What to have in mind when building a AJAX-based webapp - php

We're in the first steps of what will be a AJAX-based webapp where information and generated HTML will be sent backwards and forwards with the help of JSON/POST techniques.
We're able to get the data out quickly without putting to much load on the database with the help of a cache-layer that features memcached as well as disc-based cache. Besides that - what's essential to have in mind when designing AJAX heavy webapps?
Thanks a lot,

Probably the best thing to have in mind is that your app shouldn't be AJAX-based. It should work fine if the user's browser has scripts disabled. Only then should you start layering on AJAX. Stackoverflow is a great example of this. AJAX really improves the experience but it works when it's disabled.
Another thing I like to do is to use the same PHP validation functions for both server-side and client-side validation (as in sending an AJAX request to a script containing the same PHP function) to keep the amount of cross-language code duplication to a minimum.
Read up on Degradable AJAX.

Security for one. JavaScript has a pretty notoriously bad security profile.

These are the two that always get me:
What happens when the user clicks multiple items that may trigger multiple requests that may return out of order?
What happens when a request doesn't come back for one reason or another (timeout, server problem, etc.)? It always happens eventually, and the more gracefully your system fails the better.

Related

Push data to page without checking periodically for it?

Is there any way you can push data to a page rather than checking for it periodically?
Obviously you can check for it periodically with ajax, but is there any way you can force the page to reload when a php script is executed?
Theoretically you can improve an ajax request's speed by having a table just for when the ajax function is supposed to execute (update a value in the table when the ajax function should retrieve new data from the database) but this still requires a sizable amount of memory and a mysql connection as well as still some waiting time while the query executes even when there isn't an update/you don't want to execute the ajax function that retrieves database data.
Is there any way to either make this even more efficient than querying a database and checking the table that stores the 'if updated' data OR tell the ajax function to execute from another page?
I guess node.js or HTML5 webSocket could be a viable solution as well?
Or you could store 'if updated' data in a text file? Any suggestions are welcome.
You're basically talking about notifying the client (i.e. browser) of server-side events. It really comes down to two things:
What web server are you using? (are you limited to a particular language?)
What browsers do you need to support?
Your best option is using WebSockets to do the job, anything beyond using web-sockets is a hack. Still, many "hacks" work just fine, I suggest you try Comet or AJAX long-polling.
There's a project called Atmosphere (and many more) that provide you with a solution suited towards the web server you are using and then will automatically pick the best option depending on the user's browser.
If you aren't limited by browsers and can pick your web stack then I suggest using SocketIO + nodejs. It's just my preference right now, WebSockets is still in it's infancy and things are going to get interesting once it starts to develop more. Sometimes my entire application isn't suited for nodejs, so I'll just offload the data operation to it alone.
Good luck.
Another possibility, if you can store the data in a simple format in a file, you update a file with the data and use the web server to check its timestamp.
Then the browser can poll, making HEAD requests, which will check the update times on the file to see if it needs an updated copy.
This avoids making a DB call for anything that doesn't change the data, but at the expense of keeping file system copies of important resources. It might be a good trade-off, though, if you can do this for active data, and roll them off after some time. You will need to ensure that you manage to change this on any call that updates the data.
It shares the synchronization risks of any systems with multiple copies of the same data, but it might be worth investigating if the enhanced responsiveness is worth the risks.
There was once a technology called "server push" that kept a Web server process sitting there waiting for more output from your script and forwarding it on to the client when it appeared. This was the hot new technology of 1995 and, while you can probably still do it, nobody does because it's a freakishly terrible idea.
So yeah, you can, but when you get there you'll most likely wish you hadn't.
Well you can (or will) with HTML5 Sockets.
This page has some great info about this technology:
http://www.html5rocks.com/en/tutorials/websockets/basics/

