Browse Prior Art Database

Legacy Data Encapsulator

IP.com Disclosure Number: IPCOM000014487D
Original Publication Date: 2001-Jun-14
Included in the Prior Art Database: 2003-Jun-19
Document File: 2 page(s) / 54K

Publishing Venue

IBM

Abstract

Disclosed is a method for accessing data files from legacy computer systems. All of the properties of the data elements(e.g. data type, length, offset, etc.) are encapsulated in a single software construct. Software modules wishing to access this data only need to request the data element by referencing the field name that identifies the data element. The software construct returns the properties of the data element and the value of the data to the requesting software module. In this way, all of the details of the underlying legacy data storage systems are hidden from the requesting software module. Problem Description Many software applications that are critical to the business of commercial enterprises use data files whose design predate the widespread use of relational databases. That is, the data files consist of tens and sometimes hundreds of distinct files. These files for the most part have different record lengths, distinct mappings of fields within records, and various data types for the distinct fields. Typical data types are string data, numeric data of varying lengths and format such as integer, short, long, double, floating point, boolean, packed, date, time, and timestamp. Applications that access these files typically must know all the details about each file: its record length and storage method (e.g. sequential, indexed, relative, etc.), its field definitions, data types, lengths and format. This information is typically part of any program that accesses one or more of these files. This replicates the same information many times over in many different programs, leading to:

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 1 of 2

Legacy Data Encapsulator

Disclosed is a method for accessing data files from legacy computer systems. All of the properties of the data elements(e.g. data type, length, offset, etc.) are encapsulated in a single software construct. Software modules wishing to access this data only need to request the data element by referencing the field name that identifies the data element. The software construct returns the properties of the data element and the value of the data to the requesting software module. In this way, all of the details of the underlying legacy data storage systems are hidden from the requesting software module.

Problem Description

Many software applications that are critical to the business of commercial enterprises use data files whose design predate the widespread use of relational databases. That is, the data files consist of tens and sometimes hundreds of distinct files. These files for the most part have different record lengths, distinct mappings of fields within records, and various data types for the distinct fields. Typical data types are string data, numeric data of varying lengths and format such as integer, short, long, double, floating point, boolean, packed, date, time, and timestamp.

Applications that access these files typically must know all the details about each file: its record length and storage method (e.g. sequential, indexed, relative, etc.), its field definitions, data types, lengths and format. This information is typically part of any program that accesses one or more of these files. This replicates the same information many times over in many different programs, leading to:
1. increased maintenance cost when file definitions change to modify the same information in many places;
2. increased incidence of errors from multiple instances of the same information maintained by hand;
3. increased complexity of development and maintenance costs to propagate file access programming into business applications.

The Legacy Data Encapsulator addresses these problems.

Disclosure Description

The Legacy Data Encapsulator defines all file definition data for any number of files in one place. In object-oriented programming environments, the Encapsulator would be a class containing information about each field in each file managed. In non-object oriented environments, the Encapsulator could be a module with function calls. The information for each field is indexed by a unique name, qualified by the file name, so that each field in each file in the legacy "database" could be identified uniquely.

For each field, all information necessary to read and write the data for that field is maintained in the Encapsulator. This information includes: the unique field name, the data type of the field (see Problem Description above), the length of the field, the offset of the field within the record, and any other necessary information such as mask data for boolean values.

The Encapsulator provides methods or function calls t...