Client side Callback in GWT - java

I'm trying to create a logger for a GWT application as an exercise to evaluate GWT. What I specifically want to do is have it so that I can post messages to a client side label at any point from the server side. So, if some interesting stuff has happened on the server the client can be updated.
My First question is, is this possible, I can understand it not being.
Second, if it is possible, where should I look for information, I've tried google and their documentation and all the showcases have nothing on this.
Thanks

Well, there are a couple of Options. You need to get the data from the server... So you either need to poll the server, or use server push.
Polling is pretty easy. Just use the Timer class to repeatedly call a service to see what value it should be displaying.
Server push is done using something like comet. here is one implementation for gwt that looks somewhat promising. They basic concept behind this is the browser sends a request to the server and keeps the connection open so the server is free to keep sending data back.
Comet is the better option if you can get it working. It will probably be simpler and scale better.
Good Luck!

Polling is probably the best way to do what you're looking for. The big gotcha with GWT is that everything has to be serializable. I'm not sure if anything that can push to the browser can be easily serialized.

Related

Log all request and response data in tomcat

At every moment in the world things change, said some guy.
We're sitting on a big fat REST API server which is becoming more and more of a burden to maintain because good Java developers are hard to find, expensive, development takes much more time, so we're slowly switching various application parts to Rails, which is... well, nevermind.
There are way too many APIs (and some are poorly documented) to just clone functionality by looking at the docs, so my idea is to find a way to log everything that's passing through this service for a while, in a machine-readable format we can later index, process and report, so we can gradually rewrite the code.
Is there a "clean" not so "hacky" way of doing this with Tomcat? Or should I go down the wireshark/reverse proxy way?
TIA!
You can create a payload logging filter and configure it in your web.xml. You can also use an output filter for the response. This and this should be helpful working examples.

Database change notification app

I'm about to embark upon writing an android app which notifies the phone's user when an external mySQL DB is updated (add only) with a ticket so that the user can check if the ticket requires his attention (an attempt to reduce the buildup of tickets that he has to trawl through).
From my research, most questions suggest using a PHP web service with my program (written in java) and definitely/maybe/definitely not/it's deprecated using SQLNotification to fire the event. I've also seen something about some bloke called JSON and the brands of SOAP he uses.
What I've been unable to figure out is how all of these frameworks/toolkits/services/things work together.
My question is in two parts:
Is SQLNotification usable? If not, is there a simple way to check for changes (beyond the obvious answer of polling)
How does everything (SOAP, JSON, web service, app) fit together and have I missed anything on the frameworks front (Heard mentions of spring, hibernate, tomcat).
On my experience, I'm relatively fluent in Java, understand the basics of MySQL, am a beginner in PHP and haven't written for android before.
Thanks,
Ben
According to the information you have provided, seems like you have different options:
Push
If you want to go the push way, you will need some central architecture that can take notifications from the database and immediately send them to connected clients. It's not that easy to build such a scheme, would only recommend it if you really need immediate notification. As a start point, look at this sample: http://www.gianlucaguarini.com/blog/push-notification-server-streaming-on-a-mysql-database/
Pull
If you go the pull way (polling), you can have a service working on the phone wich polls every configured time. In that case you will need some stateless service, some simple JSON service would do great.
On both cases, be careful with security, you should secure your channel with ssl and have some decent sort of authentication.
I think a good rule of thumb is saying (it's just based on personal opinion/experience, maybe your decision path has other factors that you have to consider), if your pollig intervals don't need to be lower than 5 minutes, polling will do fine. If you need almost real-time notifications, you can implement the push architecture, but you have to know it will cost more efford to get it working as you have to take care of things like client disconnection, how to handle notifications if a client is not connected, get the real-time notifications from the database, etc.
Hope this helps as a starting point,

http server in php

I want to create http socket connection for server in php for multiple client . how can I do that ? I need some resource .
First I was trying to create server in java .I create a server in java . And trying to reach from android application .But server can't find any client.But when I create client in java .It was working. How can I solve that problem ???
Take a look at this article:
Writing Socket Servers in PHP by Zend
Also give a try with Google:
http://www.google.com/search?aq=0&oq=php+socket+server+mul&sourceid=chrome&ie=UTF-8&q=php+socket+server+multiple+clients
Personally I think this would be a pretty bad idea, as already mentioned it lacks Threading and it's Socket support (imo) isn't really that adaptable.
The only plus side is that you can use fork to fork off another PHP process to handle a client, but you're getting very complex.
Another language would be much more suited for this type of development.
Note that even if you did do this in PHP, you'd probably have to rely on external services anyway, and possibly even end up writing at least some code in another language anyway.
You're trying to use PHP to do what? Mind you, I like PHP and work with it almost every day, but please do remember PHP in and on itself is based on request and response, and not very suitable for long running processes. In a manner of exercise, it might be interesting, but if you're trying to write a webserver from scratch using PHP, you might want to reconsider your choice of language.
That said, you can create a socket acting as a server, and listen to incoming packets. I do still think you're reinventing the wheel though.
though i love php and java, i wrote my socket servers in c++ running under lamp in an amazon ec2 cloud server. it is very, very simple to code and debug and safe and you can practically just copy/paste examples.
in the long run, i will probably develop a java solution because of portability and scalability, but the initial effort to get a c++ solution working is just so much less than implementing a java solution...
the first thing you must ascertain (find out) is whether your server allows you to open custom ports. amazon ec2 does and at this point in time (feb13), can be used for free for 12 months.
so, this is for you if you are in a hurry:
this here set of examples has all that you need to be up and running in no time.
Judging from the question title (the rest only makes it more confusing) you could use an existing package like http://pear.php.net/package/HTTP_Server to implement a webserver in PHP. It already contains all the socket code to accept client connections and stuff.
So what i have to do to find the server from different client
"Finding" is too broad a topic. Depends on your actual setting. On a LAN there are some protocols for discoverability. Otherwise you should just rely on a fixed machine name and port number for your instantiated server. You can connect to it as e.g. http://localhost:8007/ or whatever you've predefined.

