Browse Prior Art Database

Semi-Automatic Program-Generation Method

IP.com Disclosure Number: IPCOM000102480D
Original Publication Date: 1990-Nov-01
Included in the Prior Art Database: 2005-Mar-17
Document File: 8 page(s) / 295K

Publishing Venue

IBM

Related People

Mertz, RT: AUTHOR

Abstract

A technique is described whereby programs are semi-automatically generated, using an open architecture, whereby all of the constituted parts, such as read, write, select, match, summarize, reformat, etc., including the generated program, are accessible, readable, modifiable and maintainable. Described is a file processing program to illustrate how a semi-automatic PL/1 program generation is accomplished. However, the concept is not confined to file processing or to PL/1 programs.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 22% of the total text.

Semi-Automatic Program-Generation Method

       A technique is described whereby programs are
semi-automatically generated, using an open architecture, whereby all
of the constituted parts, such as read, write, select, match,
summarize, reformat, etc., including the generated program, are
accessible, readable, modifiable and maintainable.  Described is a
file processing program to illustrate how a semi-automatic PL/1
program generation is accomplished.  However, the concept is not
confined to file processing or to PL/1 programs.

      File processing programs, typically, involve operations such
as:  Read, Summarize, Write, Reformat, Select, Sort and Match.  Since
all of the operations, except Sort, deal with individual records, the
Sort operation is considered to be performed in separate runs.  The
objective of the concept is to provide a method of generating
file-processing programs, using an open architecture, such that:
   a) Programs generated will be accessible, readable, modifiable and
maintainable and thus be viably independent of their mode of
generation.  Although the generation process may not have the
capability of meeting every need in file processing, modifications
can be easily made to satisfy special requirements.
   b) New operation capabilities can be added to the process with
reasonable ease.
   c) The generation process is modular and open-ended, such that it
is also accessible, readable, modifiable and maintainable.
   d) The input to the generation of the program is in the form of a
sequence of template data set, one for each operation, which are
called and filled in by the user.  This sequence can then be saved
and modified to generate a new program so as to meet new needs.  The
templates themselves can also be modified, as needed.

      Generally, file-processing programs should be readily
accessible to the system, whereby functional modules perform the
various (macro) operations, such as read or write, along with any
associated housekeeping chores.  The functional module used in this
concept involves the use of a "control path", from which function
modules are entered and exit. Each step in a path is entered and
exited only once, so that control proceeds through the path and then
returns to the beginning of the path.  Fig. 1 illustrates the
simplest path for reformatting a file.  Somewhat less simple is the
Select operation. Here, a second path is introduced so as to address
selected records, as shown in Fig.  2.

      Another variation is the Summary operation.  Here again, a
second path is introduced for summary records, as shown in Fig. 3.
More complex is the Match operation on two or more files, as shown in
Fig. 4.  This arrangement lends itself to mechanical generation of
file-processing programs, since each step is appended to an existing
path (the come from concept) or begins a new path, or a combination.
Implementation may also be provided for semi-independent operatio...