|
The Wizard of Poz posted:One of the major potential benefits I can think of that using TypeScript might offer, would be an IDE being capable of using decent Intellisense and 'instant' error checking (so if you type the wrong property name, you find out right away thanks to a red line or something). Is there an IDE that supports this? Please keep in mind I've never actually tried TypeScript. for a light-weight IDE (a bit similar to sublimetext) this is exactly what the new visual studio code IDE does. intellisense in vs code with typescript is really good and cool. all you have to do when working with some library or framework is import the corresponding type definition, and there are hundreds upon hundreds of them, imported with easy command line args, specially easy in vs code when i can just right click and open a path in console. i started using ConEmu instead of the default windows cmd window, because it started getting a bit annoying when you need multiple on at once. on the typescript website you can download the latest version for vs 2013 and 2015. i found this little project and it's teaching me a lot, and though it's angular/c# specific, it's typescript, and i'm doing my own thing from it.
|
#
?
Sep 3, 2015 11:44
|
|
|
|
| # ? Apr 24, 2024 14:41 |
|
|
I think WebStorm has type completion and stuff for TypeScript, and it's a pretty dope web IDE otherwise.
|
|
#
?
Sep 3, 2015 12:50
|
|
|
Sublime or vim and a lint/hint plugin are probably still my favorites - no idea how they play with typescript though. That example from before doesn't make sense like someone pointed out. It's setting variables and overwriting them immediately. It doesn't seem to pull in 'value' from anywhere or ever set the 'number' property. A method is, for most intents and purposes, just a function that is a property of an object. Why is the distinction important? For most entry-mid developers it really isn't. "Hiding" a method on an object that looks like a plain old number is abusing the expressiveness of Javascript in my opinion because it makes it harder to understand and doesn't really make anything more elegant or performant. I've done it a few times with special use cases I can't think of off the top of my head. If you want an object with value properties and methods, just clearly name/assign it as such.
|
|
#
?
Sep 3, 2015 13:22
|
|
|
The Wizard of Poz posted:One of the major potential benefits I can think of that using TypeScript might offer, would be an IDE being capable of using decent Intellisense and 'instant' error checking (so if you type the wrong property name, you find out right away thanks to a red line or something). Is there an IDE that supports this? Please keep in mind I've never actually tried TypeScript. https://www.jetbrains.com/webstorm/help/typescript-support.html
|
|
#
?
Sep 3, 2015 13:28
|
|
|
Visual Studio Code is basically designed for TypeScript.
|
|
#
?
Sep 3, 2015 13:40
|
|
|
Blinkz0rz posted:Visual Studio Code is basically designed for TypeScript. AND it's written in TypeScript as well, so if anyone knows about issues in TypeScript, it's the people writing VS Code.
|
|
#
?
Sep 3, 2015 15:29
|
|
|
Ochowie posted:The original version doesn't make sense to me. It has an inner function that is never called so the outer function doesn't return anything. Also, randomNumber.number would be undefined in the original version wouldn't it? I know javascript is hosed up but I don't think Math.floor adds a number property to randomNumber silently. Your version makes sense except it still doesn't return anything from newGame(). What should I add to return something for newGame()? Is it missing a return statement somewhere? Also, I was stressing out cause I got less than two weeks to get this stuff in my noggin till I have to pony up another $300 to renew my Thinkful subscription. Pressure is intense! White Light fucked around with this message at 16:18 on Sep 3, 2015 |
|
#
?
Sep 3, 2015 16:09
|
|
|
Parrotine posted:What should I add to return something for newGame()? Is it missing a return statement somewhere? Probably? In the version I posted, newGame() could return randomNumber, but that still depends on who is calling newGame() and what it is doing with it. randomNumber has enough information for a UI to hook up and run through the "guess number" game, but it's not the whole application by far.
|
|
#
?
