Browse Prior Art Database

Virtual File Access to a Binary Relational Database

IP.com Disclosure Number: IPCOM000045702D
Original Publication Date: 1983-Apr-01
Included in the Prior Art Database: 2005-Feb-07
Document File: 2 page(s) / 14K

Publishing Venue

IBM

Related People

Pullin, DJ: AUTHOR [+2]

Abstract

This article describes a mechanism for obtaining access to a binary relational database via program operations which are equivalent to record input/output operations on a conventional file. This mechanism allows a simple program to load a conventional file into a binary relational database, or unload a part of a database into a file. It also permits the simple migration of existing application programs, which access files, to operate on a database.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 2

Virtual File Access to a Binary Relational Database

This article describes a mechanism for obtaining access to a binary relational database via program operations which are equivalent to record input/output operations on a conventional file. This mechanism allows a simple program to load a conventional file into a binary relational database, or unload a part of a database into a file. It also permits the simple migration of existing application programs, which access files, to operate on a database.

Query operations on a binary relational database have been described in (1,2). A query is defined as a regular, structured collection of data items extracted from a database, and is therefore analogous to a conventional file of data records. In this article, a query is redefined as a "virtual file" which may be materialized from a data-base, data-base, although it does not exist as a physical entity within the data-base. Alternatively, records from an external source may be inserted into a virtual file, resulting in update operations on the database, although the external record is not stored as a physical entity.

Before a program can access a virtual file, some preparatory actions must be performed: A. Define a query over the database by one of the known

methods.

B. Associate with each column in the query a field name, field

width and field offset within a virtual record.

C. Store the description of the query and the description of

each virtual field (i.e., the results of steps A and B) in a

data dictionary.

D. Associate a virtual file name with the query, and store this

in the data dictionary.

Step C has the effect of linking a description of a source of data (the query) with its intended destination (the output record). Step D provides a name which may be used in an application program to refer to the virtual file so defined. Existing methods do not permit these associations to be made.

The data dictionary which is used in steps C and D should be the same data dictionary which is used to contain the description of the binary relational database structure. It is then possible to record the description of the required query as a search path which refers to specified database sets and database relationships.

The major program operations on a virtual file and the methods by which these can be implemented are: 1. OPEN virtual file name

2. CLOSE virtual file name

3. GET virtual-record FROM virtual file name

4. PUT virtual-record INTO virtual file name

5. DELETE virtual-record FROM virtual-file-name

6. INSERT virtual-record INTO virtual-file name.

1

Page 2 of 2

1. The OPEN operation searches the data dictionary for the specified virtual file name, and returns an error indication if it is not found. Otherwise, it retrieves the query definition from the data dictionary and holds it in m...