Browse Prior Art Database

Automatic Query Generation

IP.com Disclosure Number: IPCOM000120408D
Original Publication Date: 1991-Apr-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 2 page(s) / 59K

Publishing Venue

IBM

Related People

Loe, PA: AUTHOR [+2]

Abstract

A software technique is disclosed that removes the need for the user of a relational database to understand the concepts and syntax of relational algebra.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 55% of the total text.

Automatic Query Generation

      A software technique is disclosed that removes the need
for the user of a relational database to understand the concepts and
syntax of relational algebra.

      When a user defines a query on data held in a relational
database he has to define two different things. First he must define
what is wanted and then how to get it. It is reasonable to expect him
to define the first, but the second can be automated by putting
intelligence into the machine to process his query.  There are
several standard parts to a query which are listed below.
1.   Identify the various tables where the data is located.
2.   Describe which of the many standard relational
     operations are to be used to join the tables together.
3.   Describe how the data rows are to be combined to form
     new rows.
4.   Identify which rows of the resultant table are
     required.
5.   Identify the order that the results are to be presented
     in.
6.   Identify which columns of the resultant table are
     required.

      In theory, only the last three items represent the information
that the user MUST supply since these actually represent WHAT is
required.  The other steps all represent HOW the result is to be
obtained.  It is these three steps which require the user to possess
a knowledge of relational algebra, computer processes and, often,
complex language syntax.  To overcome the problems of these first
three steps is the s...