Sep 3, 2015 16:41
|
|
|
Parrotine posted:What should I add to return something for newGame()? Is it missing a return statement somewhere? what is that function supposed to return? what does the rest of the program do? my first school project and first experience programming anything was a game of simon in javascript. that was hell to figure out having no idea how any languages worked. yes, they started CS with JS.
|
|
#
?
Sep 3, 2015 17:12
|
|
|
Bruegels Fuckbooks posted:Microsoft Visual Studio 2013. I was amazed and delighted when TypeScript errors automatically popped up in the normal Build Errors section, telling me my build failed.
|
|
#
?
Sep 3, 2015 17:14
|
|
|
Opulent Ceremony posted:I was amazed and delighted when TypeScript errors automatically popped up in the normal Build Errors section, telling me my build failed. VS2013 has ok support for TypeScript , but anything beyond simple use-cases you'll probably want to use grunt or gulp to build your project, and VS2013 does not have good integration options for them. VS2015 is supposed to have significantly improved that, and also added project types which work well with TypeScript SPA, but I haven't had the chance to try them out yet. If you are just doing HTML/CSS/TS, VS Code is much more lightweight and easier to start with. If you are working with both TypeScript & C# integrated, VS2015 is your best bet. I also never found a good way to move between versions of TypeScript with VS2013. I hope VS2015 has fixed this.
|
|
#
?
Sep 3, 2015 17:35
|
|
|
pepito sanchez posted:what is that function supposed to return? what does the rest of the program do? It's just a simple number guessing game: You type in your guess on what the number is supposed to be, then the program will tell you if your guess was 'hot' or 'cold' based on how close it was to the actual number. When the user guesses the secret number, the app lets them know it, and a new game is started. Also, I believe the counter is supposed to change depending on the number of guesses you make, but i'll admit i'm not quite sure how to do that. I linked the codepen in my initial post. White Light fucked around with this message at 17:39 on Sep 3, 2015 |
|
#
?
Sep 3, 2015 17:37
|
|
|
Parrotine posted:It's just a simple number guessing game: You type in your guess on what the number is supposed to be, then the program will tell you if your guess was 'hot' or 'cold' based on how close it was to the actual number. When the user guesses the secret number, the app lets them know it, and a new game is started. Ok, well, the hot/cold part is pretty important, because you can't actually do that with any of the solutions posted so far, and there is no real point to the check() function, as it only checks if it's an exact match, and to detect the 'heat' part, you'd basically have external logic to do that, which defeats the purpose of embedding any logic inside a 'randomNumber' object. Also, again, how this ends up being bound to the UI is really important. Can't really offer any good explanations without that.
|
|
#
?
Sep 3, 2015 17:40
|
|
|
Well at least I know that what I have isn't a working solution after all. Gotta think positive!
|
|
#
?
Sep 3, 2015 17:58
|
|
|
Parrotine posted:Well at least I know that what I have isn't a working solution after all. Gotta think positive! Here is a solution that 'works' in the sense you could hook it up to a UI. You need to map your click event to call randomNumber.check(), and pass in the value from your guessing textbox. Probably want to convert it to a number first as well. The result of check() will be your yes/hot/warm/cold message, based on the absolute value of the difference between .number and value. code:
|
|
#
?
Sep 3, 2015 18:09
|
|
|
^^ i just took that and put it into something else because bored http://jsfiddle.net/bvLzyxbu/1/
|
|
#
?
Sep 3, 2015 19:40
|
|
|
pepito sanchez posted:^^ i just took that and put it into something else because bored And if you are lazy, here's yours with no typing! http://jsfiddle.net/ke9uezLn/1/
|
|
#
?
Sep 3, 2015 20:17
|
|
|
Angular question: I have this: code:code:This looks suspiciously like this problem: http://blog.brunoscopelliti.com/run-a-directive-after-the-dom-has-finished-rendering/ but that's supposed to be fixed in angular 1.2+? (I'm running 1.4.) I can probably hack my way around it (the directive calls a function on the controller so I could just use the projectId value from the controller - where it comes from in the first place), but it'd be nice to know why this is failing. edit: I now suspect it has to do something with the {{...}} in the html but I'm not sure what or how to fix it. edit: holy tables batman
|
|
#
?
