We Keep Coding

why not call service class directly from jsp

So i have a sort of design question:
I have a jsp, and a controller that fetched the data for that jsp. Some of that data come from service classes.
I know that mvc pattern tells me to use the controller to call the service class and pass that info to my view (jsp).
Why can't I call the service class from my jsp directly?

You can, and that's what developers sometimes do. But you shouldn't.
MVC is about interchangeability and separation of concerns. If you call your service from JSP, you create a tight coupling, to parameters and return types, for example.
Moreover, usually, systems are not developed singlehandedly. Let's say you have getAllAdmins() method in your service, which you use for internal logic. Do you really want another developer to use it directly in JSP, and by mistake display all your admins? Probably not.

You can. You can even put everything in one class and maybe it will work. But why? Doing like that ruin all flexibility.
You think only about little example, but you should think what advantages it gives to big applications.
Read this.

Is it a bad practice to use a ThreadLocal Object for storing web request metadata?

I am working on a j2ee webapp divided in several modules. I have some metadata such as user name and preferences that I would like to access from everywhere in the app, and maybe also gather data similar to logging information but specific to a request and store it in those metadata so that I could optionally send it back as debug information to the user.
Aside from passing a generic context object throughout every method from the upper presentation classes to the downer daos or using AOP, the only solution that came in mind was using a threadlocal "Context" object very similar to a session BTW, and add a filter for binding it on ongoing request and unbinding it on response.
But such thing feels a little hacky since this breaks several patterns and could possibly make things complicated when it comes to testing and debugging so I wanted to ask if from your experience it is ok to proceed like this?

ThreadLocal is a hack to make up for bad design and/or architecture. It's a terrible practice:
It's a pool of one or more global variables and global variables in any language are bad practice (there's a whole set of problems associated with global variables - search it on the net)
It may lead to memory leaks, in any J2EE container than manages its threads, if you don't handle it well.
What's even worse practice is to use the ThreadLocal in the various layers.
Data communicated from one layer to another should be passed using Transfer Objects (a standard pattern).
It's hard to think of a good justification for using ThreadLocal. Perhaps if you need to communicate some values between 2 layers that have a third/middle layer between them, and you don't have the means to make changes to that middle layer. But if that's the case, I would look for a better middle layer.
In any case, if you store the values in one specific point in the code and retrieve it in another single point, then it may be excusable, otherwise you just never know what side affects any executing method may have on the values in the ThreadLocal.

Personally I prefer passing a context object, as the fact that the same thread is used for processing is an artifact of the implementation, and you shouldn't rely on such artifacts. The moment you want to use other threads, you'll hit a wall.
If those states are encapsulated in a Context object, I think that's clean enough.

When it comes to testing, the best tool is dependency injection. It allows to inject fake dependencies into the object under test.
And all dependency injection frameworks (Spring, CDI, Guice) have the concept of a scope (where request is one of these scopes). Under the hood, beans stored in the request scoped are indeed associated with a ThreadLocal variable, but this is all done by the dependency injection framework.
What I would do is thus to use a DI framework, which would make request-scope objects available anywhere, but without having to look them up, which would break testability. Just inject a request-scoped object where you want to use it, and the DI framework will retrieve it for you.

You must know that a servlet container can / will re-use threads for requests so if you do use ThreadLocals, you'll need to clean up after yourself once the request is finished (perhaps using a filter)

If you are the only developer in the project and you think you gain something: just do it! Because it is your time. But, be prepared to revert the decision and reorganize the code base later, as should be always the case.
Let's say there are ten developers on the project. Everybody might like to have its thread local variable to pass on parameters like currency, locale, roles, maybe it becomes even a HashMap....
I think in the end, not everything which is feasible, should be done. Complexity will strike back on you....

ThreadLocal can lead to memory leak if we do not set null manually once its out of scope.

Reuse objects among subsequent requests to dynamic Google App Engine backend

on my GAE app, I have a servlet that performs an XSLT transformation. I used to run it as frontent, but sometimes it took too much time to finish. So I'm now running this on the backend.
This is what I did:
1/ create a file 'backends.xml' defining a dynamic public backend named 'xslt'
2/ prepend 'xslt' to the domain when calling the servlet:
http://xslt.[appname].appspot.com/getCoordinates?[params]
This works!
The typical behaviour of the app is that a series of calls to this 'getCoordinates' servlet will be made. Each request will trigger the 'doGet' method of this 'getCoordinates' servlet, which does the initialization of the Saxon processor, xsltCompiler, xsltExecutable and xsltTransformer, but all of these objects could be reused across subsequent requests!
My question: how should I program to separate this initialization code into a handler for the backend initialization request to '_ah/start'?
If I just create another servlet 'startXslt' that answers the request to '_ah/start' and initialize all the generic objects within this servlet's 'doGet' method, how will I be able to use the objects from within the 'getCoordinates' servlet's 'doGet' method?
(I'm not very experienced with java servlet programming, so I reckon this may be more like a general question on java servlet programming, and not GAE-specific, or is it?)

