The Data Dictionary Form displays all of the data dictionaries available on the host system.

Open

The data dictionary form is primarily used to select one or more variables out of the data dictionary for display on a given screen.  The list of currently selected variables shown in the top-right of the form and shows the variable name, type, description, and source (dictionary) for each selected variable. 

The top-left of this form contains the data dictionary browser which displays the currently selected data dictionary either as a hierarchical tree-based or as a list of variable names, depending on the type of data dictionary.  For a tree-based data dictionary such as TsentryGsm, the root elements in the tree represent each of the available global commons, and expanding one of these commons reveals its internal structure.  For a list-based data dictionary such as one associated with a historical trend data file, the individual variable names are presented as a flat list.  The currently displayed dictionary can be changed by clicking on the “…” button above the browser.

The bottom half of this form allows the user to search for text within the list of all variable names and descriptions.  The user can search the data dictionary for multiple strings by separating them with spaces in the Search string; in this case all of the specified words must exist in either the variable name or description in order for the variable to be displayed.

Variables can be added to the list of selected variables from either the data dictionary browser or the search results by highlighting the appropriate variable and either clicking the Add button or simply dragging the variable name up to the list.  The Remove button removed the currently selected variable from the list and the Clear button removes all variables from the list.

Right-clicking on the list displays a popup menu with several options for modifying the variables in the list.  The Prefix Variable sub-menu allows the user to apply a prefix to the variable name in case he wishes to return not the variable value itself but rather some metadata about the variable.  The following prefix options are available:

When the user clicks the Accept button each of the variables in the selection list will be passed back to the caller.  For instance, on the ProbeX screen selecting multiple variables in the data dictionary form will add a probe for each selected variable

All of the files used by the TsentryGsm data dictionary system reside in the ‘Pif’ directory of the application. (example:  \LocNtRt\sys\pif\* ).

The following table describes the files used by the TsentryGsm data dictionary. 

The data dictionary is built with the GsmPifBld program located in the \tpriNtRt\bin directory.  It requires two parameters:

1.  The fully qualified name of the sysGsm.h header file (input)

2. The fully qualified name of the sysStruct.ini file (output)

The following is an example command as would be submitted to a DOS box command line:

d:\LocNtRt\tpriNtRt\bin\GsmPifBld  d:\LocNtRt\sys\Inc\sysGsm.h  d:\LocNtRt\sys\Pif\sysStruct.ini

Execution of this program will replace the current copy of sysStruct.ini with an updated copy built from the latest application C/C++ language header files.

This program parses all header files included in sysGsm.h and generates the text sysStruct.ini file that specifies the name, type, size, dimensions, and descriptions for all structures and structure members that it finds in the header files.

The data dictionary builder automatically extracts descriptions from the comments associated with each structure member defined in the header files.  Descriptions are parsed according to the following rules:

• Any comment immediately to the right of and on the same line as a structure member declaration will be interpreted as the description for the variable.

• If a comment is found to the right of the variable declaration, then any comments on subsequent lines will be also be included in the variable description.  Description parsing is terminated when a subsequent line contains anything except a comment (e.g. a new structure member declaration or a blank line) or any comment that does not have whitespace immediately following the start-of-comment token (e.g. “//int oldVariable;” will terminate the description parsing).

These descriptions are stored in the sysStruct.ini file along with the properties (name, type, size, etc.) of each of the structure members.

Note however that these descriptions are associated with single structure members in the data dictionary, and a given structure member may exist in multiple locations in the entire data dictionary.  The data dictionary generates the description for a fully qualified variable as follows:

1. If the variable is explicitly specified in the sysVars.ini file, and the variable definition includes a description, the data dictionary uses this string as the official description.

   1. Variable descriptions can be omitted from a variable definition in the sysVars.ini file by specifying an empty string (“”) or an asterisk (*) as the description.

2. If the variable is not specified in the sysVars.ini file, or the variable definition does not include a description, the description for the fully qualified variable name is generated at runtime by concatenating together the descriptions for each structure and member in the variable.  For the global common itself, the system uses the description specified in the [GblCmnDef] section of the sysVars.ini file.  Structure member descriptions are concatenated in reverse order relative to the fully qualified variable name so that the description for the final element is shown first.

For example, suppose a system includes the following header file:

and the data dictionary is initialized with the following sysVars.ini file:

The data dictionary would indicate descriptions for each of the following variables as:

The initialization of the data dictionary, including reading the variable definitions, allocation of the global shared regions, populating the global shared regions, and saving the global shared regions is automatically taken care of by the NtRtMgr process.