AJAX exposing the API of a web app

A couple of years ago, before I knew about Stack Overflow, I was working in an office with a lot of competition between the programmers. There, I had to code a web page in PHP with Drupal, that needed to get data from another site by RSS. What happened was that there was no way to get the data beforehand: the data depended on the content of the page which itself was dynamic, so the page stopped loading for a couple of seconds while PHP went to get the RSS data. That was bad. The page depended on a couple of parameters out of a huge list. So fetching all possible combinations in davance was out of the questions. It was some sort of search page, that included the results of a sister site, I think.
The first thing I did to improve that was to set up a caching system. When the page was loaded, it launched a Javascript method that saved the RSS data back into the database for this specific page, using AJAX. That meant that if the same page was requested again, the old data would be sent immediately. and the AJAX script would get the cache updated with the new data, if needed. The Javascript pretty much opened a hidden page on the site with a GET instruction that matched the current page's parameters. It's only a couple of days later that I realised that I could have cached the data without the AJAx. (Trust me, it's easier to spot in hindsight.) But that's not the issue I'm asking about.
But I was told not to do any caching at all. I was told that my AJAX page "exposed the API". That a malicious user could hit the hidden page again and again to do a Denial of Service attack. I thought my AJAX was a temporary solution anyway, but that caching was needed. But mostly: wasn't the DoS argument true of ANY page on the site? Did the fact that my hidden page did not appear in the menus and returned no content make it worse?
As I said, there was a lot of competition between programmers, so the people around me, who were unanimous, might have been right, or they may have tried to stop me from doing something that was bad because they were not the ones doing it. (It happened a lot.) But I'm still curious. I was fully aware that my AJAX thing was a hack. I wanted to change that system as soon as I found something better, but I thought that no caching at all was even worse. Which was true? Doesn't, by that logic, ALL AJAX expose the API? If we look past the fact that my AJAX was an ugly hack, was it really that dangerous?
I'll admit again and again that it was an ugly, temporary fix, but my question is about having a "hidden" page that returns no content that makes the server do something. How horrible is that?
both sides are right. Yes, it does "expose" the api, but ajax requests can only access publicly accessible documents/scripts in the first place, so yes, all ajax requests "expose" their target script in the same way. DoS attacks are not script specific, they are server specific, so one can perform a DoS using anything pointing to the server, not just this script your ajax calls. I would tell your buddies their argument is weak and grasping at straws, and don't be jealous :P
If I read your post correctly, it seems as if the AJAX requested version of the page would know to invalidate the cache each time?
If that's the case, then I suppose your co-worker might have been saying that the hidden page would be susceptible to a DDOS attack in a way that the full pageload wasn't. I.E. The full pageload would get a cached version on each pageload after the first, where as the AJAX version would get fresh content each time. If that's the case, then s/he's right.
By "expose the API", your co-worker was saying that you were exposing the URL of a page that was doing work that should be done in the background. The outside world should not know about a URL whose sole purpose is to do some heavy lifting task. As you even said, you found a backend solution that didn't require the user's browser knowing about your worker process at all.
Yes, having no cache at all when the page relies on heavy content is worse than having an ajax version of the page do the caching, but I think the warning from your coworker was that no page, EVEN if it's AJAX, should have the power to break the cache in a way you didn't expect or intend.
The only way this would be a problem is said "hidden page that returns no content that makes the server do something" had different authentication scheme or permissioning from the rest of the pages, or if what it made the back-end do would be inordinately heavy compared to any other page on the site that posted something.

Client-side or server-side processing?

