Alright, I got classes, feel free to ping me with commits though

Alright, have fun dude

Before committing, please check your repository as it is more than possible that it is different.

@IsmaelMiguel Since the license is included in the source file, is that why you are allowed to have the actual jQuery file in your code?

@SirPython Yeah, it is a requirement of the license.

Which I'm perfectly fine with

Ah, good. I was thinking about using jQuery for my side project, but I didn't know if I was allowed to include the source file my code.

As long as you say "This is jQuery!", you are fine

But so far, what do you think?

I'm looking over the repo now.

Take your time

What is the legacy folder's purpose?

To contain the legacy code, until deciding what to do with it

Basically: old versions

Ah.

Before the massive rewrites

I have a few commits to make to my fork.

How do I update my repo so that it is even with yours?

Great question...

nhgrif knows it, I don't

I found an article that explains how to do it with the terminal, but I'm too tired to boot up my VM now...

You can install SourceTree

It is complicated to use, but when you get the hang of it, it becomes really quick

I might actually have a git bash client on my windows side...

Try it

You don't seem to have a fork from my git

github.com/SirPython/memer

Let me investigate

@SirPython Try this: hpique.com/2013/09/updating-a-fork-directly-from-github

Hmm. It looks like what Malachi does is keep a separate repo, but he doesn't worry about your commits.

Ooh.

That link looks good.

It does look good.

That's why I sent it

It isn't hard to follow.

Ah, I have updated my code.

Nice!

It worked?

Yes; perfectly.

That's some good news

How often should I create pull requests? Like, how many commits?

It really depends

There are files that you can commit like crazy

There are others that you need to be careful

As long as you check that no one is editing it, you can do any time you see it is working fine

Hello :)

:-)

I've heard of promises and I'm aware callbacks are the way to go.

I'm actually learning JavaScript to learn node :P

However, my understanding of callbacks is I need to put a callback in a callback in a callback if I want to execute multiple functions after each other synced.

I still hope I ever get the hang of it, but most of my hope is gone by now :)

From what I've learned, the thing is that JavaScript is written in the way that you free up the thread while stuff(tm) happens. When that stuff is done, you have a callback that somehow triggers the next chain in line.

But you should try to write your code as independent of each-other as possible. Else the hard part becomes testing.

If things fuck up, you'll have hell trying to figure it out :P

Something like that, yea. If network I/O is busy but you have a CPU calculation going, it will keep the calculations going till the I/O is free so it won't block.

@Gemtastic Too late, things are already fucked :P

Exactly. Therefore you kinda need to forget about the train of functions that happens in synchronous programming (and GAS)

I'm attempting to unlearn that myself. I've gotten to the part where I kinda understand in theory how it works and what I'm supposed to do, but I don't know how to do it :P

It still feels a bit random with how the function parameters end up inside the functions in the functions @_@

Basically I get a request to my Node.JS server. The query inside that request contains data I need to write to hardware. So I send the request to the appropriate method handler and write it to the bus. Now, if the query says I need to write multiple times, I want those to be executed in order...

@Gemtastic That ^^

Greetings

I have VHDL experience, which does 200 things at a time, but all deterministic. JS isn't deterministic...

@Mast Then the questionis: Why do they have to happen in order?

And I'm a Java developer. :P

@Gemtastic I was writting a review to your question, I ended up overengeneering it.

Because my hardware expects it in order or won't properly function.

@IsmaelMiguel Nothing wrong with over engineering.

@IsmaelMiguel Less is more~

;P

Actually, it is A FREAKING LOT more

@Mast Hmm... I'm not familiar with that kinda problems :S

To the point where I even make the prompt stop when it is typing.

That is overengeneering at it's finest.

Assume you're writing FizzBuzz. You'll want 1,2,Fizz,4,Buzz, not 1,Fizz,2,Buzz,4

Nah, he wrote something cool

You could have the function push to an array and then order the array and print it

Now, my hardware expects it's commands in the right order as well. Just like that.

And you seem to have forgotten the 3

@IsmaelMiguel Fizz = 3

Oh, right, you were explaining order, not FizzBuzz itself

Haha XD

@Gemtastic Does JavaScript have variable length arrays, like Vectors?

It does

Yes. Strings also have a .length

Everything is variable

But they are from differen prototypes

It's easy to learn how JavaScript (the language) is structured

@Gemtastic Structure seems to lack IMHO.

