Jim Bedenbaugh

I'm debugging a jsp using Netbeans and bundled Tomcat.

The jsp is looking for a class and can't find it. I have the class in a jar file in the /lib folder. Other classes use this class and compile just fine.

I've tried everything. Here's the snippet:

<%@ page import="com.oneok.portal.mgmtauth.Reports,<br /> com.oneok.portal.Hierarchy,<br /> com.oneok.portal.util.CommonUtils,<br /> com.oneok.db.DBConnection,<br /> java.sql.Connection,<br /> java.sql.SQLException,<br /> java.sql.ResultSet,<br /> java.sql.Statement,<br /> java.util.TreeMap,<br /> java.util.ArrayList,<br /> java.util.Arrays,<br /> java.util.Set,<br /> java.util.Stack,<br /> java.text.DecimalFormat,<br /> org.apache.log4j.Logger,<br /> com.oneok.portal.beans.HierarchyBean,<br /> com.oneok.portal.beans.MgmtAuthBean,<br /> java.util.Iterator"<br /> %>

<%!<br /> public TreeMap setMgmtAuthBeans(TreeMap tmEmployees) {
TreeMap results = new TreeMap();
reports.getMgmtAuthBeans().clear();

Connection conn = null;
Statement stmt = null;
ResultSet rs = null;
String sql = "";
try {
conn = DBConnection.getConnection("MISC_READONLY");

It craters on the connection attempt.

DBConnection is included in the compile and run libs for Netbeans. According to Netbeans doc, these libs are automagically included in the classpath, although I would think Tomcat takes care of that by putting them in the /lib folder.

Any ideas?

Okay, I haven't been to the forum in about 3 years or so, but I've been working. . .

So who has the best practice tests and books for this exam?

Are you guys getting ready to start your new book yet? I think I heard Bert say the new book would be a HF Web Services book?

True?

My point is that EJB 2.X is still around and will be for some time, plus the new spec will be backward compatible, so EJB 2.X implementations are going to be around for sometime.

Just because 3.0 is coming doesn't mean that everyone is going to spend oodles of cash rewriting EJB 2.X - there's still a mountain of COBOL code out there still running - just because Java came along didn't mean a rewrite of COBOL code.

Same thing here. So basically, no, I don't agree with you.

Sorry, but I disagree and I thik a lot of other people would too. I've been at JavaOne all week and I don't hear of anyone abandoning EJB. In fact, I talked to someone who went to a session by Gavin King (who is on the spec team for EJB 3.0) who is really high on EJB 3.0 - and he is the Hibernate king (sorry, bad pun).

In fact, I'm encouraged by what I hear - I went to Linda DeMichiel's session on Tuesday and what they are doing is great, albeit it should have been this way in the first place. The DTO anti-pattern is going away and the simplification is wonderful - no more messy, needless implementations of interfaces except in specific cases.

We are going to do a rewrite in the next year and we'll certainly keep an eye on the 3.0 deployment.

I had a chance to meet Kathy Sierra and Bert Bert Bates at the author table and in talking to Bert briefly about the HFEJB book, I told him the book will probably be cut half, but we both speculate 3.0 won't be available for implementation probably until mid to late next year some time - I see no need to abandon EJB devlopment until then.

Anyone from the forum attending JavaOne in San Francisco, June 28-July 1?

I just read a blog on the web where a developer was complaining becasue of the following instance:
MethodA and MethodB are CMT methods who are defined as RequiresNew. Supposedly, they maintain when MethodA is called through the remote interface, a new transaction starts and then when MethodB is called through the local interface, the first transaction is not suspended and a new transaction is not started because MethodB was called locally. They maintain that the new transaction only starts when the remote is called.
This flys in the face of everything I've read about EJB transactions or the instances where I've used them, however, I thought that maybe it was because they were using an EJB container that didn't conform to the specs.
When a method is declared RequiresNew, it should start a new transaction, regardless of whether it was called locally or through the remote, right? Is anyone aware of a EJB container that does what was described above?

Originally posted by Michael Ernest:
Stay ar the Argent, then. You'll be in good shape.

Is that where you stay?

Originally posted by Bert Bates:
Do you have any budget restrictions?

Not really. Part of my negogiations for my job was a yearly trip to this conference - all fees, hotel, air fare, food, etc., although I don't want to be extravagant. I don't need a splendid room. Something functional with easy access to all the session breakouts. . .
Are you going to be at the conference this year?
[ February 10, 2004: Message edited by: Jim Bedenbaugh ]

Anyone here been to the JavaOne conference and if so, what is the best hotel to stay in in terms of location of the sessions?

Originally posted by Michael Sullivan:
A couple of comments. I've been working with procedural code for years, but for the last year have had to work with Java (as a server side web technology). . .

Your experience nearly mirrors mine. For 15 years I was a mainframe COBOL/CICS/DB2 programmer/analyst. My creditials as a programmer who understood business systems were impressive enough, but I had no bona fides to prove my Java skills, except to pass the SCJP.
Interestingly enough, I recently interviewed with a consulting firm and did not fair well on the technical review. While I have been studying and using patterns, J2EE architecture, frameworks, etc. (everyone seems to think these are most important, right?) since passing the SCJP in January of 2002, I soon discovered that a large portion of what I studied for the SCJP is mostly not used in everyday experiences and I plain forgot much of it. This probably explains why experienced Java developers often fail the SCJP if they don't study for it. I remember taking the first few mock exams for the SCJP. I couldn't fathom why anyone would write code like I saw in the questions - why in the world would anyone think it's important to know how the compiler would behave in a situation that you most certainly will never face in the real world?
IMHO, J2EE is the substance of the Java world. If you have a solid OO base and decent knowledge of patterns, the language becomes secondary. I can't believe knowing the innards of the HashMap implementation are important to the work-a-day developer.

Originally posted by Leonardo Penha:
Hi,
I am currently designing my component diagram, and I feel the ServiceLocator component is cluttering it.

I'd say your question might violate forum - and Suns's - rules concerning the exam, but I'll leave it to the moderators to decide.

Detlef,
Firstly, I think that the SCJP is a prerequisite for the SCBCD, so everyone who takes the SCBCD will have already taken the SCJP.
As for a comparison between the two, consider this: The SCJP covers the entire spectrum of Java where the SCBCD is a highly specialized area of knowledge. This translates into a two different types of tests: the SCJP will cover a broad area without getting too deep, however, given as the SCBCD is a tightly defined area, you can expect the depth of knowledge necessary to pass this exam to be much greater.
Having taken both, I cannot say one is more difficult that the other, merely different, however, I think that the preparation time for both should be about the same.
Hope this helps.

We are currently closing our offices in Ft Lauderdale, Florida and will be telecommuting from various points around the nation. To overcome the communications issues, we have installed DSL in the homes, web cams and speaker phones. We are using virtual whiteboards for visuals (if we are brainstorming a design, etc.)
Meeting notes will be taken to the point of usefulness, but nothing close to what would be required if we were an ISO9000 shop. We'll eventually implement CMM guidelines - eventually we plan to get to CMM level 5, so documentation is a must.
I can't see much excuse for not keeping virtual copies of certain communications.

Originally posted by Mark Herschberg:
Besides, Natalie will be thrilled to see your article.
--Mark

Um. . .sorry, but who's Natalie?