Win a copy of Spring Boot in Practice this week in the Spring forum!

yi zhu

Ranch Hand
+ Follow
since Sep 10, 2002
Cows and Likes
Total received
In last 30 days
Total given
Total received
Received in last 30 days
Total given
Given in last 30 days
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by yi zhu

As a family can contain one single member, EJBHome is AbstractFactory.
what 's multiple and dynamic classifications?.
These are used for better reflecting the real world when you do conceptual modeling.
For an of example of multiple classification, a person having cleared SCJP and SCEA is a SCJP and a SCEA. This person has two titles. But there is not a title equivalent to the sum of SCJP and SCEA.
For an example of dynamic classifiation, when a SCJP clears SCEA two years after his obtainment of SCJP, he becomes a SCEA but he is no longer a SCJP.
How to implement this is another story...
With JAVA, I see no way to do multiple classification. But dynamic classification can be mimicked by the design pattern STATE

what 's the difference between specification and implementation diagrams?.
Specification is WHAT and implementation is HOW.
can implementation diagrams can contain aggregation/compostion symbols?
I see no reason stoping you from doing that. It can only make your diagrams easier to understand.
can navigabilty and aggregation/compostion drawn together?.
For composition, there is no need to do that because you alway gain access of a part via the whole
As said in Fowler's book, the difference between aggregation and association is hard to explain. So I suggest you consider using association instead of aggregation if you need to have navigability.
I guess it's a feature provided by a language.
With Java there is no way to change the type of an object at runtime.
I never used Smalltalk. But It seems that In Smalltalk, when you declare a variable, you need no specify the type. So any type can be assigned to this variable. I guess this means dynamique types.
ejbCreate(...) is not mandatory for entity beans!
For statful session beans only the author of beans knows the signature of ejbCreate(...) method.
My answer is D.
A is obviously wrong. If two components are mutually dependent, you should consider to put them together to form a new component.
B is logically wrong. As a component consists of a couple of packages, dependency between the two components does not imply that a package in one component depends on any package in another component.
C is nonsense in my point of view.
68% will be even more exciting!
You can edit this in vendor-specific xml files
I'm a bit dispointed by the behavior of a few people around here. I think it's of our interest to keep SCEA's value. Discussion about assignment should be limited on "WHAT". But talking about "HOW" is definitly not appropriate. Don't miss this chance to prove what you can do.
My $0.02
Hi all,
I guess I understand better your position now. (Thanks to all your explanation)
But could you show me a useful OO feature that is available in Smalltalk but not in Java due to primitive? :roll:
I'm using Tomcat 4.1. The servlet generated from my jsp did not[B] import my tag package. When my tag handler class was used, this servlet used the full qualified class name of my handler class.
However, as mentioned by Sam, this servlet has his own package called [B]org.apache.jsp
. So when your tag handler class has no package declaration, this servlet will look it up in org.apache.jsp and all imported package. This will be obviously fruitless.
You said that both "<" and "++" are actually behaviour of an int - and are implemented as methods in Smalltalk.
But you can also consider i+j or i<j as static methods too. (equivalent to something likeINT.sum(i, j) and, j). And without primitive, how can you build an object?
Well... You are really hard "full" OO GUYS!!!
To make you more comfortable with 1+1, just think it as a static method... and this expression does not tell you the low level implementation behind it. Don't you think it's more than your 1.add(2)?
Sorry, true is primitive too.
It's not about whether the code will run or not.(The answer is yes)
The point here is that EVAL_BODY_BUFFERED has nothing to do with EVAL_BODY_AGAIN despite the fact that they have the same value.
By definition, EVAL_BODY_BUFFERED is returned by doStartTag and EVAL_BODY_AGAIN is returned by doAfterBody. That's why EVAL_BODY_TAG is deprecated!

Originally posted by Chris Mathews
At no time would I make Entity Beans remoteable

Christ and I shared at least this point! I even suggest that local interfaces will always be hidden behind a session facade.

Originally posted by Chris Mathews
You can easily scale a co-located Web/EJB Application across multiple machines, in fact this is the architecture recommended by most.

Well, though this may be correct in some cases, it can not be true everywhere.
1. Web is only one of possible clients for EJB.
And if I follow your logic, I would say the most scalable fashion will be put all on client's PC. Then we will be back to old dark days, no more distributed computing.
2. For mostly read only EJBs, we can provide only one separated server and a few more web server. With the caching capability of EJB, this will be more scalable than the architecture you provided.
My $0.02
What is? Local Interfaces? Local Interfaces were one of the best things added to EJB 2.0. I don't get your point and I think you are way off on this.
Local interface is perhaps the best thing added to EJB2.0. What I'm against is to allow servlet communicating directly with local interfaces. (Sanjay Raghavan provided a far better approach...)