What is the best way to implement a website where Users can interact together

I am creating a website where users will be able to chat and send files to one another through a browser. I am using GWT for the UI and hibernate with gilead to connect to a mysql database backend.
What would be the best strategy to use so Users can interact together?
I'd say you are looking for comet/AJAX|Server push/etc. See my previous answer on this matter for some pointers. Basically you are simulating inverting the communication between server and client - it's the server that's initiating the connection here, since it wants to, for example, inform the user that his/her friend just went online, etc.
The implementations of this technique change quite rapidly, so I won't make any definitive recommendations - choose the one that best suits your needs :)
COMET is the technology that allows chatting over a web page - It is basically communicating through keep-alive connections. This allows servers to push information to the client.
There are several implementations of this on the client side with GWT.
Most servers nowadays support this, It is also part of the Servlet 3.0 spec (Which nobody has implemented yet)
While COMET is very nice, it's not the only solution! Usual polling with time intervals (as opposed to COMET long polling) is still commonly used. It's also possible to require a manual refresh by the user.
Take Stackoverflow as an example - for most things you must refresh your browser manually to see the changes. I think, it's commonly perceived as normal and expected. COMET or frequent polling are an added bonus.
The problem with COMET is, that it can easily lead to lots of threads on the server. Except, if you additionally use asynchronous processing (also called "Advanced IO"), which is not too well supported yet (e.g. doesn't work with HTTPS in Glassfish v3 due to a severe bug), can lead to problems with Apache connectors etc.
The problem with frequent polling is, that it creates additional traffic. So, it's often necessary to make the polling less frequent, which will make it less convenient for the end user.
So you will have to weigh the options for your particular situation.

Java applet communication with Rails application

I'm creating a Rails application and on it, there should be a Java Applet.
My question and problem is that the applet must be tightly integrated with the Rails parts. I must be able to get a list of all users, update an image, etc... And there's a surprisingly small amount of information available on the Internet of how to use applets with Rails. So please give me some hints. What is the best way to do it?
Send parameters to the applet?
Use Rails REST interface from the applet?
Use JRuby somehow?
Other....?
Thanks!
Can you provide more details? In the meantime, here's my take on your questions:
Send parameters to the applet?
Your rails app will be able to serve the applet, but once served I don think you'll be able to send messages to it (however you will be able to respond to messages from it, which is perhaps what your asking).
Use Rails REST interface from the applet?
You've kind of answered that one yourself. REST is an interface design and therefore can be accessed from anything that can issue a HTTP request. The trick is to correctly construct the URL so rails knows what you want to do. There's good info on configuring rails routes (REST and non-REST) here http://guides.rubyonrails.org/routing.html
Use JRuby somehow?
You could use jruby for this, but you dont need to. Your server (rails) and your client (browser/applet) talk to eachother via http and so don't need to be the same language or run on the same VM.
Hope that help....
I think the reason you probably haven't found anything specific about Applets and Rails is that they function a bit at different levels and aren't really dependent on one another. It looks like Rob was trying to clarify a few things, so I'll take it another step just to be sure we are all on the same page.
The job of Rails is to generate and serve up HTML/XML/Javascript/images via a web server to the user's web browser for rendering. Part of that HTML will be an APPLET (or possibly OBJECT) tag, which instructs the browser to load the applet. Usually, this instructs the browser to invoke the Java Plug-in and lets it handle loading the applet. Once loaded and running, however, even though the applet is displayed on the current web page in the browser (or maybe in another window even), it really isn't terribly aware of the web page it is sitting in. For the most part, applets don't care about the browser or the page they are "part of". So if an applet needs more information, or needs to ask for data, it usually will just send an HTTP request to the server it came from. It would then parse the data and update itself.
I am assuming what you probably need is for something to be clicked or entered into the applet, and that data be used to update the web page that Rails is serving to the browser. With an applet, you pretty much have 2 options:
Use the web server application to share state information
Use the Java-to-Javascript communication using JSObject as indicated at http://java.sun.com/products/plugin/1.3/docs/jsobject.html
Honestly, option number 2 comes with so many caveats, that I would never use it unless you had complete control over browser and Java versions on all potential users' systems. Even then, I'd be concerned of an update to something breaking it.
Basically, option number 1 leaves you with the Applet and the Javascript/HTML polling the web server (Rails) periodically to see if there are any updates or requests that they need to respond to for data exchange. If the applet is updated by the user, it sends a message to the web server via a URL request/post and the next query (probably via an AJAX-like call) by the web page will see the new data and the web page will be updated with it.
I hope that helps.
Two really great answers. I appreciate them a lot, thanks!
Reading your posts made me realize that the best choice is to use HTTP-requests to Rails REST interface. However, I see some downsides with this approach. But I don't see any better solution to it. One feature the applet should have, is to be able to browse and search in all products, which can be quite many. Sending a HTTP-request for each search will be expensive. Maybe I could solve this by loading all products when the applet starts. Then the browsing and searching would be fast. Or maybe do some nice caching. So once they are found, I don't fetch them again.
About not finding lots of information about this on the net. I see your point monceaux. But... I still think that there should be more. I mean, in my situation I would really like a Rails specific library that helped me send requests to correct urls. To bad Java is not that dynamic though. Kind of hard to do some stuff automatically, like in Ruby and Rails. Maybe I'll write a small library for this. I mean, I must write it anyway. So why not make a library of it? Some people might have use of it.

Categories