Raible Designs

[TSSS] BOFs, Booze and Benitar

On Friday night, I attended all three of the 7:30 Birds of a Feather sessions. The first one I went to was Spring, where Rod talked about what's coming in Spring 1.3. Rod did a 25 minute presentation on the new stuff and then opened the floor up to Q and A. The session was well attended and I skipped over to the Tapestry/Trails BOF when the Q and A started.

I was surprised to find that very few folks where at the other BOFs. While the Spring BOF had 50+ attendees, the Tapestry one only had around 15-20 and JSF had around 8. I quickly left the Tapestry/Trails BOF when Chris started walking through his Trails Video. He was doing a live version of it, and since I'd already seen it, I figured I wasn't going to learn anything new. I've also been following Trails since it first started, and was more interested in talking about Tapestry.

I walked into the JSF BOF as Ed was talking about JSF 1.2 and what's next for JSF 2.0. This was good timing as I had a few suggestions for 2.0: HTML Templates like Tapestry, bookmarkability (don't make everything a post) and thinking about tools like Tiles and SiteMesh. While neither tools is part of the spec, I think they should be remembered in case there's an opportunity to make integrating with them easier. Ed did mention that JSF 1.2 has pretty much solved the content-interweaving problem, so putting HTML in your JSF JSPs should be better supported.

The very interesting part of this BOF is that Ajax capabilities are very much on the radar for JSF 2.0. They plan on providing native XMLHttpRequest capabilities. My suggestion for this was to provide the setup and registration of requestable class methods as part of the framework, and leave writing the JavaScript to the developer. This was a good BOF and I'm pumped to see that JSF is embracing the next-gen way of developing webapps. Let's just hope JSF 2.0 is released this year and not 2 years from now.

After the BOFs, I joined Matt and Jim to wait for one of Matt's buddies (Scott) to come into town. After he arrived, we headed over to the OpenSymphony open bar at the Bellagio. There, I got to meet Patrick Lightbody and enjoyed several beers and good conversation with the likes of Seth Ladd, Thomas Risberg, Mike, Dion and Christian.

After the open bar closed, Jim, Matt, Scott and I headed just off the strip to the Gold Coast Casino. Matt and Scott wanted to find some poker (tables had a 2-hour wait on the strip) and Jim and I wanted cheap Blackjack. We were pleased to find $5 tables and stayed there for several hours. I don't know what time we headed back to our hotel, but I'm guessing 1 or 2. The rest of the night was pretty funny. Jim and I gambled until 7 in the morning at several blackjack tables. Our hotel had this "celebrity theme", so we had dealers like Pat Benitar and Stevie Wonder throughout the morning. Both of these dealers were great and I got "hooked up" on several occasions. There were at least 10 times where I asked for a card and they didn't give it to me (after which I won b/c they busted). We ended the night at 7:00-7:30 with 5 crisp $100 bills in my pocket. Total cost of the whole trip: $100. Not bad eh?

Getting home yesterday was quite an adventure. After going to bed at 7:30, I woke up by some miracle at 11:00. I don't know if I had a wake up call or what, but my buzz was still in full swing. I caught a cab and headed to the airport. I paid the cabbie with a $25 chip, which he didn't like, but after I told him to keep the change (it was a $6 cab ride) - he happily obliged. At the airport, I took a nap while waiting for my flight to board and almost missed it. They called my name over the intercom b/c I was the only passenger left to board. Luckily, I was awake and made the flight. Upon arriving in Denver, I walked to my car and promptly locked my keys in the trunk. The airport officials got them out for free and I made it home to a very happy family around 6:00 p.m. It's good to be home.

[ANN] Equinox 1.3 Released

This release is mainly a bug fix release, but it also adds support for Maven. All of the frameworks used in Equinox, as well as its build/test system is explained in Spring Live. Detailed release notes are below:

- Added missing "validator" property to "userFormController" bean in Spring MVC version.
- Added "redirect" element to success mapping to user list to prevent duplicate post problem.
- Moved "ctx" variable declaration from decorators/default.jsp to taglibs.jsp so it's available to all JSPs.
- Changed any references to UserDAO in UserWebTest.java instances to use UserManager instead (to prevent problems when transactions aren't used).
- Fixed install scripts in extras so they'd work on Windows from the command prompt. Added "fixcrlf" target for users that encounter issues.
- Added installer for Maven in "extras/maven". This can be used to replace the Ant build system.
- Dependent packages upgraded:

• Display Tag 1.0
• Hibernate 2.1.8
• iBATIS 2.0.9b.550
• JPOX 1.1.0-beta-1
• Spring 1.1.4
• Tapestry 3.0.2

Download. For more information about installing the various options, see the README.txt file.

