I want the eagarloading expand function of breezejs to give me simple json objects when querying breeze's asp.net web api , so that i can iterate it in my angular views. like {{p.users.projects.projectname}} .. but results i get is not plain json when digging deep inside projects
function getUsersPartials() {
return EntityQuery.from('users').expand('projects')
.using(manager).execute()
.to$q(querySucceeded, _queryFailed);
function querySucceeded(data) {
users = data.results;
console.log(users);
log('Retrived [Users Partials] from remote data source ', users.length, true);
return users;
}
}
you can use .noTracking() on the query in order to get a simple json object
Related
I have a very large array that looks like this:
So in this case it consist of 85 arrays with 473 counts of data in each array.
I am currently passing this array to a web api to then process this data, this is how i do it:
function DeepDataProcess(array, callback) {
console.log(array[0]);
console.log(array);
var promise = HttpRequestFactory.Go({
url: 'api/ImportData/PostDataProcessing',
method: 'POST',
data: array[0] // this is the first index in the array I want to send the full array.
});
promise.then(function (response) { callback(response); }).catch(function (err) { console.log(err); });
}
as you can tell in my javascript i am only passing one array because when i try to send the full array my web api says my parameter is null. I also don't want to call my web api 85 times to get the full array in there, note: i know this array is a multidimensional array.
and below is my web api (c#):
[Route("PostDataProcessing")]//I have tried adding string[,] as the parameter
public IHttpActionResult PostDataProcessing(string[] array) //so that it can accept a multidimensional array but I still get a null value
{
var response = setDataProcessorService().DeepDataProcess(array);
return Ok(response);
}
when i send the full array my web api says my parameter array is empty.
How do I make my c# web api accept a multidimensional array as its parameter?
Below is the request I am making:
I'm trying some basic API Connect tutorials on IBM's platform (running locally using loopback) and have got completely stuck at an early point.
I've built a basic API service with some in-memory data and setter / getter functions. I've then built a separate API which takes two GET parameters and uses one of my getter functions to perform a search based on two criteria. When I run it, I successfully get a response with the following JSON object:
[{"itemId":1,"charge":9,"itemSize":2,"id":2}]
I've then tried to add a piece of server logic that modifies the response data - at this point, I'm just trying to add an extra field. I've added a Javascript component in the Assemble view and included the following code (taken from a tutorial), which I thought should modify the message body returned by the API while still passing it through:
//APIC: get the payload
var json = apim.getvariable('message.body');
//console.error("json %s", JSON.stringify(json));
//same: code to inject new attribute
json.platform = 'Powered by IBM API Connect';
//APIC: set the payload
//message.body = json;
apim.setvariable('message.body', json);
Instead of getting an extra JSON parameter ("platform"), all I get is a 500 error when I call the service. I'm guessing that I'm doing something fundamentally wrong, but all the docs suggest these are the right variable names to use.
You can't access json.platform but at that point json variable is json type. Are you sure that you can add a property to a json type variable if your json object lacks of that property? I mean: What if you first parse the json variable of json type to a normal object, then add new property, and finally stringify to json type again for body assigning purposes?
var json = JSON.parse(apim.getvariable('message.body')); //convert to normal object
json.platform = 'Powered by IBM API Connect'; //add new property
apim.setvariable('message.body', JSON.stringify(json)); //convert to json again before setting as body value
You need to get the context in some determined format, and in this function do your logic. For example if your message is in json you need to do:
apim.readInputAsJSON(function (error, json) {
if (error)
{
// handle error
apim.error('MyError', 500, 'Internal Error', 'Some error message');
}
else
{
//APIC: get the payload
var json = apim.getvariable('message.body');
//console.error("json %s", JSON.stringify(json));
if(json){
//same: code to inject new attribute
json.platform = 'Powered by IBM API Connect';
//APIC: set the payload
//message.body = json;
apim.setvariable('message.body', json);
}
}
});
Reference:
IBM Reference
You have the message.body empty, put a invoke/proxy policy before your gateway/javascript policy for example.
When I store data returned from a service in my controller and then edit it, it also changes the data in the service.
JSFiddle Demo
/* The backend connection part and the actual markdown editor JS have been removed because it would just make the code huge and is irrelevant to the problem */
var myApp = angular.module('myApp', []);
// In my app, this service caches blog post data from my server on the client side and returns single posts from it
myApp.factory('PostService', function ($log, $filter, $http) {
var data;
// Just an example for data pulled from server
data = [{
id: 99999,
text: "Hello"
}];
// Function for returning a post out of the data array loaded from the server
var getData = function (id) {
if (id !== undefined) {
var arr = $filter('filter')(data, {
id: id
});
if (arr.length === 0) {
$log.error('PostService:: getData(' + id + '):: Post Not Found');
return 'not_found';
} else {
$log.debug('PostService:: getData(' + id + '):: Post returned');
return arr[0];
}
} else {
return data;
}
};
return {
getData: getData
};
});
function ctrl($log, $scope, PostService) {
var edit = this;
// Sample post id
edit.editingId = 99999;
// "Copy" (apparrently more "bind") the blog post data to edit.data
edit.data = PostService.getData(edit.editingId);
}
This is used for a markdown editor. I wanted to load the data from the service into the controller, then edit it, and give the service the new version on pressing a "Save" button.
