Task Design |
DEVELOPER |
 |
The programmer defines a set of PROIV Functions that can be executed by the calling application as a unit of work. Such a set of functions is termed a Task. A PROIV Task has a defined entry and exit point and specific error handling capabilities. It can operate in the background with no visible screen interface or, in certain configurations. it can automatically invoke a PROIV Windows Client on the calling machine to handle keyboard input and screen output. Tasks are stored in the bootstraps of a kernel on a machine. Multiple Tasks can be stored in a set of bootstraps. There can be many machines on the network on which Tasks are defined.
The programmer defines each Task’s name and optionally defines a set of parameters to pass data between caller and kernel (in both directions). and between functions comprising the Task. A process called 'mapping' is done at Task Definition time to map Task parameters to function variables. For UNIX environments, task definition and mapping can only be done through PROIV Developer. It is not possible to maintain tasks from a 'green screen' or 'buffered terminal environment'.
The programmer also needs to plan which Task runs on each machine, as the tasks reside in the kernel bootstraps. When the caller invokes a Task, the Task request must specify the name of the machine on which to execute the Task.
Once defined, a Task can only be executed via the PROIV Bus by non-PROIV applications, typically using the P4API toolkit. P4API is packaged as a Win32 DLL. OCX Library, Java Class Library, and Unix archive library for C and for C++. The P4API is documented in PROIV Bus Application Programming Interface.
Execution status can also be returned from the kernel back to the caller.
Section links:
Elements
of Tasks explains the purpose of Tasks and how PROIV Functions interact
with them.
The
Task Model outlines the PROIV Task model.
Parts
of a Task outlines the constituent parts of a Task
Defining
a Task explains how to create
Topic ID: 540214