COPY NAMED SELECTION

version 3

COPY NAMED SELECTION ({aTable; }name)

COPY NAMED SELECTION copies the current selection of aTable to the named selection name. The default table for the process is used if the optional table parameter is not specified. The parameter name contains a copy of the selection. The current selection and the current record of Table for the process are not changed.

A named selection does not actually contain the records, but only an ordered list of references to records. Each reference to a record takes 4 bytes in memory. This means that when a selection is copied using the COPY NAMED SELECTION command, the amount of memory required is 4 bytes multiplied by the number of records in the selection. Since named selections reside in memory, you should have enough memory for the named selection as well as the current selection of the table in the process.

Use the CLEAR NAMED SELECTION command to free the memory used by name.

The following example allows you to check if there are other overdue invoices in the [People] table. The selection is sorted and then saved. We search for all records where invoices are due. Then we reuse the selection and clear the named selection in memory. Clearing the named selection in memory is optional, in case the database designer wants to keep the sorted selection for future use:

   ALL RECORDS([People])
      `Allow the user to sort the selection 
   ORDER BY([People])
      ` Save the sorted selection as a named selection 
   COPY NAMED SELECTION([People];"UserSort")
      ` Search for records where invoices are due 
   QUERY([People];[People]InvoiceDue=True)
      ` If records are found 
   If (Records in selection([People])>0)
         ` Alert the user
      ALERT("Yes, there are overdue invoices on table.") 
   End if
      ` Reuse the sorted named selection 
   USE NAMED SELECTION("UserSort")
      ` Remove the selection from memory 
   CLEAR NAMED SELECTION("UserSort")

CLEAR NAMED SELECTION, CUT NAMED SELECTION, Identifiers, USE NAMED SELECTION.