Once the data has been retrieved and any correlatives have been performed, the system can perform additional manipulative functions on the intermediate form of the data using the function codes specified in the dictionary attribute-defining item, output-conversion. When those functions, if any, have been completed the data is in output format.
If there is a processing code in both the correlative and output conversion, the correlative result does not feed into the output conversion as its input. The correlative is processed and used in any sort or comparison operation. If there is no output conversion, the correlative result displays. If there is an output conversion, the result of the correlative is discarded, the output conversion is processed, and the result displays.
If you use MultiValued processing codes, the result of the last processing code can be used as input data by the next processing code.
More than one processing code can be specified in the attribute. Separate each processing code with a value mark.
All arithmetic done within correlatives is integer arithmetic. To use values with decimal places such as 41.25, 3.55, 8.00, 22.25, declare a fixed decimal position for all data within the file using the mlnm or mrnm code. The data can be stored as 4125, 355, 800, 2225, and so on. When the values are retrieved, the decimal place is reinserted.
If there are internal values with a varying decimal point embedded in the values such as 4.125, 3.5, 8, 22.25, scale the values from the 000.00 form to a 00000 form, perform the arithmetic, and descale it back to the 000.00 form.
The following f code scales a number with up to five decimal places:
|
Fx(G0.1);x(G1.1);(S;*‘0’);(ML%5);: |
where x is the attribute-mark count.
See Also