Andrzej Maslowski

Thanks for the reply, Michał.
I guess I don't have enough FP experience to go beyond lambdas yet. Your response tells me a lot about what I don't know:)
As it seems you really know your trade, I'm really looking forward to reading the book and expand my toolset.

Hi Michał Płachta, everyone.
Great to see another Grokking series book.
In the book introduction, you wrote "functional programming and writing maintainable software". As probably many of us, I've seen some unmaintable lambdas in my life. Code so convoluted, that it could be considered an entry for Obfuscated Code Contest.
What measures can be taken to enforce functional programming bits written in a readable way, promoting conciseness while still benefittng from the new paradigm?
It also seems that many enter this brave new world, but not everyone stay. Especially in java, some early adopters tend to write as much lambdas as they can, but later they withdraw and try to find balance. Repositories around the world could tell a lot about this kind of evolution. What course should be taken here for junior developers - should they be encourage to write as much FP as possible?
Do you have any hints that tackle this?

You already have SCJP and SCWCD, so you know how these exams work. SCBCD is not much different. I'd say four weeks should be enough if you are determined, even less if you know something about the subject.
Of course Hebert has the point here: only you and mocks can tell how close you are to achieving the goal.

Thank you all for your kind words.

Sudhansu Maharana wrote:
But I have scored 48% in whizlabs simulator practice test. Can you little guide me is this ok to just pass the OCJP 6 or i should prepare more before taking the actual exam. I have studied K&B book twice. (Is the Whizlabs simulator questions are harder than the real exam questions?)

48% is a bit too low. How was your final mock from K&B book?

After a year of postponing it, almost on the voucher deadline, I passed the 310-091, which in the meantime changed its name to 1Z0-860.
Long story short - I started to learn about four weeks ago, by skipping through the Manning's "EJB in Action". I considered it to be too exam-unrelated, so I went on to read Wiley's "Mastering EJB3 4th Edition".
I found it much more relevant and suitable to my certification needs.
About a week ago I started learning using sample questions as the basis. I used all mocks listed in the Ranch's links list for SCBCD, all free ones, and some demo versions. All doubts I had were quickly resolved by Mikalai Zaikin's Study Notes (thank you!) and EJB specification.

Just before the exam I came across the Oracle's sample questions for 1Z0-860. I was terrified by my results - just 50%... Maybe I was tired, I don't know... Question in section 8.8 has its own thread here on java ranch, question 3.3 has errors - it confuses stateless and stateful beans...
Now good news for everyone who wants to pass this exam: there are no such questions on the real exam. They are not so puzzled, they do not contain so many errors or ambiguities. In fact, most of the questions are quite easy if you know a bit about the subject. In my opinion it was easier than SCWCD.

So, thank you all for providing information on mocks and recommended books.

Good luck to all future SCBCD/OCPJBCD exam takers.

Andy

Thank you, guys, I really appreciate it.

Next stop: SCBCD

I passed the test today, finally. I've been preparing for this for months, not because it was so hard, but because I had so many other things to do.
In fact, during last three weeks I focused really on the exam.

As for the preparation, I read HF JSP twice (thanks Bert, Kathy & Bryan;)), did some online mocks (mainly the ones linked here, on javaranch), and studied APIs - servlet-2_4-fr-spec, jsp-2_0-fr-spec and jstl-1_1-mr2-spec. For my own testing purposes, I used Tomcat5.5 servlet container with Eclipse as IDE.

I must say that I'm overwhelmed by the difference between SCJP and SCWCD - in SCJP everything was clear and obvious; in SCWCD there's so much ambiguity, that even exam authors have some problems with proper approach to the presented subjects.
Three of the questions I got today on the exam, where, I must say, badly constructed. I can't remember the exact question (you can't get any notes out of the exam center), but it went something like this:
Q: You have a List object, stored as a request attribute, what is the proper way of retrieving it?
A: List ar = request.getAttribute("list");
and some other answers, all of which lacking the cast to List type. In fact, all the answers were wrong.
Another question involved autoboxing feature, that was not supposed to be tested on SCWCD5.

The same applies to the HF JSP book - look at all these errata points: some of them confirmed, some still unconfimed yet. I don't believe that it is a matter of chaotic book writing (the book itself is very good, and does its job very well). I think it is a general problem with Servlet specification. All these inconsistancies, piles of rules, and even more exceptions from the rules. I would like to quote James E. Rolfe here (Angry Video Game Nerd):
"WHAT WHERE THEY THINKING?!"

Ok, enough complaining. Some thanks following:
- Bert Bates, Kathy Sierra and Bryan Basham for their HF JSP book
- javaranch people for all the questions I didn't have to ask
- Combidata Poland (Gdynia) Test Center people which were very friendly and helpful

Good luck to all future SCWCD exam takers.

Andy

Hi Chinmaya,
thanks for your answer.

A filter's doFilter() method must call doFilter() on the input FilterChain object in order to ensure that all filters have a chance to execute.

I think its correct.

So we're both positive about the subject. I found that someone has posted an errata to this question, but it's still in 'unconfirmed' state.
page 735: http://oreilly.com/catalog/errataunconfirmed.csp?isbn=9780596516680

The only way a filter can be invoked is through a declaration in the DD.

It is true. When client made request, the container decides which filter will call first and which is second and third etc. According to this mapping the FilterChain's doFilter() method call's the next filter.

Ok, I get your point. I wasn't that sure by the time of taking this mock though. I hope questions will be clearer on the final exam.

Excuse me for digging up an old topic, but I'm currently in the process of learning to the SCWCD. I have the same doubt - why is A the correct answer? Duc explained, why B isn't correct - that's fine.
But how can you say, that "Filters may be used to create request or response wrappers."?
Your hint, Duc, labelled as 3, doesn't help very much. Even if filter code creates wrappers while it's running, it doesn't mean that "Filter may be (or is) used to create wrappers". Let me draw you an analogy:
- Craftsman creates tools to do his work
therefore
- tools are created by a craftsman
Would you say that craftsmen may used to create tools? Shouldn't you rather say that, "wood may be used to create tools"?

This whole question (5/Chapter 13) is so unclear, that really lost my nerve with it. E answer is labelled incorrectly as wrong, even though it is right in fact.
It says clearly that "A filter's doFilter() method must call doFilter() on the input FilterChain object in order to ensure that all filters have a chance to execute."
And it's correct - if you really want to ensure that all filters have a chance to execute, you MUST call doFilter on the FilterChain.

Also question 1 in this chapter is a bit unclear to me. It has an answer D, stating that:
"The only way a filter can be invoked is through a declaration in the DD."
What about invoking a filter through a doFilter(r,r,c) call?