In the comment by the link Jquery multiple Ajax Request in array loop the author said that making ajax requests in a loop may end up DDOS-ing ourselves.
Does that apply only a loop or multiple ajax calls in general? I mean may the risk of DDOS-ing be as well if I make multiple ajax requests via recursive function like
ajax(0);
ajax(index) {
var xhr = new XMLHttpRequest();
xhr.onreadystatechange = function() {
if(this.readyState == 4 && this.status == 200) {
ajax(index+1)
}
};
xhr.open('POST', 'http://example.com/ajax_handler.php');
xhr.setRequestHeader('Content-type', 'application/x-www-form-urlencoded; charset=UTF-8');
xhr.send();
}
ps. I understand that we can "congregate all data together then send that in a single request to the server", but I need to run generating static pages passing data from the client to the server. So if there are dozens of thousands of pages I must to pass to the server via AJAX, they can't be passing as one single request because of limit of POST requests.
Why so? I would just like to keep all the logic of the generator at the client and call at the server only standard operations like reading and writing files. That is the client reads templates and content via ajax and server reading function, build page html according to its logic and pass the whole html to the server to be written in a html file
The problem Rory McCrossan was describing was if you make multiple requests at once. If you have lots of requests, you might overload the server (and/or your network connection) - you shouldn't make tons of requests at once. (Probably best to not send more than 5 request a second to a server, or something like that.)
But in your code, you're not sending out the requests at once; you only have at most one request active at any time, so the issue he was describing isn't something you need to worry about.
That said,
dozens of thousands of pages I must to pass to the server via AJAX
is a pretty odd requirement and will require a lot of bandwidth even if you don't overload the network. Consider if there's any more elegant solutions to the problem, such as generating/sending a page only when that page is requested.
Related
I need to send many requests to server 50-100 requests to load data, each response has at least 0.5KB and at most 7KB of data.
I send the requests using ajax as following: (code is simplified)
for (var i=0; i<elements.length; i++) {
var element = elements[i];
// make ajax call with element as parameter and update page to show data for element
}
This works for my needs, because I dont need data to come from server in order, and it works most of the time. But sometimes the last few elements dont get loaded and I get communication link failure error in my chrome javascript console.
I am assuming that the server got overloaded, how can I regulate sending requests to make sure I get a response for each request in the shortest time possible?
Notes:
I use Spring MVC in the backend
I use ExtJS Ajax to make the requests
Try using seperate loops for your data uploading process. Overloading is the only cause for that Communication failure.
I solved this by recursively calling the each request, this way only one request is sent to the server at the time, and overload is avoided.
This might not even be an AngularJS question and could just be an AJAX question. I'm new to the "developer" side of the frontend so bear with me.
When making an AJAX call to fetch JSON data, where does the logic behind what data is returned and viewed fall? In my mind, there would be a couple of possibilities and I want to understand which is the proper choice and why.
Let's use an example of searching and playing a Youtube video.
The logic could fall to the backend (controller), where the JSON is rendered based on some logic to give you a JSON file with exactly the right data. i.e. you search "cat videos" and when making an AJAX call, the JSON file you pull has been rendered to be only cat videos.
The opposite end would be that the Angular controller has the logic. This would imply that all data is called (cat videos along with everything else... music videos, funny videos, tutorials, and so on) and then sorted through on the client side. This, to me anyway, would be more inefficient / slow for the client, so doesn't seem to make sense. I suppose still might do some filtering of the data on the client side though. So, maybe a search for "cat videos" wouldn't return ALL videos, but definitely all cat videos and any filtering based on, say, # of views, video length, and so on would be done on the client side (vs. calling the database again for a "new" set of videos).
Not sure if this is accurate, but could you have logic in your factory to return only a portion of the data? However, I believe the entire JSON file would need to be rendered, but only portions would be returned. I guess depending on where the JSON file renders (i.e. backend or frontend) this could be similar to either option #1 or #2.
Or maybe I'm misunderstanding things entirely and the way this works is entirely different!
I'm basically looking to figure out how the scenarios of 1. user searches a term and results are shown, 2. user clicks a search result and now more detailed data of the result is on it's own page. And how this ends up working out. I'm looking for help with AngularJS, but I think this ultimately an AJAX question (single page app or not) more than anything.
There's a few critical concepts you may be confused about.
First. JSON is not a file, it's a format, more simply, a type of string. It's really good for collapsing arrays and storing address-value pairs, so a lot of data flies around in that format. Strictly speaking, they are JSON objects, but they're a lot like strings and arrays. It looks like this, if I remember correctly:
{ "name" : "john doe", "pet" : "dog", "hobby" : "parasailing" }
Second, AJAX is a request to the server, made from the client (the browser) after the original page has loaded. That is, you type in 'youtube.com' and the youtube server receives the request and sends a big pile of HTML back to your browser.
