DPML has stood the test of time.
This tutorial is now about five years old and DPML has proven to be a solid and
valuable technology for NetKernel application development.
As we approach the release of NetKernel 4 we realized that we should keep the jokes for a least a while longer.
However, ...
With NetKernel 4 DPML will change dramatically to become a very powerful closure based context-aware language.
Stayed tuned for the lastest!
Author Comment Jan 2005
This tutorial was written nearly 3 years ago. It provides a purely DPML-centric view of using NetKernel. Since it was
written, NetKernel has evolved and now supports a wide choice of languages. Keep in mind that
you have more than one option available. It is convenient to use DPML as a high-level scheduling language to invoke other
services/pipelines written in a dynamic/scripting language.
This tutorial progressively reveals some of the workings of NetKernel from the perspective of DPML. This deliberately concentrates on stuff
you need to understand to use DPML and so is not a comprehensive view of the underlying concepts - please take a look at the
NetKernel overview for a more complete view of the NetKernel concepts. DPML is a very simple language
which constructs active URI requests and issues and sequences these requests to the kernel - understand
the NetKernel abstraction and you'll understand how DPML really works!
A fast-track DPML reference is available here. Finally, aren't you pleased I didn't carry on with the bad jokes
throughout the rest of the documentation?
DPML is a declarative processing language for sequencing NetKernel requests. This is a bottom up introduction to
DPML processing which provides many interactive examples and real working applications.
The highest commendation we've so far received for this book is, "For sure, you're no John Grisham", Anon.
After reading Part 2 you will know how to use intructions in general and how to specifically
perform XSLT pipeline transforms, amongst others.
You will have learnt how exception handling can be used to make a process robust.
You will have learnt how to modularize an application into components for maintainability and performance.
By this stage you will either be chomping at the bit for the next chapter or you will have abandoned it
having got the picture and written a generalized DPML-based
service scheduling language runtime with full distributed transactionality.
After reading Part 3 you will know how create conditional processes, create transactional locks and
use inderection for dynamic applications. You will be able to build scalable, robust applications.
By this stage you will playing with all the Accessor technologies and will be considering DPML as the basis
for the control-rod algorithm for your home nuclear reactor project.
Experiment in the Workbench
You can experiment with any of the examples or start trying your own ideas by rapid prototyping them in the
workbench module. This is located in the directory [install]/modules/workbench/.
You can execute your apps with your browser.
So if you created a directory 'myApp' and an idoc 'useful.idoc' you could run it with
http://localhost:1060/workbench/myApp/useful.idoc. Have fun!
SysAdmin Note ffcpl:/workbench/ is exported by the workbench module. It is configured to incorporate all of the standard
modules making all the core technologies available for experimenting with.
Tutorial Introduction
Fundamentals
Practicalites
DPML By Example
