Accessor DevelopmentAccessor Development
Introduction to Accessor Development
Home > Books > NetKernel Extensions > Accessor Development

Rate this page:
Really useful
Satisfactory
Not helpful
Confusing
Incorrect
Unsure
Extra comments:


Accessor Development

An accessor is code that acts as the end-point for resource requests. Resources are an abstract notion about information; the role of an accessor is to accept requests for resources and return concrete, immutable resource representations.

Accessors can be categorized based on intent:

  • Active Accessors provide an endpoint for a function. Functions in NetKernel either transform resources (e.g. active:xslt) or provide a service. Active accessors only handle SOURCE type requests.
  • Data Accessors provide an endpoint for data. They must support the SOURCE request verb and usually handle SINK, EXISTS, NEW, and DELETE. Data accessor generally provide an implementation of a complete URI scheme - examples are transient, ffcpl, and data.

Language Choice

Accessors can be implemented in any supported language. However, distinctions can be made between DPML, and Java and other scripting languages.

  • Java can be used to write any accessor. Java accessors will have superior performance and the greatest flexibility.
  • Scripted languages can be used to write any accessor. Typically scripted languages cannot hold state between invocations in static variables.
  • DPML can be used to write active accessors.

© 2003-2007, 1060 Research Limited. 1060 registered trademark, NetKernel trademark of 1060 Research Limited.