Simple structure != lack of structure

I basically need to translate my query and afterwards 'write' the translation like write(key, value)

I kinda feel like I could write my own JavaScript

But, in order. Translate first, write after and in right order.

Well, that's where the callback comes in; if success on translate, trigger write, else handle error.

I'm all full of theory, I need more practice though XD

Let's say the translation leaves 12 write commands. How would I execute those sequentially?

Synchronous and all.

Put them all in an array and walk over that? Or could I just walk over the resulting object?

Well, then you're doing it wrong :P

I don't think I know enough to exactly tell how to do that, but the way understand it you need to drop the whole "sequential" thing. JavaScript is single-threaded so there is a linear event-chain, but the functions fire and then the thread goes on to do other things while it waits. The point is to free it up so that it can deal with firing another long-running function.

If ordering is important, you can manage that by sorting before persisting or whatever you wanna do with the result.

@IsmaelMiguel, I think we should add some general language used on SE, not necessarily memes

Feel free to write a dictionary :P

@Quill That's why I have the _ common.json

I don't even know what a TLA is!

@IsmaelMiguel is _common.json for non-memes

Yes and no

It's for anything that's common

Top notch internet

Don't worry about the double-ping

@Gemtastic I can't drop sequential if the hardware requires sequential, can I?

@IsmaelMiguel Terribly Lazy Answer

To be fair, I've never heard of hardware that requires it

@Mast Thanks a lot

@Gemtastic First I need to initialize the hardware with a command. Second I need to send the data. Third I need to tell the hardware all data is in and it should execute it.

@Gemtastic Some things may require sequential operations. For example, waiting the value of a sensor that takes around 15ms to get a value. And you need that value to continue.

Things will still happen in order of execution, but the methods may end at different times. If you fire the same function printing "hello world" will print it in a sequential order because they take the exact same amount of time to execute.

@Mast That's when callbacks are the thing.

But if you can avoid it, do :P

Sometimes it is a good thing to use calbacks, even when not needed.

And well, I'm still a JS n00b so you shouldn't really be asking me for more precise tips :P

Just so that the browser can carry on and process other important stuff.

@Gemtastic Yea, but bus controlling isn't that exact.

Exactly my point though it may be unclear :P

@Gemtastic May is the keyword

Indeed

Maybe we should s/may/was

What? o.O

Feels like we're making 'Hitchhikers' Guide to SEN', with Memer

SEN?

Stack Exchange Network

More like CRS

We have memes from other sites too

We don't use lazers here, we have napalm.

Lots of ammo for zombies

Those are very old memes

monking #2

Greetings

@DanPantry Did you go to bed again?

No, I posted monking in the 2nd monitor as well :-)

oh, I saw that now :P

Wanna give me a review on some JS code? :D

maybe it'll soothe my headache and come down from this energy drink so sure

:)

I'm new to JS to please treat me kindly ;)

Looks like the other guys have cvoered more or less what I would have said

First thing that stood out was that it isn't re-usable, but @somethinghere has already covered that

Yeah, that's kinda what happens with recursion; takes away a little of the reusability :P

Otherwise, how do you think I faired as a beginner?

I think you did OK, but you need to separate DOM access from actual business logic.

Also, github.com/CSSLint/csslint/wiki/Disallow-IDs-in-selectors

@Gemtastic You can do recursion with a queue. Just how I did.

You shouldn't use IDs in selectors (css or javascript, IMO)

at least, not in the way that you have.

I would have passed an element into the function that prints to that element

and that element could be $(#yourid) or whatever, but I don't think you should have hard-coded the element into your function

though I would prefer to see .js-my-element over #my-element

@IsmaelMiguel There's a lot you can do, my code is the "make it work" part. Step two is, Make it better! :D

@Gemtastic Step 3 is letting me over-engeneer it

@DanPantry isn't that a jQuery thing?

by the way, @IsmaelMiguel it is "analyze"... not analise :P

@IsmaelMiguel pfft :P

you wouldn't want to google 'analise'

@DanPantry Challenge accepted

@Gemtastic yes, but there is a native alternative.. node.querySelector('#yourid')

@DanPantry That kinda glared at me as well. I've been wondering if I should have edited it XD

where node is a node like document

@DanPantry Ah, the query selector thingy. That one, was it the one that only reads what's there when you create the document? (Or did I get that wrong?)

