how to stop Subscribe? - meteor

I have two selects where the seconds values depends on the choosen value in the first one (Countrys / States) both are collections
When i subscribe with for example Belgian as country i get the states for Belgian
But when i choose another country (ex. Netherlands) the states are added !
I read that i must stop de subscription with .stop() an renew the subscription but this doesnt deems to work ???
My code :
'click #stateList' : function(event, template) {
var x = myTrim($("#countryList").val());
var y = Countrys.findOne({country: x});
var z = y.nr;
if(this.stateSub != null){
this.stateSub.stop();
}
this.stateSub = Meteor.subscribe('stateList', z);
}
It seems that this.stateSub is always undefined even if there is already a working subscribe ???
What am i doing wrong ???
thanks

In template events, this binds to the context document (in impressive ways sometimes) so you can get this._id for the document in the "HTML scope" where the event took place.
If you want to persist an object throughout events, you can use the template object that the event receives as an argument instead. This object will persist across events, as long as you're on the same template, that is.

Related

Returned value from Meteor Helper not showing up in template

I have a Meteor Helper that does a GET request and am supposed to get response back and pass it back to the Template, but its now showing up the front end. When I log it to console, it shows the value corerctly, for the life of mine I can't get this to output to the actual template.
Here is my helper:
UI.registerHelper('getDistance', function(formatted_address) {
HTTP.call( 'GET', 'https://maps.googleapis.com/maps/api/distancematrix/json? units=imperial&origins=Washington,DC&destinations='+formatted_address+'&key=MYKEY', {}, function( error, response ) {
if ( error ) {
console.log( error );
} else {
var distanceMiles = response.data.rows[0].elements[0].distance.text;
console.log(response.data.rows[0].elements[0].distance.text);
return distanceMiles;
}
});
});
In my template I pass have the following:
{{getDistance formatted_address}}
Again, this works fine and shows exactly what I need in the console, but not in the template.
Any ideas what I'm doing wrong?
I posted an article on TMC recently that you may find useful for such a pattern. In that article the problem involves executing an expensive function for each item in a list. As others have pointed out, doing asynchronous calls in a helper is not good practice.
In your case, make a local collection called Distances. If you wish, you can use your document _id to align it with your collection.
const Distances = new Mongo.collection(); // only declare this on the client
Then setup a function that either lazily computes the distance or returns it immediately if it's already been computed:
function lazyDistance(formatted_address){
let doc = Distances.findOne({ formatted_address: formatted_address });
if ( doc ){
return doc.distanceMiles;
} else {
let url = 'https://maps.googleapis.com/maps/api/distancematrix/json';
url += '?units=imperial&origins=Washington,DC&key=MYKEY&destinations=';
url += formatted_address;
HTTP.call('GET',url,{},(error,response )=>{
if ( error ) {
console.log( error );
} else {
Distances.insert({
formatted_address: formatted_address,
distanceMiles: response.data.rows[0].elements[0].distance.text
});
}
});
}
});
Now you can have a helper that just returns a cached value from that local collection:
UI.registerHelper('getDistance',formatted_address=>{
return lazyDistance(formatted_address);
});
You could also do this based on an _id instead of an address string of course. There's a tacit assumption above that formatted_address is unique.
It's Meteor's reactivity that really makes this work. The first time the helper is called the distance will be null but as it gets computed asynchronously the helper will automagically update the value.
best practice is not to do an async call in a helper. think of the #each and the helper as a way for the view to simply show the results of a prior calculation, not to get started on doing the calculation. remember that a helper might be called multiple times for a single item.
instead, in the onCreated() of your template, start the work of getting the data you need and doing your calculations. store those results in a reactive var, or reactive array. then your helper should do nothing more than look up the previously calculated results. further, should that helper be called more times than you expect, you don't have to worry about all those additional async calls being made.
The result does not show up because HTTP.call is an async function.
Use a reactiveVar in your case.
Depending on how is the formated_address param updated you can trigger the getDistance with a tracker autorun.
Regs
Yann