Sep 4, 2015 10:22
|
|
|
Traditionally when I post a question here, I find the answer myself 5 minutes later. So in keeping with this tradition: Changed the html to this: code:Directive scope binding changed to: code:And finally in the function that is actually called to check for uniqueness the 'id' parameter has become a function so I have to get the value from it with 'id()'. Got my inspiration from http://stackoverflow.com/questions/12371159/how-to-get-evaluated-attributes-inside-a-custom-directive where the code:
|
|
#
?
Sep 4, 2015 10:49
|
|
|
I've got to build a front-end framework for doing CRUD on relational data that has a schema + metadata but where the schema will change often - forms need to dynamically adjust to show/edit properties and relationships. I've already written some C# + JS that does all this really nicely but it's tied to a specifc 'admin' app where its ok if stuff looks a bit generic, whereas what we want is a framework we can give to clients or other developers to build richer customer facing forms. I'm guessing a modern JS framework might make this way easier. My javascript is decent but my experience with modern frameworks like Angular/React etc is pretty out of date - I'm reading articles and tutorials about different frameworks and they all seem to get you to a certain point really quickly and easily and seem nice enough - but I've been burned so many times by frameworks that are a total bitch to use once you get past the tutorial/common usage stage that I'm a bit wary...would a refactoring of my own lightwieght framework be best? Is Angular still a good choice otherwise? I'm going to keep wading through tutorials and spiking stuff out but any pointers would be grealty appreciated!
|
|
#
?
Sep 4, 2015 11:57
|
|
|
toiletbrush posted:I've been burned so many times by frameworks that are a total bitch to use once you get past the tutorial/common usage stage that I'm a bit wary...would a refactoring of my own lightwieght framework be best? This is exactly why I ditched Angular, waited for awhile, and then cautiously tried React. React isn't lightweight size-wise, but there's a surprisingly small amount of methods/properties/events/etc. that need to be documented. Try it for a day or two, and then take pleasure in knowing that you pretty much understand React and there won't be any huge learning curves after that, unlike Angular. Dynamically changing schema automatically makes me think of Facebook's other technology, GraphQL, but I still haven't had a chance to dig into that and it seems relatively more complex. Rolling your own is still totally feasible, but I'd recommend at least playing with both to see if they seem like something that would save you time.
|
|
#
?
Sep 4, 2015 13:12
|
|
|
Uziel posted:I have this saved but haven't tried it: I haven't checked this thread in a while (too much Metal Gear  ) but this looks pretty great!
|
|
#
?
Sep 4, 2015 15:44
|
|
|
kloa posted:I haven't checked this thread in a while (too much Metal Gear
|
|
#
?
Sep 4, 2015 16:20
|
|
|
Lumpy posted:And if you are lazy, here's yours with no typing! Where the hell did you learn how to do that? Are you a wizard 
|
|
#
?
Sep 4, 2015 17:09
|
|
|
Anyone here doing any lightweight statistical work in Javascript? I'm looking for a library that handle simple linear regression and a few other things. http://simplestatistics.org/ is the best I've come up with so far. I thought jStat or Math.js might have what I need, but I'm kind of surprised they don't seem to have a simple linear regression.
|
|
#
?
Sep 4, 2015 18:10
|
|
|
Parrotine posted:Where the hell did you learn how to do that? Are you a wizard It's just a different type of <input> tag. Though, the slider really makes the game a little too easy.
|
|
#
?
Sep 4, 2015 19:30
|
|
|
What's going on in the world of CSS? 3-4 years ago everyone was using bootstrap, I've noticed that for every "frontend framework starter kit" everyone doesn't ship with bootstrap... did everyone get sick of it? Instead of responsive design is everyone targeting mobile? Is everyone just way better at making good looking CSS than I?
|
|
#
?
