Parent Child Controls
The parent child control is a composite control with a tree on the left and a grid on the right. The control is used to provide a hierarchical view of the business data. Users can resize the tree part of the control during runtime. The tree and grid portions use the same business view, if one is attached. If an editable parent child control links to a business view and the user has edited the content, the change is saved by runtime when the form is saved.
At most, a form can contain a single parent child control. Additionally, a form cannot contain both a parent child control and a grid control. You cannot place a parent child control on a tab page; although you can nest a reusable subform containing a parent child control on a tab page.
You can add data dictionary (DD) items to the grid. Because of its similarity to a grid control, many of the properties for parent child controls are the same.
After you configure the control, you set up either a parent/child relationship or use event rules (ER) to load child nodes to the tree. The method that you use depends on whether the table has an inherent parent/child relationship.
You can use ER and system functions to customize the way in which the parent child control functions. For example, you can change the top node of a tree or change the node that appears as the first child on a tree.
To increase performance during runtime, parent child controls can make use of page-at-at-time processing. Page-at-a-time processing ensures that each fetch fetches only one page of data. Page-at-a-time processing is the default mode for all parent/child forms. During page-at-a-time processing in standard mode, the page size is the number of nodes that can fit in the current view. When the Find process begins, only one page of first-level nodes is fetched from the table and inserted in the tree. When the user scrolls down, a new page of data is fetched.
Page-at-a-time processing for the expand-all style of parent/child forms is similar to that for standard mode. When the Find process starts, one page of first-level nodes is fetched from the table and inserted into the tree. Then, all of the first-level nodes are expanded. Each expansion fetches only one page of data. Because the tree expands exponentially in expand-all mode, a deep tree might affect performance.