Re-create template while switching routes

How can we re-create template while switching routes?
For example, i have subscriber template. It detects when user scrolls down to a display and subscribes to more data. It takes several parameters.
Example:
amazing_page.html
{{#each}}
{{amazing_topic}}
{{/each}}
{{>subscriber name='topics' count=5}}
subscriber.js
//rough sample code
Template.subscriber.onCreated(function() {
var self = this;
var type = Template.currentData().name;
var count = Template.currentData().count;
var user = Template.currentData().user;
var skipCount = 0;
self.autorun(function(c){
self.subscribe(type, skipCount, user);
var block = true;
$(window).scroll(function() {
if (($(window).scrollTop() + $(window).height()) >= ($(document).height()) && block) {
block = false;
skipCount = skipCount + count;
console.log(type);
console.log(skipCount);
self.subscribe(type, skipCount, user, {
onReady: function() {
block = true;
},
onStop: function() {
console.log('stopped');
}
});
}
});
})
});
I use this template with different parameters in different routes.
The problem is if user switches some routes, and scrolls down in one page, all subscribers he gets in another pages will actualy work in this page. More, they will store increased values for them variables, and will do all included logic.
I found a bad decision when we use Route.getName (for example) comparing and name parameter of subscriber. It is not a best option. Can someone help me to find a good practice for that?:)
Simple Example:
We have 3 different routes:
1)News
2)Videos
3)Topics
These routes templates have included special subscriber-templates. And subscribtion works fine on scroll.
Ok, now let's visit all of them: News, Videos, Topics.
Good, now scroll down and... I have three instance of subscriber template what will subscribe on them own publications, because they not destroyed when we switch routes.
And, as a result - when user scrolling Topics page, he will call subscribtion for News and Videos too, and he will take data from these collections too;)
And - this is a problem:)
UPD:
Looks like we find a decision. If i use Template.instance (autorun/subscribe) it will start working expected, except some strange cases:)
First of all, when i go in another route in next iteration (scroll down) it returns me data from old, destroyed template + error. Next time (next iteration) it will start to subscribe to a correct data. Hmm...it looks like i have mistake in autorun section...or not?
Attached print screen from console
this
It sounds like you have multiple subscriptions to the same collection and that therefore the list of documents shown in various contexts can change in unexpected ways. Meteor manages multiple subscriptions on the same collection by synchronizing the union of the selected documents.
The simplest way to manage each of your views is to make sure that the data context for a particular view uses a .find() with the query you need. This will typically be the same query that your publication is using.
A different but less efficient approach is to .stop() the subscription when you leave a view.