Demos:

• Spring
• JSF
• Struts
• Tapestry
• WebWork

Rather than uploading the different combinations that are possible with Equinox, I figured I'd just wait for requests. So if you'd like things like Tapestry+Spring+JDO, or JSF+Spring+JDBC, let me know and I'll upload a pre-built version of 1.3.

JBoss ClassLoader Logic

Is there any logic to how JBoss puts all WARs, EARs and their accompanying JARs into the same ClassLoader? It seems logical that I should be able to deploy different versions of a JAR in different WARs. This works fine on Tomcat, but doesn't seem to on JBoss. Is there someway to turn this segmentation on?

RE: Ruby on Rails Koolaid

David Geary on Ruby on Rails:

Interestingly enough, ROR creates default views for your db tables. Cool. But after I thought about it for a minute, I came to the conclusion that that feature is certainly close to useless: It's nice to get you up and running, and great for seductive demos and articles, but you're going to override at least 100% of the views that ROR generates. And therein lies the rub...

...because views in ROR are a mixture of HTML and Ruby scriplets! We've been there before, of course, in the early days of JSP with HTML mixed with Java scriptlets. No thanks, I'll pass on that giant step backwards.

I've thought about this myself. I think David is right, but only to a certain point. Ruby on Rails (and its scriptlet-ridden views) will work until they get HTML designers in their writing Ruby code. Furthermore, when companies start off-shoring their Ruby development, that's when it gets ugly. JSP scriplets were a disaster because you had HTML developers writing Java code, and using if statements that exceeded the 64K block and such. JSP scriplets are hard to maintain when they're used and abused. I've seen a lot of code (especially the off-shored stuff) that has very ugly and unmaintanable scriplet code. However, using scriptlets in JSPs isn't a bad thing - it's only bad if you're coding business logic and/or, using lots of Java code in them, or having JSPs that are scriplet-only pages.

However, I believe Rails is a bit different. Not only does it enforce MVC from the get-go, but you can't use the views stand-alone (can you?). With JSPs and Servlets, you have the opportunity to use JSPs only - which encourages scriptlets. I do wish that Rails' templates used the ${...} syntax that JSP, Velocity and FreeMarker enjoy - the <% %> syntax brings bad bad thoughts of 1000+ line JSPs.

Oh, and one last thing - for David and Rick (the JSF-is-the-best-thing-since-sliced-bread-duo). At least Rails allows HTML in its view templates. JSF developers don't even get to see HTML anymore - poor guys.

Context Reloading in Tomcat 5.5.7

Today I started using Tomcat 5.5.7 instead of 5.0.28. It was fairly easy to install on my PowerBook - I just had to add the xerces and jmx JARs from the compat package to get things working. The one thing I noticed that's different from Tomcat 5.0.28 is that when I deploy any file, it reloads the context. This can be a pain when I'm just copying JSPs into the webapps directory. I'm willing to admit it could be a problem with my "maven deploy" goal, but since this didn't happen on 5.0.28, I suspect it's the newer version of Tomcat. On 5.0.28, the context was only auto-reloaded when I updated files in the classpath.

The main reason this was frustrating is because we look up a bunch of data from a web service on app startup. The API we're talking to is nice and slow and it takes almost 50 seconds to start our application. Rather than go back to the older version of Tomcat, I wrote some code to serialize all the ServletContext variables to disk, and then check for that file on startup. If the file exists, it deserializes it and puts all the objects back in the ServletContext. Works pretty good and certainly speeds up my development environment. After all this, I'll probably downgrade to 5.0.28. Auto-reloading when classpath files change seems like a better way to go.

[DJUG] JUG Central and BPEL

Last night, I attended Denver's JUG meeting. Below are my notes from the event.

