At the moment I have the following understanding (which, I assume, is incomplete and probably even wrong).
A web server receives request from a client. The requests are coming to a particular "path" ("address", "URL") and have a particular type (GET, POST and probably something else?). The GET and POST requests can also come with variables and their values (which can be though as a "dictionary" or "associate array"). The parameters of GET requests are set in the address line (for example: http://example.com?x=1&y=2) while parameters of POST requests are set by the client (user) via web forms (in other words, a user fills in a form and press "Submit" button).
In addition to that we have what is called SESSION (also known as COOKIES). This works the following way. When a web-server gets a request (of GET or POST type) it (web server) checks the values of the sent parameters and based on that it generates and sends back to the client HTML code that is displayed in a browser (and is seen by the user). In addition to that the web servers sends some parameters (which again can be imagined as "dictionary" or "associative arrays"). These parameters are saved by the browser somewhere on the client side and when a client sends a new request, he/she also sends back the session parameters received earlier from the web server. In fact server says: you get this from me, memorize it and next time when you speak to me, give it back (so, I can recognize you).
So, what I do not know is if client can see what exactly is in the session (what parameters are there and what values they have) and if client is able to modify the values of these parameters (or add or remove parameters). But what user can do, he/she might decide not to accept any cookies (or session).
There is also something called "local storage" (it is available in HTML5). It works as follows. Like SESSION it is some information sent by web-server to the client and is also memorized (saved) by the client (if client wants to). In contrast to the session it is not sent back b the client to the server. Instead, JavaScripts running on the client side (and send by web-servers as part of the HTML code) can access information from the local storage.
What I am still missing, is how AJAX is working. It is like by clicking something in the browser users sends (via Browser) a request to the web-server and waits for a response. Then the browser receives some response and use it to modify (but not to replace) the page observed by the user. What I am missing is how the browser knows how to use the response from the web-server. Is it written in the HTML code (something like: if this is clicked, send this request to the web server, and use its answer (provided content) to modify this part of the page).
I am going to answer to your questions on AJAX and LocalStorage, also on a very high level, since your definition strike me as such on a high level.
AJAX stands for Asynchronous JavaScript and XML.
Your browser uses an object called XMLHTTPRequest in order to establish an HTTP request with a remote resource.
The client, being a client, is oblivious of what the remote server entails on. All it has to do is provide the request with a URL, a method and optionally the request's payload. The payload is most commonly a parameter or a group of parameters that are received by the remote server.
The request object has several methods and properties, and it also has its ways of handling the response.
What I am missing is how the browser knows how to use the response
from the web-server.
You simply tell it what to do with the reponse. As mentioned above, the request object can also be told what to do with a response. It will listen to a response, and when such arrives, you tell the client what to do with it.
Is it (the response) written in the HTML code?
No. The response is written in whatever the server served it. Most commonly, it's Unicode. A common way to serve a response is a JSON (JavaScript Object Notation) object.
Whatever happens afterwards is a pure matter of implementation.
LocalStorage
There is also something called "local storage" (it is available in
HTML5). It works as follows. Like SESSION it is some information sent
by web-server to the client and is also memorized (saved) by the
client (if client wants to)
Not entirely accurate. Local Storage is indeed a new feature, introduced with HTML5. It is a new way of storing data in the client, and is unique to an origin. By origin, we refer to a unique protocol and a domain.
The life time of a Local Storage object on a client (again, per unique origin), is entirely up to the user. That said, of course a client application can manipulate the data and decide what's inside a local storage object. You are right about the fact that it is stored and can be used in the client through JavaScript.
Example: some web tracking tools want to have some sort of a back up plan, in case the server that collects user data is unreachable for some reason. The web tracker, sometimes introduced as a JavaScript plugin, can write any event to the local storage first, and release it only when the remote server confirmed that it received the event successfully, even if the user closed the browser.
First of all, this is just a simple explanation to clarify your mind. To explain this stuff in more detail we would need to write a book. This been said, I'll go step by step.
Request
A request is a client asking for / sending data to a server.
This request has the following parts:
An URL (Protocol + hostname/IP + path)
A Method (GET, POST, PUT, DELETE, PATCH, and so on)
Some optional parameters (the way they are sent depends on the method)
Some headers (metadata sent to the server)
Some optional cookies
An optional SESSION ID
Some explanations about this:
Cookies can be set by the client or by the server, but they are always stored by the client's browser. Therefore, the browser can decide whether to accept them or not, or to delete or modify them
Session is stored in the server. The server sends the client a session ID to be able to recongnize him in any future request.
Session and cookies are two different things. One is server side, and the other is client side.
AJAX
I'll ommit the meaning of the acronym as you can easily google it.
