IBM Visualization Data Explorer Programmer's Reference
[ Bottom of Page | Previous Page | Next
Page | Table of Contents | Partial Table of Contents | Index ]
Partial Table-of-Contents
The Module Builder is a point-and-click interface for creating
a module from user-supplied information.
The next several sections describe the general structure of modules
built with the Module Builder; its dialog box; and examples
of the "worker routine"--the interface between a
module and the user-supplied application code.
From specifications supplied by the user, the Module Builder
generates three files:
- a module description file (with the extension
.mdf);
- a C-code framework file that implements the structure of the
module; and
- a makefile.
The C-code framework file itself consists of three major routines:
- The first routine checks input parameters and creates output
Objects.
It is named by prefixing m_ to the module name
(e.g., the entry point for a module named Example is
m_Example).
- The second routine traverses hierarchically defined
data Objects.
When this routine (Traverse) encounters a Data
Explorer
Field Object, it calls the third routine.
- The third routine (doLeaf) creates a "worker
routine" as an interface to the user's application-specific
code: it extracts the input arguments, sets up the outputs,
and calls the user-supplied code.
(See 3.7 , "Worker Routine" and "Worker Routine Examples".).
The worker routine appears at the end of the C-code framework file.
To complete a customized module, the user need only:
- add the application-specific code to the worker routine (after the
line "User's code goes here" at the end of the C-code
framework file) by:
- using an "include" file, or
- adding the application code directly to the framework file.
Using an include file is recommended because (1) if you rerun the
module builder, it will overwrite the .c
file, and (2) it makes replacing your own code easy.
Notes:
- The Module Builder is designed to accommodate the most common form
of customized module, in which the output Object is a copy of
the input, but the data component is modified.
- Other inputs can be constant parameters or other hierarchically
defined data Objects
(note that the hierarchy of the data Objects must match exactly).
[ Top of Page | Previous Page | Next
Page | Table of Contents | Partial Table of Contents | Index ]
[Data Explorer Documentation | QuickStart Guide | User's Guide | User's Reference | Programmer's Reference | Installation and Configuration
Guide ]
[Data Explorer Home
Page]
[IBM Home Page | Order | Search | Contact IBM | Legal ]