Yes, it's general questions, and there is really a hundreds of ways of doing that. Btw, most projects are based on some framework, and it depends on it. If you're startd with plain raw servlets - i strongly recommend you to take a look at other options. For GAE there is Gaelyk. Or Spring MVC as most populart (is it?) general usage framework.
Btw, if you need an solution right now, I can recommend one of the following:
init in init() method (it will be called on app startup)
store it at class static field, and init in static {} block (called at class initialization, shared between instances)
make an singleton for this transformers (you can init it once, at first call)

Using a backend is a good idea, since you could control that only a single instance would be used and re-used when addressing the backend.
In this way, all servlets would be executed within the same JVM instance, and you could therefore have a shared object by using a Singleton pattern as suggested by splix on the other answer.
As I understand it, your question pertains on how you could hook on the backend initialization to initialize your own objects. If that is the case, you could implement a ServletContextListener and put your code on the contextInitialized(ServletContextEvent) method.
This method would be invoked once every time a new instance is created, be it at the front end or at the back end.

Servlet 3 spec and ThreadLocal

As far as I know, Servlet 3 spec introduces asynchronous processing feature. Among other things, this will mean that the same thread can and will be reused for processing another, concurrent, HTTP request(s). This isn't revolutionary, at least for people who worked with NIO before.
Anyway, this leads to another important thing: no ThreadLocal variables as a temporary storage for the request data. Because if the same thread suddenly becomes the carrier thread to a different HTTP request, request-local data will be exposed to another request.
All of that is my pure speculation based on reading articles, I haven't got time to play with any Servlet 3 implementations (Tomcat 7, GlassFish 3.0.X, etc.).
So, the questions:
Am I correct to assume that ThreadLocal will cease to be a convenient hack to keep the request data?
Has anybody played with any of Servlet 3 implementations and tried using ThreadLocals to prove the above?
Apart from storing data inside HTTP Session, are there any other similar easy-to-reach hacks you could possibly advise?
EDIT: don't get me wrong. I completely understand the dangers and ThreadLocal being a hack. In fact, I always advise against using it in similar context. However, believe it or not, thread context has been used far more frequently than you probably imagine. A good example would be Spring's OpenSessionInViewFilter which, according to its Javadoc:
This filter makes Hibernate Sessions
available via the current thread,
which will be autodetected by
transaction managers.
This isn't strictly ThreadLocal (haven't checked the source) but already sounds alarming. I can think of more similar scenarios, and the abundance of web frameworks makes this much more likely.
Briefly speaking, many people have built their sand castles on top of this hack, with or without awareness. Therefore Stephen's answer is understandable but not quite what I'm after. I would like to get a confirmation whether anyone has actually tried and was able to reproduce failing behaviour so this question could be used as a reference point to others trapped by the same problem.

Async processing shouldn't bother you unless you explcitly ask for it.
For example, request can't be made async if servlet or any of filters in request's filter chain is not marked with <async-supported>true</async-supported>. Therefore, you can still use regular practices for regular requests.
Of couse, if you actually need async processing, you need to use appropriate practices. Basically, when request is processed asynchronously, its processing is broken into parts. These parts don't share thread-local state, however, you can still use thread-local state inside each of that parts, though you have to manage the state manually between the parts.

(Caveat: I've not read the Servlet 3 spec in detail, so I cannot say for sure that the spec says what you think it does. I'm just assuming that it does ...)
Am I correct to assume that ThreadLocal will cease to be a convenient hack to keep the request data?
Using ThreadLocal was always a poor approach, because you always ran the risk that information would leak when a worker thread finished one request and started on another one. Storing stuff as attributes in the ServletRequest object was always a better idea.
Now you've simply got another reason to do it the "right" way.
Has anybody played with any of Servlet 3 implementations and tried using ThreadLocals to prove the above?
That's not the right approach. It only tells you about the particular behaviour of a particular implementation under the particular circumstances of your test. You cannot generalize.
The correct approach is to assume that it will sometimes happen if the spec says it can ... and design your webapp to take account of it.
(Fear not! Apparently, in this case, this does not happen by default. Your webapp has to explicitly enable the async processing feature. If your code is infested with thread locals, you would be advised not to do this ...)
Apart from storing data inside HTTP Session, are there any other similar easy-to-reach hacks you could possibly advise.
Nope. The only right answer is storing request-specific data in the ServletRequest or ServletResponse object. Even storing it in the HTTP Session can be wrong, since there can be multiple requests active at the same time for a given session.