@Gemtastic it will retrieve the elements that are loaded in the DOM at time of execution, yes. however you should only be running scripts after the DOM is initialized anyway.

querySelector returns the first instance of that selector matching.

querySelectorAll will return all instances

@DanPantry I understood it as that it doesn't read in changes made by things like "selectElementById"

If you use document.querySelector('#id'), you are doing it wrong.

What is selectElementById

@IsmaelMiguel I agree, it was just the side-by-side comparison. You shouldn't be using IDs in JavaScript anyway (in my opinion)

The only reason you should use IDs (again, in my opinion) is for hash-based anchors in urls

oopd I mean "getElementById"

You shouldn't hard-code an element

Sorry, I haven't memorized all the names yet blush

@Gemtastic querySelector will return id-ed elements, its just it doesnt make sense to use querySelector when retrieving an element with an iD

querySelector mainly exists because there's no getElementWithClass

Yes there is one O.o

@DanPantry There's getElementsByClassName()

developer.mozilla.org/en-US/docs/Web/API/Document/…

hur durrr

I woke up at 6am

forgive me

Don't worry

However, I chose to use an ID here as class by the very point is supposed to use when there could be more than one

Like having many console-wannabes

The point being

Wheres I only wanted to touch that one div

the element you choose should not be chosen in the same function that does the heavy lifting.

you should have function print(element) {}

@DanPantry I feel like that every day ;P

and have the caller pass the chosen element (ID-ed or not) to the function

Yeah, that's the more OOP way of doing it

I daresay that's even DI

bbl gotta take a break in the sun ;P

that's not OOP :-)

it's just separation of concerns. stable in all dev :)

staple

@IsmaelMiguel, instead of <br> in the .md files, you add two spaces after the end of the sentence

tempted to apply to groupon

thoughts?

to?

Groupon

Don't.

Heard too many bad stories about it.

@Mast elaborate?

(or cite)

People not getting their money back, people not getting what they paid for, stuff like that.

I'd be working in the development team, not buying/selling stuff

Oh.

Well, perhaps you can fix their sh*t so the complaints go down.

:p

@DanPantry Neigh ;D I'm a horsegrammer :O

wat @Gemtastic

Stable

Where you put horses :D

I dunno I blame the sun

So, @DanPantry do you have any recommendations for assignments to try to practice doing callbacks?

@Gemtastic my recommendation for callbacks is "don't, learn promises" :p

That's actually my idea

That I'd learn callbacks so that I know them before I even sniff on promises

There's not much to learn about callbacks, frankly lol

I never learned to successfully implement promises. I did manage callbacks, albeit ugly.

@Mast you can just use Bluebird to manage the 'implementation' for you

Promises abstracts things away. I feel that I don't learn as well when I "only" learn the abstractions

Callbacks feel like a good way to really learn JavaScript

@DanPantry I'm not going to use additional libraries. I'm already neck deep in that stuff. NodeJS is what I got.

@Mast Promise is part of the ES6 spec

Bluebird just lets you convert 'node-style' callbacks into promise-based ones easily

(it's also faster than the ES6 spec)

@Gemtastic Well, as I said, there's not much to learn. Callbacks are functions that are passed as the last argument to a function that is asynchronous.

That's like making a crazy language even crazier.

The callback is invoked once the asynchronous action is completed. node-style callbacks take two arguments, (err, result).

err is null if there was no error. If err is not null then an error is expected to have been occured.

@Mast Bluebird is one of the most widely used libraries. It's that good.

Well, I still feel like I need to get some assignment to learn how to handle them properly.

Andy ou can safely ignore the extra features of Bluebird if you really don't want to use it, but I recommend you do.

@Gemtastic I'm not giving you one :p

How does that work? How does the err pass into the function?

I don't use cbs and even if I did I can't think of an example on its own

That feels a bit like magic to me

you invoke it like a normal function, @Gemtastic.

cbs?

cbs = callbacks

example async function...

function doSomethingMagic(cb) {
  setTimeout(function() {
    cb(null);
  }, 1000);
}

doSomethingMagic(function(err) { if(err) throw err; console.log('foo'); });

In promises, this looks like this.

Well, I suppose I don't fully grasp how it works with normal functions then; when you pass (err, callback) to a function, how is the err put into the err?

@DanPantry As said, extra libraries aren't an option. I think it's doable in vanilla Node.JS.