So, I'm new to dynamic web design (my sites have been mostly static with some PHP), and I'm trying to learn the latest technologies in web development (which seems to be AJAX), and I was wondering, if you're transferring a lot of data, is it better to construct the page on the server and "push" it to the user, or is it better to "pull" the data needed and create the HTML around it on the clientside using JavaScript?
More specifically, I'm using CodeIgniter as my PHP framework, and jQuery for JavaScript, and if I wanted to display a table of data to the user (dynamically), would it be better to format the HTML using CodeIgniter (create the tables, add CSS classes to elements, etc..), or would it be better to just serve the raw data using JSON and then build it into a table with jQuery? My intuition says to do it clientside, as it would save bandwidth and the page would probably load quicker with the new JavaScript optimizations all these browsers have now, however, then the site would break for someone not using JavaScript...
Thanks for the help
Congratulations for moving to dynamic sites! I would say the following conditions have to be met for you to do client-side layout (it goes without saying that you should always be doing things like filtering DB queries and controlling access rights server side):
Client browser and connection capabilities are up to snuff for the vast majority of use cases
SEO and mobile/legacy browser degradation are not a big concern (much easier when you synthesize HTML server side)
Even then, doing client-side layout makes testing a lot harder. It also produces rather troublesome synchronization issues. With an AJAX site that loads partials, if part of the page screws up, you might never know, but with regular server-side composition, the entire page is reloaded on every request. It also adds additional challenges to error/timeout handling, session/cookie handling, caching, and navigation (browser back/forward).
Finally, it's a bit harder to produce perma-URLs in case someone wants to share a link with their friends or bookmark a link for themselves. I go over a workaround in my blog post here, or you can have a prominent "permalink" button that displays a dynamically rendered permalink.
Overall, especially when starting out, I would say go with the more kosher, better supported, more tutorialed, traditional approach of putting together the HTML server side. Then dip in some AJAX here and there (maybe start out with form validation or auto-completion), and then move on up.
Good luck!
It is much better to do the heavy lifting on the server side.
In CodeIgniter you create a view, looping through all the rows in the table adding in the classes or whatever else you would need. There is no reason at all to do this in Javascript.
Javascript is a sickly abused language with unfortunate syntax. Why on earth would you want to load a page, and then issue a AJAX call to load up some JSON objects to push into a table is beyond me. There is little reason to do that.
Javascript (and jQuery) is for end user enhancement. Make things slide, flash, disappear! It is not for data processing in even the mildest of loads. The end user experience would be crap because you're relying on their machine to process all the data when you have a server that is infinitely more capable and even designed for this specifically.
It depends on your target market and the goal of your site.
I strongly believe in using the client side where ever you can to offload work from the server. Obviously its important you do it correctly so it remains fast for the end user.
On sites where no-js support is important (public websites, etc), you can have fallbacks to the server. You end up doubling code in these situations but the gains are very beneficial.
For advanced web applications, you can decided if making JS a requirement is worth the trade of losing a (very) few users. For me, if I have some control over the target market, I make it a requirement and move on. It almost never makes sense to spend a ton of time to support a small percentage of potential audience. (Unless the time is spent on accessibility which is different, and VERY important regardless of how many people fit into this group on your site.)
The important thing to remember, is touch the DOM as little as possible to get the job done. This often means building up an HTML string and using a single append action to add it to the page vs looping through a large table and adding one row at a time.
It's better to do as much as possible on the server-side because 1) you don't know if the client will even have JavaScript enabled and 2) you don't know how fast the client-side processing will be. If they have a slow computer and you make them process the entire site, they're going to get pretty ticked off. JavaScript/jQuery is only supposed to be used to enhance your site, not process it.
You got the trade-off correctly. However, keep in mind that you can activate compression in the server side, which will probably make adding repetitive markup to format the table a small bandwidth cost.
Keep also in mind that writing Javascript that works in all browsers (including hand-helds) is more complicated than doing the same server side in PHP. And don't forget that the "new JavaScript optimizations" do not apply to the same extent to browsers of handheld devices.
I do a great deal of AJAX app development and I can tell you this from my experience. a good balance between the two is key.
do the raw data server-side but use javascript to make any modifications that you would need to it. such as paging, column sorting, row striping, etc.
I absolutely love doing everything in AJAX heh.. but there are some short falls to doing it using AJAX, and that's SEO. search engines do not read javascript, so for the sake of your website's page rank, I would say have all data served up server side and then formatted and made look cool client-side.
The reason I love AJAX so much is because it drastically speeds up your APP usage by the user as it only loads the data you need to load where you need to load it, versus load the entire page every time you do something... you can do a whole bunch of stuff, such as hide/show rows/columns (we are talking about table manipulation here because you mentioned a table) and even with these show/hide actions add delete actions where when you click a delete row or button it deletes that row not only visually but in the database all done via AJAX calls to server-side code.
in short.
raw data: server-side sending to the client the raw data in html layout (tables for table structured data, however I do everything else in divs and other flexible html tags, only do tables for column/row style data)
data formatting: client-side which also includes any means of interacting with the data. adding to it, deleting from it, sorting it differently etc. This achieves two things. SEO, and User Experience (UX).

When is it appropriate to use AJAX?