Sep 28, 2015 00:34
|
|
|
Hed posted:What's going on in the world of CSS? 3-4 years ago everyone was using bootstrap, I've noticed that for every "frontend framework starter kit" everyone doesn't ship with bootstrap... did everyone get sick of it? Instead of responsive design is everyone targeting mobile? Is everyone just way better at making good looking CSS than I? I just figured it's because every Bootstrap app looks just like Bootstrap, so people got tired of it. At least that's why I don't use it any more. Plus there are other options: Foundation, Ionic, etc.
|
|
#
?
Sep 28, 2015 02:36
|
|
|
It's very easy to make bootstrap not look like bootstrap, but of course you never notice that those sites are using bootstrap unless you're actively digging into the source.
|
|
#
?
Sep 28, 2015 03:06
|
|
|
Plorkyeran posted:It's very easy to make bootstrap not look like bootstrap, but of course you never notice that those sites are using bootstrap unless you're actively digging into the source. Yeah, and people have been learning to do this more, so of course it looks like BS is disappearing.
|
|
#
?
Sep 28, 2015 15:39
|
|
|
Yeah I would expect usage of bootstrap going up. Also, "what's going on in the world of CSS" is not a question that really applies to what is happening with bootstrap.
|
|
#
?
Sep 28, 2015 15:43
|
|
|
Plorkyeran posted:It's very easy to make bootstrap not look like bootstrap, but of course you never notice that those sites are using bootstrap unless you're actively digging into the source.
|
|
#
?
Sep 28, 2015 15:49
|
|
|
Karthe posted:What does this involve? Is it just overriding default styles? Or are people only using parts of Bootstrap like the grid system and then styling everything else like usual? Best way to use bootstrap nowadays is to download the SASS files; that way you can choose which modules you want and customize the variables that define how everything looks so you don't have to tear your hair out fighting Bootstrap's specificity.
|
|
#
?
Sep 28, 2015 16:00
|
|
|
The React side at least is busy exploring CSS modules. They are not actually specific to React and you can use them with Angular or plain JS as well. http://glenmaddern.com/articles/css-modules It works sort of similar to BEM / Suit style, but is much easier to maintain. Definitely one of the better things to happen to CSS tooling in the last year in my opinion.
|
|
#
?
Sep 28, 2015 19:11
|
|
|
I'm working on moving from browserify to webpack, and I've run into a snag. My mocha tests are a bunch of files under test, sometimes more than one level deep. This is no problem in browserify with mochify From the webpack docs, it looks like mocha-loader is what I want...but it seems like this requires a single entry point into my tests. I do not have that. I found a couple people asking the same question on mocha-loader's issue tracker with no resolution. It seems odd that I can't easily find anyone describing the solution. I mean, if you look at mocha's examples, all of them have their tests set up like mine...multiple test files without a single entry point. How do I run these tests with webpack?
|
|
#
?
Oct 1, 2015 16:51
|
|
|
You can use require.context to have a single entry file which requires every test file:JavaScript code:
|
|
#
?
Oct 2, 2015 02:19
|
|
|
That's exactly what I needed, thanks! Of course now I find out that sinon doesn't play nice with webpack. I hate modern front-end Web development.
|
|
#
?
Oct 2, 2015 05:08
|
|
|
Is there an alternative to URI.js that plays nice with requirejs? Why is parsing a URI not part of standard JavaScript 
|
|
#
?
Oct 2, 2015 23:24
|
|
|
Thermopyle posted:That's exactly what I needed, thanks! Just be like most front-end devs and don't test. Problem solved.
|
|
#
?
Oct 5, 2015 13:09
|
|
|
|
| # ? Apr 24, 2024 14:41 |
|
|
fletcher posted:Is there an alternative to URI.js that plays nice with requirejs? I've used url() with success in the past. It's not an AMD module so you need to shim it though.
|
|
#
?
Oct 6, 2015 12:37
|
|
