#39;;
    }
    // ...
    if (something) {
        something.doStuff();
    }

The above code does not compile; it throws `SyntaxError: ... &#39;import&#39; and &#39;export&#39; may only appear at the top level`.

I tried using `System.import` as shown [here](http://www.2ality.com/2013/07/es6-modules.html#module-loader-api), but I don&#39;t know where `System` comes from. Is it an ES6 proposal that didn&#39;t end up being accepted? The link to &quot;programmatic API&quot; from that article dumps me to a [deprecated docs page](http://wiki.ecmascript.org/doku.php?id=harmony:module_loaders).



How can I conditionally import an ES6 module?

The typical way to loop `x` times in JavaScript is:

    for (var i = 0; i &lt; x; i++)
      doStuff(i);

But I don&#39;t want to use the `++` operator or have any mutable variables at all. So is there a way, in ES6, to loop `x` times another way? I love Ruby&#39;s mechanism:

    x.times do |i|
      do_stuff(i)
    end

Anything similar in JavaScript/ES6? I could kind of cheat and make my own generator:

    function* times(x) {
      for (var i = 0; i &lt; x; i++)
        yield i;
    }
    
    for (var i of times(5)) {
      console.log(i);
    }

Of course I&#39;m still using `i++`. At least it&#39;s out of sight :), but I&#39;m hoping there&#39;s a better mechanism in ES6.

Is there a mechanism to loop x times in ES6 (ECMAScript 6) without mutable variables?

I&#39;m writing a script which moves dropdown below or above input depending on height of dropdown and position of the input on the screen. Also I want to set modifier to dropdown according to its direction.
But using `setState` inside of the `componentDidUpdate` creates an infinite loop(which is obvious)

I&#39;ve found a solution in using `getDOMNode` and setting classname to dropdown directly, but i feel that there should be a better solution using React tools. Can anybody help me?

Here is a part of working code with `getDOMNode` (i 
a little bit neglected positioning logic to simplify code)

    let SearchDropdown = React.createClass({
    	componentDidUpdate(params) {
    		let el = this.getDOMNode();
    		el.classList.remove(&#39;dropDown-top&#39;);
            if(needToMoveOnTop(el)) {
                el.top = newTopValue;
                el.right = newRightValue;
                el.classList.add(&#39;dropDown-top&#39;);
            }
        },
    	render() {
            let dataFeed = this.props.dataFeed;
    	    return (
    			&lt;DropDown &gt;
    			    {dataFeed.map((data, i) =&gt; {
    					return (&lt;DropDownRow key={response.symbol} data={data}/&gt;);
    				})}
    			&lt;/DropDown&gt;
    		);
    	}
    });

and here is code with setstate (which creates an infinite loop)

    let SearchDropdown = React.createClass({
    	getInitialState() {
    		return {
    			top: false
    		};
    	},
    	componentDidUpdate(params) {
    		let el = this.getDOMNode();
    		if (this.state.top) {
               this.setState({top: false});
            }
            if(needToMoveOnTop(el)) {
                el.top = newTopValue;
                el.right = newRightValue;
                if (!this.state.top) {
                  this.setState({top: true});
               }
            }
        },
    	render() {
            let dataFeed = this.props.dataFeed;
            let class = cx({&#39;dropDown-top&#39; : this.state.top});
    	    return (
    			&lt;DropDown className={class} &gt;
    			    {dataFeed.map((data, i) =&gt; {
    					return (&lt;DropDownRow key={response.symbol} data={data}/&gt;);
    				})}
    			&lt;/DropDown&gt;
    		);
    	}
    });


setState() inside of componentDidUpdate()

I stumbled upon a code example which was using this comparison:

    var someVar = 0;
    Object.is(false, someVar); //Returns false 

I know `false == 0` will be `true` that&#39;s why we have `===`.

How is `Object.is` different from `===`? 

Object.is vs ===

The WeakSet is supposed to store elements by weak reference. That is, if an object is not referenced by anything else, it should be cleaned from the WeakSet.

I have written the following test:

    var weakset = new WeakSet(),
        numbers = [1, 2, 3];
    
    weakset.add(numbers);
    weakset.add({name: &quot;Charlie&quot;});
    
    console.log(weakset);
    
    numbers = undefined;
    
    console.log(weakset);
Even though my `[1, 2, 3]` array is not referenced by anything, it&#39;s not being removed from the WeakSet. The console prints:

    WeakSet {[1, 2, 3], Object {name: &quot;Charlie&quot;}}
    WeakSet {[1, 2, 3], Object {name: &quot;Charlie&quot;}}

Why is that?

Plus, I have one more question. What is the point of adding objects to WeakSets directly, like this:

    weakset.add({name: &quot;Charlie&quot;});
Are those Traceur&#39;s glitches or am I missing something?

And finally, what is the practical use of WeakSet if we cannot even iterate through it nor get the current size?

ECMAScript 6: what is WeakSet for?

When is it important to pass `props` to `super()`, and why?

    class MyComponent extends React.Component {
      constructor(props) {
        super(); // or super(props) ?
      }
    }

Content Type	Original Author	Original Content on Stackoverflow
Question	Félix Sanz	View Question on Stackoverflow
Solution 1 - Javascript	Bergi	View Answer on Stackoverflow
Solution 2 - Javascript	loganfsmyth	View Answer on Stackoverflow

Importing modules using ES6 syntax and dynamic path

Javascript Problem Overview

Javascript Solutions

Solution 1 - Javascript

Solution 2 - Javascript

Today Extension height for iPad much larger than specified

How to clone a struct storing a boxed trait object?

Attributions