Browse Prior Art Database

Compilation of Nested Decision Tables

IP.com Disclosure Number: IPCOM000088835D
Original Publication Date: 1977-Aug-01
Included in the Prior Art Database: 2005-Mar-04
Document File: 2 page(s) / 13K

Publishing Venue

IBM

Related People

Larson, LE: AUTHOR

Abstract

A processor provides a quick and easy scheme to create code productions for processes described in more than one decision table, and provides the fastest linkage technique between the processes described therein. The logic strings produced satisfy the conventions of structured programming from a concise table formatted program specification. Since the resulting code is completely label free, it can be incorporated into any program without concern for sequence label conflict. The processor creates optimized code strings which satisfy the theory of structured coding. Thus, the code is easy to read and maintain. In addition, the code is produced by machine at a rate much faster and more reliable than can be accomplished manually. The net result is that the total cost of code production is reduced.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

Compilation of Nested Decision Tables

A processor provides a quick and easy scheme to create code productions for processes described in more than one decision table, and provides the fastest linkage technique between the processes described therein. The logic strings produced satisfy the conventions of structured programming from a concise table formatted program specification. Since the resulting code is completely label free, it can be incorporated into any program without concern for sequence label conflict. The processor creates optimized code strings which satisfy the theory of structured coding. Thus, the code is easy to read and maintain. In addition, the code is produced by machine at a rate much faster and more reliable than can be accomplished manually. The net result is that the total cost of code production is reduced.

To implement this idea, a programming problem is expressed as a set of decision tables such that some of their stubs make a reference to other tables by having action stubs which read "INCLUDE table name". The decision tables are presented to the processor one at a time and each table is compiled into an internal text. The internal text is constructed in such a manner that the nesting of logic conditions is maintained. Furthermore, the interaction of logic conditions and program actions is likewise maintained. Each set of actions to be executed for a given set of conditions is bounded by a DO-END pair and each condition is translated to a true path and false path pair bounded by an IF-THEN-ELSE triplet.

The code of each paired true path and false path is examined for common code within corresponding DO-END pairs. If the common code is at the beginning of the DO-END pair, then the code is placed prior to the IF-THEN- ELSE triplet. If the common code is at the end of the DO-END pair, then the code is placed after the DO-END pair on the false path. The redundant code string is then removed from the text. This entire process is accomplished by reading the program text in a rear-to-front fashion. Each such compilation results in saving...