Facebook like load new posts in meteor

I'm in the process of learning meteor. I followed the tutorial to create microscope. If some one submits a post meteor will re render the template for all users. This could be very annoying if there are hundreds of posts then the user will come back to the top of the page and loose track of where he was. I want to implement something similar to what facebook has. When a new post is submitted template isn't rendered rather, a button or link will appear. Clicking it will cause the template to re-render and show the new posts.
I was thinking of using observeChanges on the collection to detect any changes and it does stop the page from showing new posts but only way to show them is to reload the page.
Meteor.publish('posts', function(options) {
var self = this, postHandle = null;
var initializing = true;
postHandle = Posts.find({}, options).observeChanges({
added: function(id, post) {
if (initializing){
self.added('posts', id, post);
}
},
changed: function(id, fields) {
self.changed('posts', id, fields);
}
});
self.ready();
initializing = false;
self.onStop(function() { postHandle.stop(); });
});
Is this the right path to take? If yes, how do I alert the user of new posts? Else, what would be a better way to implement this?
Thank you
This is a tricky question but also valuable as it pertains to a design pattern that is applicable in many instances. One of the key aspects is wanting to know that there is new data but not wanting to show it (yet) to the user. We can also assume that when the user does want to see the data, they probably don't want to wait for it to be loaded into the client (just like Facebook). This means that the client still needs to cache the data as it arrives, just not display it immediately.
Therefore, you probably don't want to restrict the data displayed in the publication - because this won't send the data to the client. Rather, you want to send all the (relevant) data to the client and cache it there until it is ready.
The easiest way involves having a timestamp in your data to work from. You can then couple this with a Reactive Variable to only add new documents to your displayed set when that Reactive Variable changes. Something like this (code will probably be in different files):
// Within the template where you want to show your data
Template.myTemplate.onCreated(function() {
var self = this;
var options = null; // Define non-time options
// Subscribe to the data so everything is loaded into the client
// Include relevant options to limit data but exclude timestamps
self.subscribe("posts", options);
// Create and initialise a reactive variable with the current date
self.loadedTime = new ReactiveVar(new Date());
// Create a reactive variable to see when new data is available
// Create an autorun for whenever the subscription changes ready() state
// Ignore the first run as ready() should be false
// Subsequent false values indicate new data is arriving
self.newData = new ReactiveVar(false);
self.autorun(function(computation) {
if(!computation.firstRun) {
if(!self.subscriptionsReady()) {
self.newData.set(true);
}
}
});
});
// Fetch the relevant data from that subscribed (cached) within the client
// Assume this will be within the template helper
// Use the value (get()) of the Reactive Variable
Template.myTemplate.helpers({
displayedPosts = function() {
return Posts.find({timestamp: {$lt: Template.instance().loadedTime.get()}});
},
// Second helper to determine whether or not new data is available
// Can be used in the template to notify the user
newData = function() {
return Template.instance().newData.get();
});
// Update the Reactive Variable to the current time
// Assume this takes place within the template helper
// Assume you have button (or similar) with a "reload" class
Template.myTemplate.events({
'click .reLoad' = function(event, template) {
template.loadedTime.set(new Date());
}
});
I think this is the simplest pattern to cover all of the points you raise. It gets more complicated if you don't have a timestamp, you have multiple subscriptions (then need to use the subscription handles) etc. Hope this helps!
As Duncan said in his answer, ReactiveVar is the way to go. I've actually implemented a simple facebook feed page with meteor where I display the public posts from a certain page. I use infinite scroll to keep adding posts to the bottom of the page and store them in a ReactiveVar. Check the sources on github here and the live demo here. Hope it helps!

What's the Difference between Template.Instance() vs template.data?

When creating a Meteor event handler, what's the difference between...
'click .something': function(e,t){
var data = t.data
}
vs
'click .something': function(e,t){
var data = Template.instance().data
}
They both seem to bring up the same data. Is there a reason why I should one or the other?
Similar question here:
Difference between Template.instance() and this
The thing to realize is that:
In the template life-cycle functions (onCreated, onRendered...) this is equal to Template.instance() so this.data is the same as Template.instance().data AT THAT TIME!
In a helper or event, this is the current data context.
So, note an important thing here: the Data context can change over time if your data changes upstream:
If you pass data to a template, the template will be re-rendered with the new data. New data = new data context.
So if you do something like:
Template.example.onCreated(function() {
this.data.myKey = "my example data set on template creation"; //WRONG!
// or equivalently:
Template.instance().data.myOtherKey = "another key"; //WRONG!
})
well, this data may be under this (i.e. the data context) in your helper (this.myKey) but only as long as the upstream data does not change.
As soon as the upstream data changes, this will be the new data context, and will NOT contain your added data.
So, in summary:
If you need to add context to your template in onCreated or onRendered, make sure you do NOT bind it to the current data context, but to the Template.instance()
you should do:
Template.example.onCreated(function() {
this.myKey = "my example data set on template creation";
// or equivalently:
Template.instance().myOtherKey = "another key";
})
and you can access this data in helper and events via
Template.instance().myKey
It's actually Template.instance() (with a lower i), and as this function returns the current template instance in scope (the one where the event originated), there's no difference with the second parameter of an event handler, which also holds the current template instance, this is why you can access the template data indifferently using Template.instance().data or t.data in an event handler.
There is however a simpler way to access the current data context inside an event handler : the this object is bound to the data context where the event was triggered.