I'm at DJUG listening to Christian and Kris (from Adigio) talk about their experience with using Spring, WebWork, Hibernate, Lucene and SiteMesh to develop JUG Central (I wonder if they knew this name and concept already exists?). JSPs are for the view and MySQL powers the data. This presentation is designed to explain a bit about each framework, and also tips/tricks and pitfalls they experienced when developing the site. They started working on the application in August of last year and deployed it into production in December.

Christian said they weren't going to go into the how for each framework, but Kris has had quite a few slides on SiteMesh so far. I don't blame him - it's a great tool and only a handful of folks (of about 50-60) have heard of it.

SiteMesh Pitfalls: Poor integration with Velocity and some other frameworks. BTW, if you're using Tapestry - Erik Hatcher recently created a JIRA patch with a Tapestry Decorator.

Now Kris is talking about WebWork and since he's a framework junkie, apparently this is going to be the largest part of the presentation. I think one of the nicest parts of WebWork is its auto-type conversion. The only other frameworks I've seen that have this is are JSF and Tapestry. For those that like WebWork and don't like JSF - you might find it disturbing that the WebWork actions (and their tests) in AppFuse are very similar to the JSF managed beans. I would take it as a compliment if I were a WebWork developer.

One nice thing about XWork's action configuration is you can specify a "method" parameter for a particular action. Struts recently added this with its MappingDispatchAction. I'm using this on my current project and it works quite well. Kris really likes WebWork's front-page controller pattern - where you use the <ww:action> tag to execute the action when the page is loaded. Personally, I don't have a problem with going through actions to get to my view templates. Kris finished up his WebWork piece with a plug for AppFuse (thanks!) and WebWork in Action. Congrats to all the authors - wonder if it'll be published before WebWork Live?

Now Christian is talking about Hibernate and its mapping files - and how you can generate your database schema from them - or generate your mapping files from a database. They used XDoclet to generate the mapping files in this particular project.

Hibernate Pitfalls: Think about lazy-loading early. Problems arise when you try to share Hibernate-managed objects across (Hibernate) sessions transparently. Christian mentions that Spring's OpenSessionInViewFilter is a nice way to solve the problem.

Hibernate Tips: Spring simplifies using Hibernate and makes declarative transactions easy. Read Hibernate in Action before starting development. Plan to spend some time learning how to express your data model with Hibernate relationships (one-to-one, one-to-many, many-to-many, etc.).

Christian is now talking about Spring and how it works. After thinking and writing about Spring so much in the last year, I'll just skip over regurgitating this part. ;-) His main recommendation: use real injection instead of appContext.getBean("beanName").

Other tools used: Lucene for searching and POI for indexing Word, Excel and PowerPoint files. Velocity used for templating e-mail messages.

Service-oriented Architecture (SOA) with Business Process Execution Language (BPEL)
Presented by Kevin Geminiuc and Owen Newnan from Policy Studies

This point of this presentation is to communicate what it's like to implement BPEL in a J2EE Container. BPEL is a layer on top of web services. BPEL is a programming language that you can use to program business processes. Allows you to divorce your business process from being human-centric to being document-centric. At Policy Studies, they're using iLog JRules rules engine and Oracle's BPEL implementation.

Benefits:

• Process Visiblity
• Process Agility
• Powerful Language
• Open
• Backed by "the Big Boys" (BEA, Microsoft, IBM)

History: Formerly knows as BPEL4WS, WSBPEL. Open standards based. Orchestrates web services with SOA.

Where we are today: Emerging technology (prepare to bang your head against the wall). .NET and Java products exist, as well as J2EE container integration.

BPEL & WS Standards: BPEL, XPath, WSDL, WS-Addressing, SOAP, XML-Schema, WSIF (Axis), TBD (WS-ReliableMessaging and JSRs 207/208). Note that since BPEL depends on web services (which is not a truly reliable service). Because of this, there are some proprietary extensions available.

At this point, I became bored with the presentation and quit taking notes. While the speakers had good intentions with their knowledge sharing, their delivery needed some work. The code walkthrough and demos were presented with a monotonous and unexcited tone, and a handful of folks left during this part. In summary, BPEL looks like a good way to orchestrate your various business processes. It allows you to call web services, EJBs and whatnot simply by defining their locations and methods in XML.

