A variable is a programming object defined in the Object Dictionary for the application.
Variables are used to store values temporarily during an application session.
A variable is assigned a size and a data type in its definition, but until assigned a value during an application session, the variable is valueless (that is, $Null).
A named constant is a programming object defined in the Object Dictionary for the application. The constant’s value is part of its definition.
By using named constants in place of literal constants in application programs, you ensure that a constant’s value can be quickly, easily, and universally changed, if necessary.
The Alt+F6 keystroke combination cycles the user through all available windows if the current window is not modal. Its effect is similar to clicking the mouse to make a non-current window current.
The Ctrl+F6 keystroke combination cycles through all available sibling windows of a child window.
Within a quoted character string, a backslash immediately followed by two characters that are either digits from 0 to 9 or letters from A(a) to F(f) are treated as a hex code. The text string containing the hex code must be enclosed in quotation marks.
set output printer
output “ F”; % selects a condensed font
…report commands…
output “12”; % restores regular font
set output terminal
The results of value expressions are rounded automatically, but only after the entire expression has been evaluated. (During evaluation, full decimal values are used.)
The final result takes on the number of decimal places found in the operand with the most decimal places.
The rounded result is further rounded if it is being assigned to a named constant, a variable, a field, or a form field defined to contain fewer decimal places.
To force rounding in particular circumstances, use the $round function.
Examples of Rounding
5/2
Result is 3.
5/2.0
Result is 2.5.
1+1.01
Result is 2.01.
1.01+(5/2)
Result is 3.51.
Integer=1.01+(5/2)
Result is 4 if Integer is defined to contain no decimal places.
The available pattern symbols for constructing date validation rules are
The available syntax characters for constructing date validation rules are
| – | Denote a range of dates. |
| | | Separates alternative dates or ranges of dates. |
Examples of Date Validation Rules
Date values are stored in the format YYYYMMDD. Use DataMask to control how the date is to be input or displayed in the field. So the rule can be expressed as follows:
19990101-19991231
Input values that are accepted as legal include any date during 1999.
When certain atomic expressions (variable name, field name, form field name, or formal parameter name) are used before being assigned values, or if the values are unknown, those expressions are considered to be $Null.
$Null is a property indicating a state of valuelessness; $Null is not itself a value. The null string is a value and is therefore not the same as the $Null property.
An arithmetic or functional expression that includes a $Null argument or operand always evaluates to $Null.
A logic expression that includes a $Null operand always evaluates to $False.
Because $Null is not a value, you cannot use the conditional operators = (Equals) and <> (Not Equals) to determine if an expression is $Null. Instead, the operator IS [NOT] [$]NULL is used as shown in the following example:
Salary is $null
Salary is not $null
Note: $Null can be written without the dollar sign.
Names for objects defined in the Object Dictionary must conform to the following conventions:
Examples of Object Names
Examples of valid names are
Compare these to the following invalid names:
The available pattern symbols for constructing numeric validation rules are
The available syntax characters for constructing numeric validation rules are
| – | Denote a range of numbers. |
| | | Separates alternative numbers or ranges of numbers. |
Examples of Numeric Validation Rules
The following table contains examples of numeric validation rules, and the values that can be legal or illegal under those rules:
| Validation Rule | Legal Value | Illegal Value |
| 1|2|3|5 | 1 or 2 or 3 or 5 | 0 or 4 or 6 |
| 10-20 | From 10 to 20, inclusive. 10.123 is also legal. (Use DataMask to control the number of decimal places permitted in the field.) | 0-9 or 21 |
| 10-20|100-200 | From 10 to 20, or from 100 to 200, inclusive | 0-9 or 21-99 or 201 |
| -30–5|1|1.3-1.7 | From -30 to -5, or 1, or from 1.3 to 1.7 | -4-0 or 1.8 or 2 |
The syntax of a call to a procedure is
procname («expression»)
| procname | The name assigned to the procedure in its PROCEDURE or LOCALPROCEDURE command. |
| expression | Any value expression. |
The list of expressions must be enclosed in parentheses; even when no expressions are required, the parentheses must still appear.
At the prompt level, you can issue a call to the main procedure of any procedure program.
Within a procedure program, you can issue a call to
Any procedure in a procedure program can issue a call to itself.
In a call to a procedure, you must supply one expression for each parameter specified in the PROCEDURE or LOCALPROCEDURE command that describes the called procedure. The number and type of expressions listed in the call to a particular procedure must match the number and type of parameters in the PROCEDURE command in order for the procedure to execute. The value of expression can be assigned to the corresponding parameter (IN parameters), or expression can take on the value of the corresponding parameter when the procedure ends (OUT parameters), or both (INOUT parameters).
When you specify more than one expression, each must be separated from the next by a comma.
See Also
English 
Portuguese