Lumpy posted:

Phone posting so not much detail for you but you want to use a "for in" loop, coupled with hasOWnProperty.

ufarn posted:

I have a form that outputs a formatted version of the input in plain HTML with basic jQuery, but as it's meant for sharing, the ideal would be to generate an image file. How would that work for a static website?

Wheany posted:

You have a text input that gets rendered as formatted html?

You want to output an image with words on it?

That is a server side thing, not a Javascript thing. You need to use some image output library for your server framework.

(Or if you like how your html output looks like in the browser, I guess you could use PhantomJS to render the html into an image and serve that)

ufarn posted:

I take it that would involve rasterize.js, but does that require the rendering to be on its own page, or can you fetch the specific DOM element generated by jQuery to render?

Pollyanna posted:

Can someone tell me why the hell this fiddle doesn't work? Everything seems to be created just fine, but something goes horribly wrong during rendering and I can't tell what's getting broken. I just get a grey square for my Map.

Wheany posted:

Gray square = you have not set your zoom level and center point.

https://www.youtube.com/watch?v=21PDd17dnnI

Thermopyle posted:

I feel like I'm not getting something about ES6 Promises.

Fish Ladder Theory posted:

first of all $.get returns a jquery promise so there's no reason to wrap it in a native promise that i can see.

Thermopyle posted:

I'm just thinking about how once you return a promise, you've always got to append a .then(function (result) {blahblahblah}) to use that value and its turtles all the way down.

Kobayashi posted:

Speaking of promises, it seems like all the error handling of a promise chain collapses down to a single catch(). For example, if I want to 1) start up a database and then 2) call some API if and only if the DB is ready, I'm not sure how to chain those together such that the first promise can reject in a DB-specific way and the second promise rejects in an API-specific way.

As far as I can tell, I can either create some complicated, catch-all error handler, create two separate promises that fire independently, or I can nest the second promise inside the first, leading to the same "rightward drift" that I'd get with callbacks. Am I missing some other option? Does that make any sense?

Thermopyle posted:

Suspicious Dish posted:

Yes, that's sort of the point. A Promise represents some value in the future. There's no value currently, only a value in the future.

Note that if the promise is already resolved it will call your callback with the value immediately, so there's no chance to "miss" the value gone by. That might be your misunderstanding.

RobertKerans posted:

Promise.resolve first, then chain.

Sedro posted:

It's as simple as wrapping your code in a function and immediately invoking it:

JavaScript code:

(function() {

var paste = gatherpaste();

... your code ...

}());

// paste is not accessible out here

Raskolnikov2089 posted:

So here's something I've struggled with for awhile. The whole "Global variables = death" mantra has been successfully drilled into me.

Where I'm at a loss is understanding the best way to access input if I can't store it globally.

I made a calculator that does a large part of my job, taking into account the many conflicting state, local and federal laws that we have to juggle when reviewing programs. The calculator takes pasted user input, parses it into a usable form and stores it in an array for use further down the line in the calculations.

http://jsfiddle.net/fgccLa81/

HTML code:

<input type="text" id="paste" placeholder="Paste Things Inside of Me" />

code:

function gatherpaste(){
   var pasted = $('#paste').val().split(' ');
    return pasted;
}

function doCalculations(){
    var getArray = gatherpaste();
    var newArray = [];
    for(var x=0; x<getArray.length; x++){
        if(getArray[x] < 1750){
            newArray.push(getArray[x]);
        }
        else{
            getArray[x] = "Illegal";
        }
    }
}

function doDifferentCalculations(){
    var getArray = gatherpaste();
    //etc etc
}

There's dozens of different functions that need to refer back to the array generated from gatherpaste() to perform their own discrete calculations. It seems so incredibly inefficient and wrong to have each function call gatherpaste(), querying the dom each time. But outside of declaring a global variable, i.e.

code:

 var globalGatherPaste = gatherpaste();

I don't know how else to do it?

HappyHippo posted:

Where are these functions called? When the user pushes a button? If so then gatherpaste() there and pass the result to each DoCalc function as a parameter.

Kobayashi posted:

^ I recently bookmarked keymaster.js, a dependency-free library for keyboard shortcuts. I figure that's probably easier than figuring out all the various idiosyncrasies across platforms and browsers.

MrMoo posted:

That seems a little problematic on scalability, you can capture standard console logging in Chrome or Firefox under Linux in its parent console. The biggest problem I found recently is that logging objects, e.g. console.log ("log this:", object) keeps a reference on the object until the log is wiped from memory! Incredibly dangerous implementation for long running pages.

Thermopyle posted:

I'm looking for a nice javascript logging library/framework.

The two things I'm most looking for:

1. Easily enable/disable console logging for dev/production.
2. The nicest thing would be something that integrates with some cloud logging solution like loggly/papertrail/whatever that will let me log selected events from clients.

Any recommendations here?

edit: BTW, I'm just using logentries javascript client now. Just wondering if anyone has any experience with something they like more.

Sitting Bull posted:

I'm trying to detect when a user has entered an @ symbol, within a contenteditable div, for the purpose of doing mentions. Is this possible using only keyup? I was using keypress, but this didn't allow me to check whether they had just hit enter, or backspace, etc.

The way I'm doing it is keeping a last key, and second to last key buffer. if last key was @, and second to last was whitespace, I'd allow them to mention. This is incredibly bad, though, and messes up in lots of scenarios. If its possible just using keyup, as opposed to having to gently caress around with keypress, I'd love to do it that way.

Sitting Bull posted:

I have another question WRT to this:

What is an intelligent way of determining whether or not the user has entered @ to mention somebody, after hitting backspace. Short of keeping track of the character to the left of the caret, which I don't know how to do this, since the user is entering text into a contenteditable div, not an input. The contenteditable div may have many paragraphs, containing other paragraphs, or spans.

down with slavery posted:

https://code.google.com/p/rangy/

edit. I guess https://github.com/timdown/rangy is the current URL now

Fish Ladder Theory posted:

The native Promise has valueOf() in Chrome, which appears to do the same thing, but it doesn't actually appear to be documented anywhere, and I actually can't figure out how to use it like .inspect()

code:

var a = new Promise(function(resolve, reject) { resolve(5)})
a.valueOf()
// Promise {[[PromiseStatus]]: "resolved", [[PromiseValue]]: 5}