Laravel App not always showing my changes in real time.. caching? - javascript

I am learning AngularJS and having a weird problem whereby sometimes if I make a change in my JS files, the change doesn't apply.
I can see the GET request to the file through the console however it still contains old content. I can even remove everything inside of the file and the application still shows old content. It's only when I delete the file that it recognizes that something is wrong.
Is there some sort of caching going on that I need to know about?
I am using Laravel 5.1 however at this point, Laravel is really only handling the routing at this point.
Eventually, the changes come through. Am I going crazy or is this one of the gotcha's I should know about with AngulasJS?

Try:
$scope.$apply()
The way angularjs works is when one of the binded fields (ng-model) is changed in the browser, angularjs updateds the data model in the scope (controller or directive).
If you have code with callbacks when the new data is to be assigned to the data model, you'll need to run $scope.$apply(). This statement tells anguarljs to rebind the data model to the view (html).
Depends on where you have the $apply statement, you might want to test for $scope.$$phase because if the digest/apply is already in process and you run $apply, you'll get a Javascript run time error.

Related

Custom Webpack -- Assign properties to data from Angular Service within Webpack/Write Files based on Angular Template within Webpack

My situation is that I have multiple SVG templates that have data binding on my Angular project, and I want to somehow be able to determine if the data that is going to be bound to these areas is going to fit or not (not get cut off/cover something else basically),somehow mark the products(the data coming in from the angular service) to symbolizes which template it best fits into, or even write/modify the template on an individual product basis and save a copy for each product that I could have the component reference instead.
I am partly able to do this with some basic jquery within the component checking the natural width and height of objects coming into the templates, but it REALLY slows down page load, so it's not viable
I came to the conclusion that I should use a custom webpack to do this, possibly write a loader that analyzes the data coming in from the angular service and figure out which template that specific product fits best. The loader could do the calculations to figure out what fits where.
The reason I want to do this in the webpack is because I ultimately need to do the calculations anyway, and it would really bog down the load time to try and do this much logic dynamically on user load, not to mention it's just redundant in my situation to figure this out more than once, as the product info will not change until the website is updated with a new API call anyway.
I currently have a property hard coded that indicates which template to use(a number corresponding to a template), but doing it this way limits the scope of possible templates, and opens up many possible programmer goofs, as well as the fact that it doesn't cover smaller issues that ARE noticeable, and becoming a problem.
So, my question has three parts:
Is it possible to call an Angular Service/API call in general within a loader.js/ts file?
Can I manipulate the data coming in from an Angular Service/API call and have it preserve those changes (changing a flag property of each product coming in from the service)? The interface in angular that handles the data coming in has an extra property that I could assign potentially in the webpack to denote which template to use.
Could I write html files to a folder on the local project/to a database using that data from a service and the base templates?
I apologize for the wordiness, but the situation has a lot of unique parts that I feel are necessary to include. I don't need code examples so much as I just need to know if Webpack/Loaders can/should even do this. I'm obviously open to any suggestions as well as to other ways to solve this problem.
Notes:
All the code I really have so far for the webpack would be configuring my angular.json to run off of a custom-webpack.config.js file I created, it doesn't currently do anything.
The service executes an http request and is subscribed to by angular components that consume it's data, but I could possibly write a js promise and recreate the interface for the scope of the webpack/loader, which is why I think it may be possible.

Does JIRA have a method to call JQuery(window).ready?

We have a Jura plugin written by ourselves, and in it's vm template at the very end there's the following code:
AJS.$(window).ready(function(){
doSomeThing();
});
Inside of this method we are loading some server side data and initializing internal js objects. For some strange reason this specific method doSomeThing is being called twice. Moreover, vm template is also being called twice, owerwriting first template initialization state (but template may be already initialized and contain some data at this point). I don't get it why it's made this way and how to work around this. If someone faced similar thing before and knows what to deal with it - please respond. Much appreciated.
We've found a reason of such bahavior - it's Backbone. Jira creates an element using Backbone View, which calls AJS.$().ready second time during initialization. We stopped usage of this element after our investigation