You watch your video, make a rating, and the browser doesn't reload the page but instead sends a separate request back to the youtube server with your rating. There's a parameter in the request that says "send it to ratingspage.php". This request is AJAX.
Now, the logic happens (server-side). ratingspage.php receives your request. It contacts the databases, updates or fails or whatever, and sends back a response to your browser. This response may be in JSON format.
Finally, your browser parses that response and updates the DOM (HTML document) as appropriate.
At this point, it's worth noting that if the logic happened on the client-side (browser), the user could see it - this is a security problem! So, sensitive operations should be carried out on the server side, where you can test and sanitize the request data.
In summary:
AJAX is separate from the initial load event.
Information sent is gathered from the client browser
Logic happens server-side
Logic can use whatever language the server understands (PHP, Java, Ruby, etc.)
Information is returned to the browser
Information sent and received may use JSON format
Everything client-side happens in Javascript
Here's a bare-bones ajax request (done in Javascript) with comments. This has no exception handling, state checking, or anything so don't use it! But it gives you the basic idea.
// Make a new request
var req = new XMLHttpRequest(); }
// Requests will have various states depending on whether they're processing,
// finished, error, etc. We'll assume everything went OK.
// We need to establish a handler before the request
// is sent so it knows what to do.
req.onreadystatechange = function() {
// Here's what the server sent back to the browser
alert(req.responseText);
}
// Using the GET method, set up some parameters
req.open("GET", "somelogicpage.php?blah=blee&bloo=bar", true);
// Send the request
req.send(null);
Server-side, somelogicpage.php may look like:
<?php
if ($_GET['blah'] != 'blee']) {
// This is the response text!
echo "Sorry, you need to blee when you blah.";
}
else {
// (or this)
echo "I'm ecstatic to report nothing is wrong!";
}
?>
Your alert(req.responseText) from the handler function in the previous Javascript will say whatever the PHP has dumped out.
So yes, you can use whatever portion of the request you like, and return whatever you like. Javascript kicks bleep.
I am trying to write a plugin which will work a lot with my server. Every page load will invoke an AJAX call to my server for data, the server should return a simple string.
Now I am trying to understand what would be the best aproach for this type of program.
Should I just create an AJAX call every time I need the data or is there some method I could create an open connection (despite the change of webpages) to save on server power?
Should I somehow listen to some port or something of the sort?
Do I have other options or what should I do to do this the most efficient way?
You can use HTML5 websockets (http://www.html5rocks.com/en/tutorials/websockets/basics/)
If you use this approach, then you will need to re-think the way you program your webserver, since websockets don't follow the request-response paradigm AJAX do. Instead they use a connection to stream data so you will need to open a port on your server and listen to it, the way to do it depends on the language or framework you are using. This is fast and responsive but will only work on most modern browsers.
Other approach is using Long Polling (http://techoctave.com/c7/posts/60-simple-long-polling-example-with-javascript-and-jquery). This is used by some chat clients. It works sending an AJAX request to the server, the server receives it and keeps it waiting until the data is available and then the response is sent. Then the client makes another request, waits and repeats.
Probably you will almost never want to send simple strings to the client. It's almost always better to use XML or JSON to encode the response.
Just create a simple AJAX call and put it on each page, or save it as it's own file and put a server include on each page in the header. Simple as that!
$(document).load(function(){
$.ajax({
type: "POST",
url: "/where_your_string_is.php",
success: function(msg){
$("#stringHolder").html(msg);
}
});
});
Websockets API allows bi-directional communication, but I've just found that there's another option called HTML SSE that might be used if you only need to pull data. So if you've stumbled upon this question, consider this option as well.
Let's imagine the situation that we've sent two similar (almost similar) async ajax requests to server one by one. Because of lag in network, the second request was executed before first request.
Ajax request #1: /change/?object_id=1&position=5
Ajax request #2: /change/?object_id=1&position=6
In result, we have object_id=1 position set to position=5, but we want position=6 because Ajax request #2 was executed after Ajax request #1 by us.
What is the best practice to avoid this on server side and client side?
Are you worried about racing conditions from the same client or from multiple clients?
If from the same client, I would think the safest bet would be to include a unix timestamp in the ajax request and log this value on the server. If a request comes with a timestamp that is older than the last logged value, ignore the request (or send a warning back to the browser).
I'm not sure how you would handle multiple clients with unsynchronized clocks...
For situations like this, I usually put a check in my success handler to make sure that the value being returned is still the one that I want. This will require sending up the parameter you're searching across in the results object.
