A Simple Computer Data Base System for UNIX.
Abstract
This computer program that allows users to maintain and access a file containing addresses, inventory items, or other units of text information grouped in blocks separated by blank lines. Any string within a file may be used to find and print the block(s) of text containing the string. A file is created, maintained, and accessed by a group of UNIX programs which have been designed for speed and simplicity. Besides being useful in themselves, these programs illustrate cooperative use of C programs and shell command files. A history of the development will also be given since this is of general interest to programmers. This program uses standard UNIX techniques, except for the Boyer-Moore string matching algorithm. It offers a simple and extensible approach to the type of database represented by the rolodex file found in many offices. This simple flat-file database has proven valuable as a way of maintaining and accessing an inventory file and an address file. The data file is a plain text file containing no control characters aside from new lines. The file is therefore easy to create and maintain using ordinary text editors, though a program is provided to facilitate item entry for users. On an unloaded VAX 780 it takes about 1.5 seconds to search a data file of 150,000 characters. On a system which is about 10 users, this time is about 3 seconds.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1988
- Accession Number
- ADA193375
Entities
People
- Michael M. Mcdonnell
Organizations
- Geospatial Research Laboratory