When is it appropriate to use AJAX?
what are the pros and cons of using AJAX?
In response to my last question: some people seemed very adamant that I should only use AJAX if the situation was appropriate:
Should I add AJAX logic to my PHP classes/scripts?
In response to Chad Birch's answer:
Yes, I'm referring to when developing a "standard" site that would employ AJAX for its benefits, and wouldn't be crippled by its application. Using AJAX in a way that would kill search rankings would not be acceptable. So if "keeping the site intact" requires more work, than that would be a "con".
It's a pretty large subject, but you should be using AJAX to enhance the user experience, without making the site totally dependent on it. Remember that search engines and some other visitors won't be able to execute the AJAX, so if you rely on it to load your content, that will not work in your favor.
For example, you might think that it would be nice to have users visit your blog, and then have the page dynamically load the newest article(s) with AJAX once they're already there. However, when Google tries to index your blog, it's just going to get the blank site.
A good search term to find resources related to this subject is "progressive enhancement". There's plenty of good stuff out there, spend some time following the links around. Here's one to start you off:
http://www.alistapart.com/articles/progressiveenhancementwithjavascript/
When you are only updating part of a page or perhaps performing an action that doesn't update the page at all AJAX can be a very good tool. It's much more lightweight than an entire page refresh for something like this. Conversely, if your entire page reloads or you change to a different view, you really should just link (or post) to the new page rather than download it via AJAX and replace the entire contents.
One downside to using AJAX is that it requires javascript to be working OR you to construct your view in such a way that the UI still works without it. This is more complicated than doing it just via normal links/posts.
AJAX is usually used to perform an HTTP request while the page is already loaded (without loading another page).
The most common use is to update part of the view. Note that this does not include refreshing the whole view since you could just navigate to a new page.
Another common use is to submit forms. In all cases, but especially for forms, it is important to have good ways of handling browsers that do not have javascript or where it is disabled.
I think the advantage of using ajax technologies isn't only for creating better user-experiences, the ability to make server calls for only specific data is a huge performance benefit.
Imagine having a huge bandwidth eater site (like stackoverflow), most of the navigation done by users is done through page reloads, and data that is continuously sent over HTTP.
Of course caching and other techniques help this bandwidth over-head problem, but personally I think that sending huge chunks of HTML everytime is really a waste.
Cons are SEO (which doesn't work with highly based ajax sites) and people that have JavaScript disabled.
When your application (or your users) demand a richer user experience than a traditional webpage is able to provide.
Ajax gives you two big things:
Responsiveness - you can update only parts of a web page at a time if need be (saving the time to re-load a page). It also makes it easier to page data that is presented in a table for instance.
User Experience - This goes along with responsiveness. With AJAX you can add animations, cooler popups and special effects to give your web pages a newer, cleaner and cooler look and feel. If no one thinks this is important then look to the iPhone. User Experience draws people into an application and make them want to use it, one of the key steps in ensuring an application's success.
For a good case study, look at this site. AJAX effects like animating your new Answer when posted, popups to tell you you can't do certain things and hints that new answers have been posted since you started your own answer are all part of drawing people into this site and making it successful.
Javascript should always just be an addition to the functionality of your website. You should be able to use and navigate the site without any Javascript involved. You can use Javascript as an addition to existing functionality, for example to avoid full-page reloads. This is an important factor for accessibility. Javascript should never be used as the only possibility to reach or complete a request on your site.
As AJAX makes use of Javascript, the same applies here.
Ajax is primarily used when you want to reload part of a page without reposting all the information to the server.
Cons:
More complicated than doing a normal post (working with different browsers, writing server side code to hadle partial postbacks)
Introduces potential security vulnerabilities (
You are introducing additional code that interacts with the server. This can be a problem on both the client and server.
On the client, you need ways of sending and receiving responses. It's another way of interacting with the browser which means there is another point of entry that has to be guarded. Executing arbritary code, posting data to a non-intended source etc. There are several exploits for Ajax apps that have been plugged over time, but there will always be more.
)
Pros:
It looks flashier to end users
Allows a lot of information to be displayed on the page without having to load all at the same time
Page is more interactive.

Adding some custom session variables to a JavaScript object

I currently have a custom session handler class which simply builds on php's session functionality (and ties in some mySQL tables).
I have a wide variety of session variables that best suits my application (primarily kept on the server side). Although I am also using jQuery to improve the usability of the front-end, and I was wondering if feeding some of the session variables (some basics and some browse preference id's) to a JS object would be a bad way to go.
Currently if I need to access any of this information at the front-end I do a ajax request to a php page specifically written to provide the appropriate response, although I am unsure if this is the best practice (actually I'm pretty sure this just creates a excess number of Ajax requests).
Has anyone got any comments on this? Would this be the best way to have this sort of information available to the client side?
I really guess it depends on many factors. I'm always having "premature optimization ..." in the back of my head.
In earlier years I rushed every little idea that came to my mind into the app. That often lead to "i made it cool but I didn't took time to fully grasp the problem I'm trying to solve; was there a problem anyway?"
Nowadays I use the obvious approach (like yours) which is fast (without scarifying performance completely on the first try) and then analyze if I'm getting into problems or not.
In other words:
How often do you need to access this information from different kind of loaded pages (because if you load the information once without the user reloading there's probably not much point in re-fetching it anyway) multiplied by number of concurrent clients?
If you write the information into a client side cookie for fast JS access, can harm be done to your application if abused (modified without application consent)? Replace "JS" and "cookie" without any kind of offline storage like WHATWG proposes it, if #1 applies.
The "fast" approach suits me, because often there's not the big investment into prior-development research. If you've done that carefully ... but then you would probably know that answer already ;)
As 3. you could always push the HTML to your client already including the data you need in JS, maybe that can work in your case. Will be interesting to see what other suggestions will come!
As I side note: I've had PHP sessions stored in DB too, until I moved them over to memcached (alert: it's a cache and not a persistent store so may be not a good idea for you case, I can live with it, I just make sure it's always running) to realize a average drop of 20% of database queries and and through this a 90% drop of write queries. And I wasn't even using any fancy Ajax yet, just the number of concurrent users.
I would say that's definately an overkill of AJAX, are these sessions private or important not to show to a visitor? Just to throw it out there; a cookie is the easiest when it comes to both, to have the data in a javascript object makes it just as easily readable to a visitor, and when it comes down to cookies being enabled or not, without cookies you wouldn't have sessions anyway.
http://www.quirksmode.org/js/cookies.html is a good source about cookie handling in JS and includes two functions for reading and writing cookies.

Categories