The great thing about AJAX is the very first A, that stands for asynchronous, what means that the JS engine (in this case built in the browser) won't block until the response gets back.
To understand how AJAX works, you have to know that it's very much alike a common request, but with the difference that it can be triggered without reloading the web page.
The content of the response it whatever you want it to be. From some HTML code, to a JSON string. Even some plain text.
The way the response is treated depends on the implementation and programming. As an example, you could simply alert() the result of an AJAX call, or you and append it to a DOM element.
Local Storage
This doesn't have much to do with anything.
Local storage is just some disk space offered by the browser so you can save data in the browser that persists even if the page or the browser is closed.
An example
Chrome offers a javascript API to manage local storage. It's client side, and you can programmatically access to this storage and make CRUD opperations. It's just like a non-sql non-relational DB in the browser.
I wil summarize your main questions along with a brief answer right below them:
Q1:
Can the client see what exactly is in the session?
A: No. The client only knows the "SessionID", which is meta-data (all other session-data is stored on server only, and client can't see or alter it). The SessionID is used by the server only to identify the client and to map the application process to it's previous state.
HTTP is a stateless protocol, and this classic technique enables it as stateful.
There are very rare cases when the complete session data is stored on client-side (but in such cases, the server should also encrypt the session data so that the client can't see/alter it).
On the other hand, there are web clients that don't have the capability to store cookies at all, or they have features that prevent storing cookie data (e.g. the ability of the user to reject cookies from domains). In such cases, the workaround is to inject the SessionID into URL parameters, by using HTTP redirects.
Q2:
What's the difference between HTML5 LocalStorage and Session?
LocalStorage can be viewed as the client's own 'session' data, or better said a local data store where the client can save/persist data. Only the client (mainly from javascript) can access and alter the data. Think of it as javascript-controlled persistent storage (with the advantage over cookies that you can control what data, it's structure and the format you want to store it). It's also more advantageous than storing data to cookies - which have their own limitations such as data size and structure.
Q3:
How AJAX works?
In very simple words, AJAX means loading on-demand data on top of an already loaded (HTML) page. A typical http request would load the whole data of a page, while an ajax request would load and update just a portion of the (already-loaded) page.
This being said, an AJAX request is very similar to a standard HTTP Request.
Ajax requests are controlled by the javascript code and it can enrich the interaction with the page. You can request specific segments of data and update sections of the page.
Now, if we remember the old days when any interaction with a website (eg. signing in, navigating to other pages etc.) required a complete page reload? Back then, a lot of unnecessary traffic occurred just to perform any simple action. This in turn impacted site responsiveness, user experience, network traffic etc.
This happened due to browsers incapability (at that time) to [a.] perform a parallel HTTP request to the server and [b]render a partial HTML view.
Modern browsers come with these two features that enables AJAX technology - that is, invoking asynchronous(parallel) HTTP Requests (Ajax HTTP Requests) and they also provide on-the-fly DOM alteration mechanism via javascript (real-time HTML Document Object Model manipulation).
Please let me know if you need more info on these topics, or if there's anything else I can help with.
For a more profound understanding, I also recommend this nice web history article as it explains how everything started from when HTML was created and what was it's purpose (to define [at the time] rich documents), and then how HTTP was initially created and what problem it solved (at the time - to "transfer" static HTML). That explains why it is a stateless protocol.
Later on, as HTML and the WEB evolved, other needs emerged (such as the need to interact with the end-user) - and then the Cookie mechanism enhanced the protocol to enable stateful client-server communication by using session cookies. Then Ajax followed. Nowadays, the cookies come with their own limitations too and we have LocalStorage. Did I also mention WebSockets?
1. Establishing a Connection
The most common way web servers and clients communicate is through a connection which follows Transmission Control Protocol, or TCP. Basically, when using TCP, a connection is established between client and server machines through a series of back-and-forth checks. Once the connection is established and open, data can be sent between client and server. This connection can also be termed a Session.
There is also UDP, or User Datagram Protocol which has a slightly different way of communicating and comes with its own set of pros and cons. I believe recently some browsers may have begun to use a combination of the two in order to get the best results.
There is a ton more to be said here, but unless you are going to be writing a browser (or become a hacker) this should not concern you too much beyond the basics.
2. Sending Packets
Once the client-server connection is established, packets of data can be sent between the two. TCP packets contain various bits of information to assist in communication between the two ports. For web programmers, the most important part of the packet will be the section which includes the HTTP request.
HTTP, Hypertext Transfer Protocol is another protocol which describes what the makeup/format of these client-server communications should be.