For example:
var query = $('input').val();
$.get('/search', { query: query }, function(res) {
if(res.query == $('input').val()) {
//show search results
}
});
I don't know the particulars of your use case, but this general pattern should help.
On the server :
Build a request table to map request id to timestamp
Log any request to the server, expect all requests come with timestamp
If any request comes out of order (e.g. position 6 comes before 5)
Check the request table, if it is an earlier request (timestamp) then do not process the request and send an ignore flag
If it comes in order
This is fine, proceed as usual and no need to send any ignore order
On the client:
When request comes back in, check the ignore flag. If it is there. Don't do anything to the client
Otherwise proceed as usual by processing the data
Note that this implementation that I suggested requires you to send back and forth data (such as JSON) and not the presentation code (such as HTML fragment) as you would need to check for the ignore flag on the client side.
This answer is similar to what #Farray suggestion of using timestamp.
How do I design a Django/Javascript application to provide for conditional Ajax responses to conventional HTTP requests?
On the server, I have a custom-built Form object. When the browser POSTS the form's data, the server checks the submitted data against existing data and rules (eg, if the form adds some entity to a database, does that entity already exist in the database?). If the data passes, the server saves, generates an ID number and adds it to the form's data, and passes the form and data back to the browser.
if request.method == 'POST':
formClass = form_code.getCustomForm()
thisForm = formClass(data=request.POST)
if thisForm.isvalid():
saveCheck = thisForm.saveData()
t = loader.get_template("CustomerForm.html")
c = Context({ 'expectedFormObj': thisForm })
(Note that my custom logic checking is in saveData() and is separate from the html validation done by isvalid().)
So far, standard Django (I hope). But if the data doesn't pass, I want to send a message to the browser. I suppose saveData() could put the message in an attribute of the form, and the template could check for that attribute, embed its data as javascript variable and include a javascript function to display the message. But passing all that form html back, just to add one message, seems inelegant (as does the standard Django form submission process, but never mind). In that case I'd like to just pass back the message.
Now I suppose I could tie a Javascript function to the html form's onsubmit event, and have that issue an XMLHttpRequest, and have the server respond to that based on the output of the saveData() call. But then the browser has two requests to the server outstanding (POST and XHR). Maybe a successful saveData() would rewrite the whole page and erase any potential for conflict. But I'd also have to get the server to sequence its response to the XHR to follow the response to the POST, and figure out how to communicate the saveData outcome to the response to the XHR. I suppose that is doable, even without the thread programming I don't know, but it seems messy.
I speculate that I might use javascript to make the browser's response conditional to something in the response to the POST request (either rewrite the whole page, or just display a message). But I suspect that the page's javascript hands control over the browser with the POST request, and that any response to the POST would just rewrite the page.
So can I design a process to pass back the whole form only if the server-side saveData() works, and a message that is displayed without rewriting the entire form if saveData() doesn't? If so, how?
Although you can arrange for your views to examine the request data to decide if the response should be an AJAXish or plain HTML, I don't really recommend it. Put AJAX request handlers in a separate URL structure, for instance all your regular html views have urls like /foo/bar and a corresponding api call for the same info would be /ajax/foo/bar.
Since most views will examine the request data, then do some processing, then create a python dictionary and pass that to the template engine, you can factor out the common parts to make this a little easier. the first few steps could be a generic sort of function that just returns the python dictionary, and then actual responses are composed by wrapping the handler functions in a template renderer or json encoder.
My usual workflow is to initially assume that the client has no javascript, (which is still a valid assumption; many mobile browsers have no JS) and implement the app as static GET and POST handlers. From there I start looking for the places where my app can benefit from a little client side scripting. For instance I'll usually redesign the forms to submit via AJAX type calls without reloading a page. These will not send their requests to the same URL/django view as the plain html form version would, since the response needs to be a simple success message in plain text or html fragment.
Similarly, getting data from the server is also redesigned to respond with a concise JSoN document to be processed into the page on the client. This also would be a separate URL/django view as the corresponding plain html for that resource.
When dealing with AJAX, I use this:
from django.utils import simplejson
...
status = simplejson.dumps({'status': "success"})
return HttpResponse(status, mimetype="application/json")
Then, AJAX (jQuery) can do what it wants based on the return value of 'status'.
I'm not sure exactly what you want with regards to forms. If you want an easier, and better form experience, I suggest checking out uni-form. Pinax has a good implementation of this in their voting app.
FYI, this isn't an answer...but it might help you think about it a different way
Here's the problem I'm running into...Google App Engine + jQuery Ajax = 405 Method Not Allowed.
So basically I get the thing to work using the outlined code, then I can't make the AJAX request :(.