In his demo, Kevin used Oracle's BPEL Designer, which is an Eclipse plugin that has a nice drag-n-drop editor for managing your BPEL XML files. He also used Oracle's BPEL Process Manager, which seemed to be a lot like Jetspeed - you just drop in the .ear and then deploy your processes to it. The only bad part about the Process Manager is it's administration/deployment interface only runs in IE.

If you're using BPEL in your projects, I'd be interested to hear the tools you're using. As far as open-source BPEL process engines, they mentioned Twister and ActiveBPEL.

[DJUG] Testing and Handling Exceptions in the Web Tier

I'm attending Denver's JUG tonight, where Scott Davis is talking about Unit Testing the Web Tier. His opening slide says he's going to cover HttpUnit, Canoo WebTest and JMeter. I'm most interested in the JMeter stuff as I've been meaning to integrate it into AppFuse. I've used HttpUnit and it's a little verbose for me. I prefer using Canoo WebTest or jWebUnit over HttpUnit. On my current project, we're considering using jWebUnit or HttpUnit to act as a browser when interacting with a 3rd-party system.

All of these tools run functional tests - which are much different from unit tests. Unit tests usually tests the bricks, whereas functional tests test the building. For unit and functional tests to be truly effective, they must be:

• Scriptable
• Repeatable
• Automated
• Darn close to 100% coverage

The tools Scott is talking about tonight have passed his basic tests:

• Can I learn it in 10 minutes?
• Does it play nicely with my existing test environment?
• Does it play nicely with my existing production environment?

I didn't take any notes about HttpUnit or Canoo WebTest because I didn't really learn anything new. Scott did do a nice job in his HttpUnit examples - he made it look a lot simpler than I've previously seen. I've used it HttpUnit before and it seems a bit verbose. I've always used jWebUnit, which simplifies HttpUnit's API.

JMeter allows you to do the same thing as HttpUnit and Canoo WebTest. It's a standalone GUI, for the complete non-programmer. It does not plug into Ant/JUnit and is mostly used for load testing. I thought it was exclusively used for load testing - and I think it has an Ant task. I could be wrong.

<sidenote>Scott uses Smultron for XML editing on his Mac.</sidenote>

The basic building block of JMeter is a "Thread Group". The Thread Group allows you to control the number of users/threads that run a particular test. You can test gets, posts, change the protocol and even upload files. For load testing, make sure and check "Retrieve all Embedded objects in HTML files". You have to view the "result windows" to view that your tests actually ran - there's no "green bar" feature.

I think JMeter has improved a lot since I last looked at it. Scott's overview and demonstration make it look very straight forward and easy to use. One guy asked if it's possible to see a a global view of all tests run. Scott thinks it's possible by adding a Listener to the Thread Group and creating a graph (one of the options). Scott is now showing a lot of the options in JMeter - there's a ton! It's almost overwhelming.

Next up is "Exceptional" Web Apps by Stephen A. Stelting, a senior Instructor and Author from Sun. His latest book is "Robust Java". Stephen has spent the last year and 1/2 figuring out how to make Java fail.

Objectives:

• Describe the types of errors that occur in the web
• Explain how exceptions and errors can be handled
• Describe the web container response to exceptions
• Present best practices to address web tier exceptions
• Show how web frameworks handle exceptions

Payoff: As a result of this talk, you'll have a better understanding of how to use exceptions in Servlets and JSPs, improving the robustness of your webapps.

I'm a little skeptical at this point. I think most folks don't do exception handling in their webapps. I hope Stephen has some good tips and tricks for those of us who are familiar with handling exceptions. I wonder how he feels about Spring and its runtime exceptions?

There are two types of exceptions in the web tier: HTTP Errors and Java Exceptions. Standard HTTP Errors are handled by the web server. You can also send your own HTTP errors by calling HttpServletResponse.sendError(). If you're using response.sendError(), make sure and call it before you commit the output. The web.xml file allows you to specify errors and exceptions with the <error-page> element.

Servlets and Filters have similar exception behavior. Both declare exceptions in both of their lifecycle methods: init and service (doFilter for Filters). Developers throw exceptions in lifecycle methods to "tell" the container about problems.