A most basic example of the relevant portion of a packet sent from a client to a server is as follows:
GET /index.html HTTP/1.1
Host: www.example.com
The first line here is called the Response line. GET describes the method to be used, (others include POST, HEAD, PUT, DELETE, etc.) /index.html describes the resource requested. Finally, HTTP/1.11 describes the protocol being used.
The second line is in this case the only header field in the request, and in this case it is the HOST field which is sort of an alias for the IP address of the server, given by the DNS.
[Since you mentioned it, the difference between a GET request and a POST request is simply that in a GET request the parameters (ex: form data) is included as part of the Response Line, whereas in a POST request the parameters will be included as part of the Message Body (see below).]
3. Receiving Packets
Depending on the request sent to the server, the server will scratch its head, think about what you asked it, and respond accordingly (aka whatever you program it to do).
Here is an example of a response packet send from the server:
HTTP/1.1 200 OK
Content-Type: text/html; charset=UTF-8
...
<html>
<head>
<title>A response from a server</title>
</head>
<body>
<h1>Hello World!</h1>
</body>
</html>
The first line here is the Status Line which includes a numerical code along with a brief text description. 200 OK obviously means success. Most people are probably also familiar with 404 Not Found, for example.
The second line is the first of the Response Header Fields. Other fields often added include date, Content-Length, and other useful metadata.
Below the headers and the necessary empty line is finally the (optional) Message Body. Of course this is usually the most exciting part of the response, as it will contain things like HTML for our browsers to display for us, JSON data, or pretty much anything you can code in a return statement.
4. AJAX, Asynchronous JavaScript and XML
Based off all of that, AJAX is fairly simple to understand. In fact, the packets sent and received can look identical to non-ajax requests.
The only difference is how and when the browser decides to send a request packet. Normally, upon page refresh a browser will send a request to the server. However, when issuing an AJAX request, the programmer simply tells the browser to please send a packet to the server NOW as opposed to on page refresh.
However, given the nature of AJAX requests, usually the Message Body won't contain an entire HTML document, but will request smaller, more specific bits of data, such as a query from a database.
Then, your JavaScript which calls the Ajax can also act based off the response. Any JavaScript method is available as making an Ajax call is just another JavaScript function. Thus, you can do things like innerHTML to add/replace content on your page with some HTML returned by the Ajax call. Alternatively though, you could also do something like make an Ajax call which simply should return True or False, and then you could call some JavaScript function with an if else statement. As you can hopefully see, Ajax has nothing to do with HTML per say, it is just a JavaScript function which makes a request from the server and returns the response, whatever it may be.
5. Cookies
HTTP Protocol is an example of a Stateless Protocol. Basically, this means that each pair of Request and Response (like we described) is treated independently of other requests and responses. Thus, the server does not have to keep track of all the thousands of users who are currently demanding attention. Instead, it can just respond to each request individually.
However, sometimes we wish the server would remember us. How annoying would it be if every time I waned to check my Gmail I had to log in all over again because the server forgot about me?
To solve this problem a server can send Cookies to be stored on the client's machine. The server can send a response which tells the client to store a cookie and what exactly it should contain. The client's browser is in charge of storing these cookies on the client's system, thus the location of these cookies will vary depending on your browser and OS. It is important to realize though that these are just small files stored on the client machine which are in fact readable and writable by anyone who knows how to locate and understand them. As you can imagine, this poses a few different potential security threats. One solution is to encrypt the data stored inside these cookies so that a malicious user won't be able to take advantage of the information you made available. (Since your browser is setting these cookies, there is usually a setting within your browser which you can modify to either accept, reject, or perhaps set a new location for cookies.
This way, when the client makes a request from the server, it can include the Cookie within one of the Request Header Fields which will tell the server, "Hey I am an authenticated user, my name is Bob, and I was just in the middle of writing an extremely captivating blog post before my laptop died," or, "I have 3 designer suits picked out in my shopping cart but I am still planning on searching your site tomorrow for a 4th," for example.
6. Local Storage
HTML5 introduced Local Storage as a more secure alternative to Cookies. Unlike cookies, with local storage data is not actually sent to the server. Instead, the browser itself keeps track of State.
This alternative also allows much larger amounts of data to be stored, as there is no requirement for it to be passed across the internet between client and server.
7. Keep Researching
That should cover the basics and give a pretty clear picture as to what is going on between clients and servers. There is more to be said on each of these points, and you can find plenty of information with a simple Google search.
I currently have a problem where I send an asynchronous ajax request to a .NET controller in order to start a database search. This request makes it to the server which kicks off the search and immediately (less than a second) replies to the callback with a search ID, at which point I begin sending ajax requests every 10 seconds to check if the search has finished. This method works fine, and has been tested successfully with multiple users sending simultaneous requests.