Angular controller sporadically not loading

I have a website set up with RequireJS and Angular, but at seemingly random moments, it will decide to either not load part of the Javascript or not databind part of the angular code.
The page in question has several different components on it with their own angular controllers. Everything loads fine about 90% of the time, but sometimes something will happen that prevents databinding (angular brackets visible, ng-hide not working,...). Additionally, the 'network' tab in Chrome devtools shows no files being loaded, though they are listed in 'sources'. I don't know if that is relevant somehow. I get no errors in the console at all.
Digging around in the JS console I've found that one of the broken controllers in question does exist ( = I get an object using angular.element(...).scope() ) but when I try to access one of its properties, they are either undefined OR in case of the init() function I use in all my controllers, it returns the function of the parent controller.
What could cause the controller to be loaded but its scope variables to be undefined at seemingly random times?
EDIT:
The only way I've found to sort of reproduce this issue without errors showing up in the console is to initialise the controller as an empty function. This produces similar scope behavior in the console, but it doesn't cause the angular curly braces to show up. I will accept any clue that leads me to the cause of the issue or a viable workaround.
It appears that RequireJS sometimes loaded everyting fast enough for the angular.bootstrap call to be executed before the DOM was completely loaded. This lead to angular processing what is already loaded and ignoring whatever came next. I therefore added a domReady requirement to the setup so I only bootstrap angular when I know the whole page will be there. Since it is hard to know for sure, I can only hope this was the cause and that we won't be seeing this issue again.
You might have to use AngularAMD. Works great for our angular website.

Can't get Angularjs to update view when model updates

I am an Angular newbie and I'm stuck. I have some code on jsbin.com --->(http://jsbin.com/ratageza/7/edit?html,js,output)
It's not pretty but it's showing essentially what I am doing.
In my real code I am using $http.get calls a RESTful backend to load data from a database, and $http.post calls to add to the database. All works fine as far as the database fetches and updates. But after I create a new object using the "Create" form, if I click "List All" my tournaments object is not updated.
I'm confused about whether I would need $apply after the $http.post call? I've tried using it but am either using it wrong or that's not my problem here. I've looked all around stackoverflow and can't seem to find the answer to my problem.
If anybody can see what I'm doing wrong I would really appreciate the help.
Thanks!
Check out the slight changes I made here: http://jsbin.com/ratageza/8/edit?html,js,output
The change is how you define your controller usage. You use TournamentController in two different places which actually gives you two instances of that controller with two separate, isolated scopes. Your List All view was bound to one instance while the Create view was bound to a completely different instance.
Also, your question about $apply. As a general rule, you should very seldom have to use $apply and you should NEVER use it in a controller as it's not safe to use there. Directives are really the only safe place to use $apply and even there it should only be used if data is being modified outside the scope of angular. The order in which things are processed, you find that if you use $scope.$apply() in a controller, you will frequently get exceptions about already being in a digest cycle.
http://jsbin.com/ratageza/10/edit this works.
It's not a $apply or $diget issue. Because you put TournamentController into to part of views. One is create and the other one is to show. That makes a different copy of $scope.tournament;
So I add a div to wrapper the Form and the table. Put the controller to it. Remove the form and table controller.
Hope this can solve your problem. :)

AngularJS and location.path()

I'm working on a webpage based on AngularJS and sometimes I need to change the path (the shebang if you prefer it). The thing is that sometimes $location.path("/my_path_here") works, but sometimes I need to call $scope.$apply() after calling $location.path to make the webbrowser switch to the new path.
Why is this happening?
EDIT:
Example http://pastebin.com/d1SjfCHd
Take a look at this question and Misko's answer: How does data binding work in AngularJS?
That answer explains the process in technical great detail. So, I'm gonna tell it in layman's terms.
AngularJS makes itself work by using dirty checking, there are sets of values that angular is watching. Everytime something big happens, (a click, a function call in controller), angular runs a digest cycle, comparing the watched values to see if there is any change. If there is a change, depending on the watcher, angular will take necessary action (update view, fire a callback, update route).
When you use default directives and no jquery event handling in controller, you will be fine.
However, if you do, you need to know that you are NOT in the angular's digest cycle. Which means the watchers will not fire until a digest occurs.
You need to know when are you updating a variable that is being watched. Mostly it is custom DOM event (or jquery events). When it is the case, you need to let angular know that something is changed and it needs to re-check what happened (ie. update watchers).
This is by doing scope.$apply() after you have changed something.
Bear in mind that you cannot run an $apply() if you are already in the angular's digest cycle. It will throw an error like $digest already in progress.

Categories

Resources