@Gemtastic you pass itl ike a normal function.... lol

function throwAnError(cb) {
  setTimeout(function() {
    cb(new Error());
  }, 1000);
}

errors are objects just like any other object in javascript

@Mast Most things are doable in NodeJS, but it's probably pretty complex

@Mast you can take that stance if you want, but it's a silly approach to take.

NodeJS has the best packagem anager inthe world, arguably - npm

and it is heavily focused on reusing packages

You are putting yourself at a disadvantage by not standing on the shoulders of giants

Besides, you need a 3rd party library if you ever want to compose asynchronous operations - except Promise.all() - so

@DanPantry I'm out of time to implement new stuff. 2 months ago I'd have agreed with you.

For example, the 'plain old node' way only used callbacks and there was no way to neatly compose those until async came along.

I feel dumb, I don't get it :/

I really need to get this stuff working and new libraries always take time to learn.

Doing a quick Memer test, don't mind me

@Mast I guarantee it will take you more time to implement your own promsie aggregation thinger than it will learning the bluebird API.

TLA

Winter Bash

@DanPantry Well, so far I don't understand both.

Duga

@Mast var timeoutAsync = Bluebird.promisify(setTimeout);

timeoutAsync(1000).then(function() { console.log('hello!'); });

or if you want to get tricky with Es6 generators

function* () { yield timeoutAsync(1000); console.log('hello!'); }

Which is perfectly unreadable JS, just like the rest of the language. I just need to execute a shitload of functions in the right order. I'm not interesting in timing, just in making sure the previous function has properly executed before starting the next.

Which seems overly complicated in JS.

It's freaking easy in C++.

that's because everything in JS is asynchronous.

Are you using promises or callbacks, @Mast?

@Mast Well, c++ is sequential and synchronous :P

@DanPantry Uh, one. When I'm getting a request which is actually meant for another server. It starts a new request, awaits it's response and than copies the result to the original client.

If you're using promises its super easy

@Gemtastic Yea, I love deterministic, sequential languages.

@DanPantry No, that's a callback.

You need a third party library for callbacks.

Or a callback pyramid (AVOID AVOID AVOID).

@Mast I can understand that, since you apparently don't work asynchronous :P

@Mast get this github.com/caolan/async

paste me the code you want to execute sequentially and I'll transform it for you.

SNR

@DanPantry I'll look up how it's currently done, it's not relying on a third party.

callbacks can only be executed sequentially with a) async or b) callback pyramid.

callback pyramids are bad.

So use async.

Your stubbornness to use a third party library is understandable, coming from C++.

But you need to let it go (THE COLD DIDN'T BOTHER ME ANYWAY) in JS.

handleGetSecondaryRequest = function(callback, res, qu) {
	var OPTIONS = {
		hostname: 'localhost', //10.33.62.68
		port: '8502',
		path: '/?' + qu
    //path: '/?GetSensors=True&GetActuators=False'
	}

	var data = null;

	http.get(OPTIONS, function(func, data) {
		func.on('data', function(chunk) {
      data += chunk;
		});
		func.on('end', function() {
			callback(res, data);
		});
	}).on('error', function(e) {
		callback(res, e);
	})
};

var handleGetSecondaryRequestCallback = function(res, receivedData)
{
	res.write(receivedData);
	res.end("END\n");
};

Which is a callback pyramid

Which is bad.

http.get - is that http from node?

ie, require('http')

Yea, it's kind of essential since it's part of a HTTP server.

ok, one second.

wait

does this not work? why dod you need it to be changed?

or is it just an example

It's one of the working parts.

You just asked if I used callbacks. Well, one, here.

Yes, but I mean

Whats the issue with this code? Like, you're struggilng to execute it in sequence

where is the bit that is calling it?

That part isn't working as intended currently. I could provide an example.

There are a lot of functions to be executed, which currently go like this:

function initI2C(o) {
  Servo.writeReg(0, 0);
  Servo.writeReg(2, 0);
  console.log("Init");
  runNextMethod(o);
}

Where o is the variable for my query.

Which is totally ugly.