If the aforementioned behaviour is correct in the sense of Angular's databinding, what is a better solution to achieve what I want?
Update
Based on PSL's comment and Thibaud Sowa's answer I changed the getData() function to return a copy using angular.copy(). However, it seems not to be possible to copy one object out of an array (like angular.copy(arr[0])), as it will still return the whole array. See the updated JSFiddle.
Update 2
Well, I was dumb. I corrected it in the fiddle. Thank you for your answers.
This is because you are returning an object. In javascript when you do that it's like if you are passing a pointer.
You should use angular.copy to copy the object field by field, like this:
return angular.copy(data);
See the doc here https://docs.angularjs.org/api/ng/function/angular.copy
Response to your update
Well, I edited your fiddle, to show you that you can copy an item of an array. Actually it's seems that every thing works like you want... (Or I didn't understand your need!)
The updated fiddle:
https://jsfiddle.net/thibaudsowa/aybLwa2L/3/
There is a very simple solution to your problem:
If you dont want to change the data you get from the service make a copy
There are plenty of threads on SO discussing the fastes or most elegant way to deep copy a Javascript object. A simple and rather fast solution is using json parse and stringify like this:
var copyOfA = JSON.parse(JSON.stringify(a));
Apply this to the data you get from your service and you are good to go :)
See the sample code below - in this case, the objectId for the record I am trying to retrieve is known.
My question is, if I don't know the Parse.com objectId, how would I implement the code below?
var Artwork = Parse.Object.extend("Artwork");
var query = new Parse.Query(Artwork);
query.get(objectId, {
success: function(artwork) {
// The object was retrieved successfully.
// do something with it
},
error: function(object, error) {
// The object was not retrieved successfully.
// warn the user
}
});
Query.get() is used when you already know the Parse object id.
Otherwise, one can use query.find() to get objects based on the query parameters.
Sure, you can use Parse Query to search for objects based on their properties.
The thing that wasn't clear to me in the documentation is that once you get the object in the query, you would need to do:
With Query (can return multiple objects):
artwork[0].get('someField');
With 'first' or 'get':
artwork.get('someField');
You cannot do something like artwork.someField like I assumed you would
I have a collection which holds some of the users. Some information that is needed is how many total there are, how many pages, etc. How do I pass these back to the client? Or do they have to come from a separate view in which case I will need more than one ajax call? I'd like to have the collection fetch() and also receive some of this "meta data". What's a good way for handling this?
Generally, you need to handle this in the collection class' parse method. Its responsibility is to take the response and return back an array of model attributes. However, you could do more than that if you wished if you didn't mind the parse method having this additional responsibility.
UserList = Backbone.Collection.extend({
model: User,
url: '/users',
parse: function(data) {
if (!data) {
this.registered_users = 0;
return [];
}
this.registered_users = data.registered_users;
var users = _(data.users).map(
function(user_data) {
var user = {};
user['name'] = user_data.name;
return user;
}
);
return users;
}
});
So in the trivial example above, presume the server returns a response which contains a count of registered users and and an array of user attributes. You would both parse through and return the user attributes and you would pick off the registered user count and just set it as a variable on the model.
The parse method would get called as part of a fetch. So no need to modify the fetch, just use the built-in hook method that you have.
Purists would say that you are giving the parse method a secondary responsibility which might surprise some people (e.g. returning something and modifying model state). However, I think this is okay.
One way to do this is to override the Collection::fetch() method so that it parses this metadata out of the response. You could have your backend return a response like this:
{
"collection": [
{ ... model 1 ... },
{ ... model 2 ... },
...
],
"total_rows": 98765,
"pages": 43
}
In your fetch method which overrides the original Backbone.Collection::fetch() method, you can handle each property of the object separately. Here's you could do the override with a slightly modified fetch method:
_.extend(Backbone.Collection.prototype, {
fetch : function(options) {
options || (options = {});
var collection = this;
var success = options.success;
options.success = function(resp) {
// Capture metadata
if (resp.total_rows) collection.total_rows = resp.total_rows;
if (resp.pages) collection.pages = resp.pages;
// Capture actual model data
collection[options.add ? 'add' : 'refresh'](
collection.parse(resp.collection), options);
// Call success callback if necessary
if (success) success(collection, resp);
};
options.error = wrapError(options.error, collection, options);
(this.sync || Backbone.sync).call(this, 'read', this, options);
return this;
});
Note that this approach using _.extend will affect all your classes which extend Backbone.Collection.
This way, you don't have to make 2 separate calls to the backend.
I would bootstrap the information at pagecreation. Write the information into the html document when the server creates the site. Like that you don't have to have an ajax call at all. I do that with the whole collection in ordner not to first load the page and then wait for the ajax call to return the information needed.
Code example with Python:
Line 64: https://github.com/ichbinadrian/maps/blob/master/python/main.py <- from here
Line 43: https://github.com/ichbinadrian/maps/blob/master/templates/index.html <- into here