Archive for June, 2008

I have no idea when they made it available but, while searching on Oracle E-delivery for a complete Oracle BPA package, I saw that a lot of BEA software can be downloaded via Oracle’s E-Delivery channel.

It look like it concerns all BEA software packages, although I don’t know their full product stack by heart.

It’s time to play…(and re-learn what has been lost)

Actually this is old stuff (2006), but it got lost in a comment section. I think this can still be useful to some and I also post this here for prosperity.

Somewhere in 2006 my colleague Lucas Jellema wrote a post on the AMIS Technology site about querying rss feeds from the database. My colleague Anton Scheffer and I commented on that article with our XMLDB functionality mindset.

Later on in 2007 Lucas wrote another useful post called "Querying RSS feeds in SQL…". Peter Wolf commented that he had also written a very nice blog post about using XMLDB functionality while Integrating Yahoo Pipes into APEX.

I think this is still useful stuff to a lot of us, although you should keep in mind that the table(xmlsequence(extract())) construct will be, in time, out lived by the XMLTABLE function. Also XMLTABLE supports XPath V2, the table(xmlsequence(extract())) doesn’t. The XMLTABLE function is available from Oracle database version 10.2.

I reconstructed my statement in the comment section from the post in 2006 and "solved" the small puzzle that I couldn’t solve at that time. Of course, especially via XQuery, there are more good solutions, but the following is a first step.

Anton’s example that is still very useful from database version 9.2.0.3 up to 10.1 (if not only remembering the "old" Oracle 9.2 syntax).

Read the rest of this entry »

The ODTUG Kaleidoscope conference in New Orleans has just finished with the Apex vs ADF session of Lucas Jellema and Dimitri Gielis. They both demonstrated that you can create a fully functional and appealing application within a day. They made it very clear that is was not a shoot-out. Both frameworks have their strengths and their own playing fields and they can even be viable alternatives for each other. With once in a while a teasing remark it was real fun and a  last session.
In my opinion ths year’s ODTUG was definitely worthwhile. Last year we were a bit disappointed because it seemed that the conference was not innovative enough and remained stuck with the ‘old’ Oracle toolset. This year was much better, lots of sessions about Apex, BI, ADF, the newly aquired Essbase, PL/SQL and much more. It was even a bit frustrating that many ADF sessions were about or used JDeveloper 11g, although quite stable, is not yet released and nobody would say anything about a release date.
I was especially happy with the many sessions that were dedicated to good practices, tips and trics and things they don’t teach you in the courses and demos (by Kenton Ho). I think these kind of sessions truly show the ODTUG Kaleidoscope’s identity. The other aspect is the unformal and easy atmosphere. It is extremely easy to speak to anybody, from participants to speakers and the experts in the field. In addition, New Orleans is a real nice place to stay and Bourbon Street is hard to forget .

My highlights :

• Kenton Ho with ‘Our first ADF/JSF based product: What they didn’t teach you in the courses and demos’. He is a good speaker and provided us with 10 (and more) good practices and tips on ADF development from their own experiences.
• Dan Norris with ‘Oracle identity management – The total identity solution. A subject I am not familiar with and it was great to get a clear picture of all aspects of identity management like user management, authentication and single sign on, role management, access management etc. etc. Oracle provides a complete solution, mainly based on products that they have aquired in recent years.
• Steven Feuerstein with ‘Weird PL/SQL’. Unrivaled as a speaker, he shared his notions about funny aspects of PL/SQL, for example about Oracle’s exception numbering that is both negative and positive at the same time.
• Éric Marcoux with ‘Managing Oracle Forms, PL/SQL and ADF source files in a single environment’. He presented about their experiences with improving the development process, amongst others with maven based continuous integration, a subject I am heavily interested in and which forms a key component in the AMIS software factory (or a bit friendlier ’software studio’).
• Steven Davelaar with ‘Beyond drag and drop: ADF Faces advanced techniques’. Great presentation about complicated stuff like the pagelifecycle, the immediate property, partial page triggers and the difference in behaviour between the disabled and rendered property of ADF Faces components.
• Paul Dorsey with ‘Thick database techniques for Fusion (and other) Web developers’. He claims that putting business logic (and more) in the database and reducing the middle tier to its bare minimum improves a webapplication by factors. It’s scales better, uses less resources, increases development productivity and reduces the amount of code. His method relies heavily on hiding the complexity in database views with instead-off triggers and pl/sql collections. It is certainly a very interesting and viable approach, but somehow I don’t think the debate is over yet.

.. and ofcourse my own presentation ‘ADF development: live from the trenches‘. In which I share my / AMIS’s experiences with ADF development, like naming conventions, code examples, things to avoid, the importance of unit-testing and continuous integration etc. etc.

I can’t wait for the ODTUG Kaleidoscope next year, probably in Monterey on the west coast.