Page 782
NOTE |
For depiction purposes at the analysis level, it makes sense to depict domains as entities without attributes. All of these entities could have a similar title, such as Types. For example, you could use Payroll Type, Employee Type, or Status Type to represent attribute types that will become domains in your physical model. |
WARNING |
|
Often, there are multiple diagrams for one application in CASE. For example, a business might have an accounting department and a payroll department. Both departments could have separate diagrams. A third diagram could depict all of the departments in the business. This technique is very useful when the system is very large and there are specific areas of concentration involved. It also allows the users and fellow developers a more isolated and concentrated approach to each component, which ultimately aids in better understanding.
When you make changes to an entity or its relationships, the changes can be made in either the Entity Relationship Diagrammer or RON. The changes will not be reflected on other diagrams that use the same entities. There might be a reason for the user to keep the changes off. The changes might reflect an earlier point in time or reflect the business opinions of another department. However, it is possible to take into account changes from other users and consolidate them into the current diagram. To do so, you must choose Edit Consolidate from the menu bar. At this point, you can choose whether to consolidate a specific entity, relationship, or the entire diagram.
Page 783
NOTE |
If you do not consolidate an element that has changed, you cannot edit it. |
There are several ways to make the diagram easier for your associates (the developer or the customer) to follow. You can customize your diagram with various text options and line widths as well as color .
In all of your digrams, coloring various items on your screen, changing the fonts of words, varying the line width of outlines, relationships, and so on, can be useful techniques, both for you and for others who read your diagram. These changes operate similarly in the various Designer/2000 tools in this chapter, as you'll see in each section.
For example, for entity relationships, all Type entities (the ones that can represent domains) can be filled in with blue. Another example is to use colors to depict your comfort level with the completion of an entity. For instance, if there are still many outstanding questions, the entity would be red. If there are few questions, the entity would be yellow. If the entity were complete, the entity would be green. Interface entities would be blue.
NOTE |
The colors on the diagram show up only if you have a color printer. Lighter colors from the color palette usually work better with color printers, for maximum readability. If you are using a black-and-white printer, however, colors show up as different shades of gray, so one color differentiation can be used. |
You can change all of the diagram or just a specific entity, relationship, or combination. To change the entire diagram, choose Edit Select All from the menu bar. To select one or more entities or relationships, select one and hold down the Ctrl key until all the items are selected.
To change the line width, select the Line Width icon. Choose a line width from the items listed. To change the line color, choose the Line Color icon. Choose a line color from the colors
listed. To change the fill color of entities, choose the Fill Color icon. Choose a color from the items listed. To change the font of any text selected, choose the Text icon. The font, the font style, and the size of text all can be changed.
Page 784
From Edit Navigate on the menu bar, select either an entity or a relationship. The cursor selects the item requested, and the focus of the screen includes the item requested .
TIP |
If the screen has been minimized, the Navigate function might not appear to work. Enlarge the screen and retry . |
This section provides a brief overview of how to model information using the Functional Hierarchy Diagrammer. Figure 32.1 shows a sample setup in the Function Hierarchy Diagrammer tool and some of the toolbar icons you'll be learning in this section.
Figure 32.1.
A sample layout in the
Function Hierarchy
Diagrammer.
You can create functions on the diagram or bring them in from the Repository Object Navigator. To create a function on the diagram, select the Function icon. After selecting the icon, click a place on the diagram where the function should be placed. The system will prompt you for a short name and title for the function.
Page 785
TIP |
If you want to create more than one function without having to select the button each time, hold down the Shift key when you select the Function button. When you are ready to move on to another activity, press another button on the icon bar. |
The short name of the function can contain information about the function ”such as ACCOUNT for "Keep track of user accounts" ”or it can represent the functional breakdown of the function within the hierarchy, such as REP for all reports , and REP01, REP02, and REP03 for each report. I use a two-digit application abbreviation followed by either an R for report, S for screen, M for menu, U for utility, and use the remaining spaces with MORG for Maintain Organization. There are many valid standards available. Choose one that works best for you and your organization.
CAUTION |
In the Repository Reports, the Function Definition Report has an option for listing all functions on a diagram. Currently, this option does not work unless you give all functions a short name that starts with the same alphanumeric string. Make sure that no function outside this diagram starts with this string. You can ask for the report by functions that start with the string. |
NOTE |
The short name of a function is the default for the short name of a module, which is the default for the name of a Windows file. Therefore, the name should be limited to eight characters to conform to Windows standards. |
After you add the function, you can include other information about the function by double-clicking within the confines of the function box. The Edit Function window pops up, in which you can edit the definition of the function, any notes to be added, the frequency of use, whether the function is elementary and/or atomic, the entities used by the function, the attributes, the CRUD (Create, Retrieve Update, Delete usages of the entity and attributes), and triggers.
To enter or edit entities and attributes, select the entity table from the Edit Function box. Select the entity/entities from the list that are included in the function. Enter the CRUD for each entity. Next, select the Attribute tab while the cursor is on the entity for which to enter attributes. The entities' CRUD will be reflected in each attribute selected. Edit the CRUD for each attribute. Continue this process until all entities and their attributes have been inserted. Once you have left the Entities tab, when you return the entities will be placed in order alphabetically .