z/OS Block Mode Screen Design › External Data Representation › Algebraic Notation Rules

Algebraic Notation Rules

When the input of a numeric field cannot be matched to the supplied edit pattern pictures (positive, negative, or zero), then standard algebraic and quantity notation is attempted as a last course of action. If this fails, then the string is marked as error, and the user is prompted to specify the value again.

The following list shows the rules for interpretation of an input value that does not match any pattern:

• If the edit pattern indicates a signed quantity (a format 2 or 3 edit pattern), the input value is checked for the presence of a leading or a trailing sign, specified using the “+” or “-” characters. If either is found, then the sign is set accordingly and the character is removed from consideration in the input string. If no sign character is found, then the sign is assumed to be positive. If the variable is unsigned, then no check is made for a leading or trailing sign.
• Prefix and postfix literals are not allowed, and if present, fail the edit.
• Digits are extracted from the input string until an infix literal character is encountered or the end of the input is reached. If an infix literal is encountered, then it must match an infix literal in the supplied edit pattern, and the data is aligned over the edit pattern on that infix literal character. This supplies decimal point or separator alignment. After it is aligned, the input digit string must match the digit and infix literal specifiers of the edit pattern from that point forward.
• If no infix literals are encountered in the digit string, then the value is assumed to be an integer value. If assignment is to a variable that contains decimal digits, then the input value is aligned to the indicated decimal position.

The following table shows how the system interprets data that does not match a supplied pattern. The table shows various positive, negative, and zero edit pattern pictures. For each pattern, the table shows which pattern is matched, its variable width and decimal digits (w,d), and the resulting value extracted.