Adding a model to Collection in Backbone [duplicate]

I'm running into an odd issue with a Backbone.js Model where an array member is being shown as blank. It looks something like this:
var Session = Backbone.Model.extend({
defaults: {
// ...
widgets: []
},
addWidget: function (widget) {
var widgets = this.get("widgets");
widgets.push(widget);
this.trigger("change:widgets", this, widgets);
},
// ...
// I have a method on the model to grabbing a member of the array
getWidget: function (id) {
console.log(this.attributes);
console.log(this.attributes.widgets);
// ...
}
});
I then add a widget via addWidget. When trying getWidget the result I get (in Chrome) is this:
Object
widgets: Array[1]
0: child
length: 1
__proto__: Array[0]
__proto__: Object
[]
It's showing that widgets is not empty when logging this.attributes but it's shown as empty when logging this.attributes.widgets. Does anyone know what would cause this?
EDIT
I've changed the model to instantiate the widgets array in the initialization method to avoid references across multiple instances, and I started using backbone-nested with no luck.
Be careful about trusting the console, there is often asynchronous behavior that can trip you up.
You're expecting console.log(x) to behave like this:
You call console.log(x).
x is dumped to the console.
Execution continues on with the statement immediately following your console.log(x) call.
But that's not what happens, the reality is more like this:
You call console.log(x).
The browser grabs a reference to x, and queues up the "real" console.log call for later.
Various other bits of JavaScript run (or not).
Later, the console.log call from (2) gets around to dumping the current state of x into the console but this x won't necessarily match the x as it was in (2).
In your case, you're doing this:
console.log(this.attributes);
console.log(this.attributes.widgets);
So you have something like this at (2):
attributes.widgets
^ ^
| |
console.log -+ |
console.log -----------+
and then something is happening in (3) which effectively does this.attributes.widgets = [...] (i.e. changes the attributes.widget reference) and so, when (4) comes around, you have this:
attributes.widgets // the new one from (3)
^
|
console.log -+
console.log -----------> widgets // the original from (1)
This leaves you seeing two different versions of widgets: the new one which received something in (3) and the original which is empty.
When you do this:
console.log(_(this.attributes).clone());
console.log(_(this.attributes.widgets).clone());
you're grabbing copies of this.attributes and this.attributes.widgets that are attached to the console.log calls so (3) won't interfere with your references and you see sensible results in the console.
That's the answer to this:
It's showing that widgets is not empty when logging this.attributes but it's shown as empty when logging this.attributes.widgets. Does anyone know what would cause this?
As far as the underlying problem goes, you probably have a fetch call somewhere and you're not taking its asynchronous behavior into account. The solution is probably to bind to an "add" or "reset" event.
Remember that [] in JS is just an alias to new Array(), and since objects are passed by reference, every instance of your Session model will share the same array object. This leads to all kinds of problems, including arrays appearing to be empty.
To make this work the way you want, you need to initialize your widgets array in the constructor. This will create a unique widget array for each Session object, and should alleviate your problem:
var Session = Backbone.Model.extend({
defaults: {
// ...
widgets: false
},
initialize: function(){
this.set('widgets',[]);
},
addWidget: function (widget) {
var widgets = this.get("widgets");
widgets.push(widget);
this.trigger("change:widgets", this, widgets);
},
// ...
// I have a method on the model to grabbing a member of the array
getWidget: function (id) {
console.log(this.attributes);
console.log(this.attributes.widgets);
// ...
}
});
Tested in a fiddle with Chrome and Firefox: http://jsfiddle.net/imsky/XBKYZ/
var s = new Session;
s.addWidget({"name":"test"});
s.getWidget()
Console output:
Object
widgets: Array[1]
__proto__: Object
[
Object
name: "test"
__proto__: Object
]

Resources