(0 comments, 18 posts)
This user hasn't shared any profile information
Posts by Peter Ebell
My former collegue and JHeadstart Team member Steven Davelaar will join us at AMIS on wednesday May 19th, at 18:00, to present on a particular feature of the new JHeadstart 11g release: the JHeadstart Forms2ADF Generator (JFG). Forms to Java and/or ADF migration tools have been around for a long time – for instance, VGO Software has been doing this for quite some time now with their EVO tool, and their latest release also offers ADF 11g as a possible “target” technology. But there are many more. And with every Forms2Jave migration tool, the critical issue is always: “What about the PL/SQL logic??”. (more…)
One of the most annoying things with XML Schema validation is that it will not accept empty elements of certain types, like xsd:date and xsd:double, even though the element is defined as optional (minOccurs=”0″). OK, I do understand that an empty element _does_ occur, and that this behaviour is therefore correct, but the implication is that developers will have to add <xsl:if> tags in their XSLT transformations around each and every element where this may cause a problem.
This article will show how using a small and generic Java class and the “Java Callout” feature of the 11g Mediator, we can automatically “fix” these empty elements so that they will not cause problems with schema validation any more. (more…)
My OpenWorld 2009 started out this morning with a keynote by Tom Kyte named "What Are We Still Doing Wrong?". A lighthearted presentation with lots of funny examples of bad coding practises. After that, things heated up with an "OSB Deep Dive", presented in her usual inspired – and inspiring – fashion by Deb Ayers, who clearly had way too little time to tell everyone about the cool stuff her team has been working on the last year. The feature that – judging by the amount of questions – raised the most interest was the "Result Cache": a "checkbox-easy" feature using Coherence to cache results from earlier service calls. (more…)
Just wanted to draw your attention to a "Knowledge Center" session at AMIS this wednessday (25 feb 2009, starting at 16:30). It concerns the new Oracle Service Bus (previously the BEA Aqualogic Service Bus), and will be presented by former collegue Jeroen van Wilgenburg, now working at Xebia. Jeroen has worked with the OSB on his current project, so he will be able to give us real-life experiences, not just a "summary of features"! For more information, and information on how to register, take a look at the AMIS Agenda: http://www.amis.nl/activiteiten.php?id=681&preview=&draft=
The other day, I was using the JDeveloper debugger to locally debug some code in an application that uses (and therefore becomes part of) an Oracle Coherence data grid. As it turned out, using step-by-step debugging in a JVM that is part of a Coherence cluster has significant impact on the rest of the grid. The other nodes see the node constantly alternating between being responsive and being frozen, and keep taking measures to ensure the "health" of the grid. As Coherence values "health" or better, "consensus" within the grid over response times, this regularly causes processing on the other nodes to temporarily stop as well.
To remedy this problem, I wanted to alter the "grid initialization" logic in my application to check whether the JVM is running in debug mode or not, and if so, to not do the grid initialization. I checked the java command line with which JDeveloper runs an application in debug mode, and found that there is a JVM argument "-Xdebug" that is indicative of this fact. But as this is not a "custom argument", i.e. it does not follow the "-Dname=value" syntax, you won’t be able to obtain this argument at runtime using System.getProperty(). More >
Even when JSF was not around and ADF Faces was still called UIX, Oracle already had the cool "Partial Page Request" (PPR) feature. What this boils down to is that when a certain component in the page issues a "partial event" , an (AJAX-like) request will be sent to the server that (by default) only refreshes the component itself based on the server response, rather than refreshing the entire page. What makes this feature really useful, however, is that you can also specify that other components in the same page can be refreshed along with the component issuing the "partial event". Take the classic "dependent poplists" case, for example a poplist with "Car Manufacturer" and a second poplist "Car Model" that only shows the cars made by the selected manufacturer. When that first "manufacturer" poplist changes, we need to refresh the values in the "model" poplist as well (while the rest of the page remains in place and unchanged). (more…)