Next would be

function dataI2C(o) {
  Servo.writeReg((parseInt(o.ActuatorNumber*2))-1, parseInt(o.Speed));
  Servo.writeReg((parseInt(o.ActuatorNumber*2)), parseInt(o.Direction, 10));
  console.log((o.ActuatorNumber*2)-1, o.Speed);
  console.log(o.ActuatorNumber*2, o.Direction);
  console.log("Data");
  runNextMethod(o);
}

This all gets handled by a monster:

var commandsI2C={
  'initI2C':initI2C,
  'dataI2C':dataI2C,
  'executeI2C':executeI2C,
  'stopI2C':stopI2C,
}

var nextMethodIndex=0;

var methodOrder=['initI2C', 'dataI2C', 'executeI2C']

function runNextMethod(o)
{
    console.log(o);
    console.log(typeof o);
    if(nextMethodIndex<methodOrder.length)
    {
         var func = commandsI2C[methodOrder[nextMethodIndex]];
         nextMethodIndex++;
         console.log("Next method:", nextMethodIndex);
         setTimeout(func(o), 3000)
    }

jesus

RBA

let me transform this section of code to give you an idea, OK/

@DanPantry Exactly.

SNR

what bit runs handleGetSecondaryRequest?

Why is RBA not getting flagged *sigh*

Winter Bash

Teacher's Lounge

Now, what I actually want is to generate a list of writeReg(key,data) commands which should be executed in the correct order and is generated based on the query.

@DanPantry Oh, that only runs if it's not for this server.

It's the proxy function.

Jamal Mat'sMug rolfl 200_Success SimonAndréForsburg

So I can have a client calling this server, or a client calling this server relayed by a server by a server by a server.

In theory...

@Quill That's a test guys, don't worry

@Quill Forsberg is with an E

@Mast here's a little improved version of what you already had

// callback MUST be last argument
function handleGetSecondaryRequest(res, qu, callback) {
  // You know you could pass in arguments like this:
  // { GetSensors: true, GetAcutators: false }
  // And set that as 'qs' in options, right?
  var OPTIONS = {
    hostname: 'localhost',
    port: 8502,
    path: '/?' + qu
  };

  var uri = url.format(OPTIONS);
  // Pass the error and result stream to the callback.
  httpGet(uri, callback);
}

// Helper method because the in-built http.get is weird
function httpGet(uri, callback) {

@Mast Thank you

Once you set the callback as the last argument you can use composing libraries like async

But it's also common convention to set the callback as last argument setTimeout is the exception for legacy reasons

I would really recommend you stop saying JavaScript is a bad language when the real problem is that you are just coding it wrong

:S

JavaScript is not C++ and a lot of the synchronous idioms you use in C++ just don't apply here

@DanPantry I'm not saying it's a bad language, I just have trouble getting used to it. Me hating it doesn't make it a bad language.

yield (ES6) and Promises can help with making asynchronous code appear more synchronous

@DanPantry That looks a lot better than what I got.

You can use async.waterfall to help with your composition

Somebody use something abbreviationish in chat

BTW

BTW.abc is one

@DanPantry Waterfall puts them all after each other?

@mast waterfall looks like this

async.waterfall(tasks, callback)

it will execute each task in the tasks array in order. The result of each task will be passed to the next task in the chain

if there are any errors, or the task chain finishes, then the result and /or error will be passed to callback.

Now that's worth getting a library for, exactly what I need.

async.waterfall([
  function(cb) { cb(null, 1) },
  function(previousValue, cb) { cb(null, previousValue + 1); }
], function(error, finalValue) {
  console.log(finalValue);
});

this will, for example, print 2

github.com/caolan/async#waterfall

github.com/caolan/async#waterfall

Sorry for double post

@IsmaelMiguel, I added way more memes and fixed a few bugs. There's a pull request there.

I have a major code re-haul planned for Friday afternoon, if you like I'll keep you updated with the result.

@Mast sure. :-)

@Quill I'm off for the night, though

@quill cya cyndaQUILL c;

@DanPantry marshtomp ftw

I like this guy

@Quill Burn it, with fire!

@DanPantry So you start with defining the callback and then follows the list of functions?

@Mast async.waterfall starts by defining all of the functions first, and teh last argument is the 'finished' callback as such

It's kinda hard to envision which is why I prefer promises

So promises would be even better than async?

Well that depends

it's extra work to implement promises obviously

Does it improve readability in the end?

In my opinion, yes

getFirstRequest().then(getSecondRequest).then(reactToSecondRequest)

talk in a couple hours, i need to work

Thank you for your help so far, it has been invaluable.

now that one works!

anyone want to help me get some JS working with ASP.NET

?

@Mat'sMug!!!!

I have no experience with ASP.NET and I suck at JS, but you could post it anyway.

meeting in 10 minutes

I forgot all about it too

@IsmaelMiguel I can't right click in chat now