NOTE: Hacks follow. Use with caution, or really just don't use.
So long as you continue to understand which thread your code is executing in, there's no reason you can't use a ThreadLocal safely.
try {
tl.set(value);
doStuffUsingThreadLocal();
} finally {
tl.remove();
}
It's not as if your call stack is switched out randomly. Heck, if there are ThreadLocal values you want to set deep in the call stack and then use further out, you can hack that too:
public class Nasty {
static ThreadLocal<Set<ThreadLocal<?>>> cleanMe =
new ThreadLocal<Set<ThreadLocal<?>>>() {
protected Set<ThreadLocal<?>> initialValue() {
return new HashSet<ThreadLocal<?>>();
}
};
static void register(ThreadLocal<?> toClean) {
cleanMe.get().add(toClean);
}
static void cleanup() {
for(ThreadLocal<?> tl : toClean)
tl.remove();
toClean.clear();
}
}
Then you register your ThreadLocals as you set them, and cleanup in a finally clause somewhere. This is all shameful wankery that you shouldn't probably do. I'm sorry I wrote it but it's too late :/

I'm still wondering why people use the rotten javax.servlet API to actually implement their servlets. What I do:
I have a base class HttpRequestHandler which has private fields for request, response and a handle() method that can throw Exception plus some utility methods to get/set parameters, attributes, etc. I rarely need more than 5-10% of the servlet API, so this isn't as much work as it sounds.
In the servlet handler, I create an instance of this class and then forget about the servlet API.
I can extend this handler class and add all the fields and data that I need for the job. No huge parameter lists, no thread local hacking, no worries about concurrency.
I have a utility class for unit tests that creates a HttpRequestHandler with mock implementations of request and response. This way, I don't need a servlet environment to test my code.
This solves all my problems because I can get the DB session and other things in the init() method or I can insert a factory between the servlet and the real handler to do more complex things.

You are psychic ! (+1 for that)
My aim is ... to get a proof this has stopped working in Servlet 3.0 container
Here is the proof that you were asking for.
Incidentally, it is using the exact same OEMIV filter that you mentioned in your question and, guess what, it breaks Async servlet processing !
Edit: Here is another proof.

One solution is to not use ThreadLocal but rather use a singleton that contains a static array of the objects you want to make global. This object would contain a "threadName" field that you set. You first set the current thread's name (in doGet, doPost) to some random unique value (like a UUID), then store it as part of the object that contains the data you want stored in the singleton. Then whenever some part of your code needs to access the data, it simply goes through the array and checks for the object with the threadName that is currently running and retrieve the object. You'll need to add some cleanup code to remove the object from the array when the http request completes.

It is possible to call a servlet from a Java class?

I'm using IBM WebSphere as my servlet container. My application has several servlets and Java classes. My intent is to call one of those servlets directly from a Java class. Doing some research I figured out that is possible to use the RequestDispatcher interface to achieve this. But it is necessary to pass the objects ServletRequest and ServletResponse as arguments to the method forward(). There is some way to bypass this safely and "nicely"? By "nicely" I meant to say preserving good programming and design patterns.

The only way to do this nicely is to decouple the desired logic from the servlets. This requirement is a sign that the servlets are too tight coupled with business/domain code logic which apparently needs to be used as well outside the webapplication context.
Refactor the original servlet code into reuseable Java class(es) and method(s) (which in turn does not use anything from the javax.servlet package) so that you can finally import and invoke it from both the servlet class and the "plain vanilla" Java class.

It would help to get some more background as to why you are trying to do this. I am assuming you want to invoke some piece of business logic in the servlet. This is a sign that the application is poorly designed.
Are you familiar with MVC architecture? If your "model" code was loosely coupled, you would be able to call it directly.

You could write a filter that stores the current request and response in static ThreadLocal, so that you can use them later from within the same request. You can then implement your own static method forward that uses them and dispatch to another page.
This is somehow the approach taken in JSF where FacesContext.getCurrentInstance can be accessed anytime.
But I wouldn't qualify that as an elegant design. Rather try to follow #BalusC advice and refactor your logic.