• javax.servlet.ServletException
• javax.servlet.UnavailableException
• java.io.IOException

Stephen is now describing the init() method and the exceptions it can throw. Yawn. I think most Java web developers use frameworks these days. Because of this, most developers probably don't use these methods because they don't write plain ol' servlets. One thing I didn't know is that UnavailableException takes a time parameter - if you throw an UnavailableException with this parameter, the container will retry after the specified amount of time.

Result of Exceptions in init(): The destroy() method is never called, since the initialization did not complete. Client calls during component unavailability render a 500 error.

I stopped taking notes at this point because my laptop battery was dying. I didn't really learn much in the rest of the presentation. While I can appreciate Stephen's enthusiasm, it was obvious that he was an instructor and not an in-the-trenches developer. He explained a lot of what and didn't have any code to show how to do stuff. There wasn't a single demo in the entire presentation.

Most of the exception handling stuff Stephen talked about for the rest of the session was common sense (IMO). It also centered around the Servlet and JSP API, which most folks probably don't mess with. The Struts and JSF coverage at the end was cool. If nothing else, it was to nice to hear a Sun employee confirm that JSF is quite deficient in its hooks to allow easy framework-configurable exception handling.

Now that I'm working at home, and working/interacting with friends all day - it seems that the DJUG meetings aren't as exciting. They used to be fun because I could get out of the house and have a few beers with friends. Maybe it was the lack of learning anything new tonight.

Pick the web framework you think is cool

Ever since I started adding additional web frameworks into AppFuse, people have asked me "which framework should I use?" I've often told them "use what you know." If you have in-house knowledge of Struts, use it. I thought this was good advice because I believed that existing knowledge leads to greater productivity.

Lately, I've started to change my philosophy. I'm starting to think it's more important to use the web framework you're passionate about. The one you want to learn more about. After reading Kathy Sierra's "Does it really matter if your tool is cool?", it seems this is a good idea. She writes:

Coolness (or just perceived coolness, it really doesn't matter) is linked to passion. The cooler you perceive your tools to be, the more passionate you are about those tools. And passion, while it might lead to the "everything is a nail" syndrome, has an extraordinary amount of value!

Obviously there's quality of life... a life with passion is certainly more fun than one without. And the more passion, the greater the chances that a person has what psychologists label optimal experiences. And the more optimal experiences one has, the more likely one is to describe life as being "happy". So, passion = optimal experiences = happiness. And research says happy people are generally more productive. Certainly they're more spirited and fun to be around...

So I guess passion leads to greater productivity, not existing knowledge. So which web framework do you think is cool? Which one are you passionate about?

If I had to choose based on my passionate choice, and the one that I think is the coolest, I'd have to go with Tapestry or possibly JSF (JSF would be a lot cooler if it let me put my JSPs in the WEB-INF directory instead of in the root). These are the frameworks I want to learn more about. 6 months from now? Maybe Laszlo or JDNC.

The Future of the J2EE Web Tier

There's a chat in a few hours on the future of the J2EE Web Tier. I'll be sitting in and hoping to learn something. The alignment of the JSP EL and JSF EL will be nice - but we really want HTML Templating for JSF - like Tapestry. After developing Tag Soup JSPs for JSF and HTML-editor-friendly pages for Tapestry, I can say there's definitely a big difference. Tapestry templates are much easier to read. However, if you have an existing JSP app, you can migrate to JSF easier.

The AppFuse Generator

I found the AppFuse Generator project in my comments last night. Richard So seems to have created a nice code-generation tool for AppFuse. It looks like it can create/modify all the files you need to CRUD a database table. I took a brief look at the code and it's all based on Velocity templates, so you should be able to easily tweak things.

This tool does seem to generate the DAOs/Managers - which can lead to more code to maintain. It would be nice if it could be modified to be similar to AppGen where there's a "detailed" version (that generates everything) and a "default" version (which only generates the web tier classes). The default version could use the generic "dao" and "manager" beans. I wonder if this tool supports running/installing the same table more than once? This currently doesn't work too well with AppGen, but shouldn't be too hard to fix.

In reality, I'm pumped that Richard has created this, as well as a project to back it up. I especially like that he's going to try and create an Eclipse Plug-in as part of this project. Nice work Richard!