If I send a second search request from the same user before the first search is finished, this call will not make it to the controller endpoint until after the first search has completed, which can take up to a minute. I can see the request leave chrome (or FF/IE) in the dev tools, and using Fiddler as a proxy I can see the request hit the machine that the application is running on, however it will not hit the breakpoint on the first line of the endpoint until after the first call returns.
At the point this call is blocking, there are typically up to 3 pending requests from the browser. Does IIS or the .NET architecture have some mechanism that is queuing my request? Or if not, what else would be between the request leaving the proxy and entering the controller? I'm at a bit of a loss for how to debug this.
I was able to find the issue. It turns out that despite my endpoint being defined asynchronously, ASP.NET controllers by default synchronize by session. So while my endpoints were able to be executed simultaneously across sessions, within the same session it would only allow one call at a time. I was able to fix the issue by setting the controller SessionState attribute to Read Only, allowing my calls to come through without blocking.
[SessionState(System.Web.SessionState.SessionStateBehavior.ReadOnly)]
I'm been reading up on Ajax and would like to see from the stackoverflow community if I'm understanding everything correctly.
So the normal client server interaction is a user pulls up a web browser types in a url and a HTTP request is sent to the server requesting the page and resources( css, pics ) from the web server. The web server responds to the client via HTTP the page/resources requested and the browser renders the html/JavaScript for the user to view the page.
1) So would it be safe to say that XMLHttpRequest( XHR ) object is doing the same process as the browser except your not requesting html from the server, your requesting text in some type of format?
2) Is it true that a XHR object is much like a regular object that can be manipulated by the program creating the object( like a normal object ), but also sends and receives data with another program( web server ) via HTTP?
3) So in my mind when a XHR is created it is loaded into memory and we setup some of the objects arguments when we do the request.open(“GET”, url, true). Once we do a request.send(null) the object basically attempts to “GET” the url via HTTP and once we get the data back from the server it is put in the responseText argument. Am I understanding this correctly?
4) Also synchronous vs asynchronous. When I think of synchronous I think of steps having to be followed in order. For example, I push a button, data gets sent to server, and I have to wait for data to come back before I can do anything else. With asynchronous connections I would push button, data gets sent to server, I do what ever I want while data gets sent back. Is this a good analogy?
1) Nope. The XMLHttpRequest object does exactly what its name implies -- it initiates an HTTP request. This request can be in XML, or HTML, or PHP. At the end of the day, the browser doesn't care, because in an AJAX request, it doesn't parse the request -- you have to do it yourself. So it doesn't automatically render the HTML from an AJAX request.
2) I'm not sure about manipulation (the XHR object may be immutable) but possibly. Would you ever need to extend it or manipulate it?
Yes, you can change properties of the object and so on. I apologize. I didn't understand you at first :)
3) Yep.
4) That's a great analogy. It's exactly what happens. Another analogy is a 4 lane highway is to asynchronous as a one-way street is to synchronous. If one car breaks down on the 4 lane highway, the rest can keep moving at their normal speed -- but if one breaks down on the one-way road, everything freezes. :)
Here I leave you a good graphic to see clearly the behavior differences between the synchronous and asynchronous application models:
(source: adaptivepath.com)
It would appear that you have a job grasp of how AJAX works. I can't see much to disagree with in your summary of the plumbing of an AJAX application.
I would say however that with the XMLHttpRequest object you aren't restricted to GET. You can also use POST and other HTTP verbs.
With async calls you register a callback function, the XMLHttpRequest object calls your method when the async request completes.
Seems ok to me.
Your first point though is not entirely correct, you can request html from the server using ajax is doesn't have to text, json or xml like most examples show.
I am making a request to a CGI program using AJAX. The response sends me content-length. My purpose is to dynamically exhibit the response progress. For that I need to start a function on onreadystate value of XHR object to be 3. But the request doesn't seems to acquire that status number. Instead it goes directly from state 1 to state 4.
What am I missing?
The response could be going so quickly that you just don't notice it at state 3. Especially if you are running it on localhost, the response could be transmitted very quickly. You could try setting an alert when it gets to stage 3 to test whether it's actually getting there. Also, I belive internet explorer says that it is a mistake to access the response in stage 3 so there could be compatibility issues.
If you're running on localhost, then probably the browser is never getting a chance to run between the time it sends the request and the time it gets the response...
browser opens connection, sets readyState to 1
browser sends packet to server process
server process receives packet, gets priority from scheduler
server returns data to browser, and yields control of the CPU. Browser continues execution.
browser sees all data has been received, sets readyState to 4.
Long story short: don't count on going into the "receiving" state.