ASF Freeware Quick Guide

Summary

The ASF distribution provides the minimal capabilities for making a set of documents searchable. It includes Isearch, a powerful and robust full-text search engine, and ASFserv, a fully-functional V3 Z39.50 server for searching and retrieving documents, including GILS-based data.

The search engine is connected to ASFserv through a flexible Application Programming Interface called SAI. SAI provides a generic, easy-to-use interface that is intended to be independent of any particular search engine. The ASF distributions include the Isearch search engine, however, the SAI is intended to allow for the use of other search engines, as desired.

More advanced capabilities include a Web-to-Z39.50 user interface to ASFserv and a Web-based crawler and harvester facility. With the Standalone distribution, you can index collections of documents on your own machine and make them searchable through a Web interface on your own server, and index collections of documents at other interesting Web sites, making them searchable through the Web, as well. Figure 2 depicts a typical Standalone Distribution configuration.

Standalone Distribution Configuration
Figure 1 - Typical Configuration

HtDig is provided as the Web-based crawler in the Standalone distribution. The interface between HtDig and the harvesting utility is a simple ASCII file containing a white-space delimited list of URLs. This interface is designed to allow for the use of other Web-based crawlers as desired.

Basic Features

Search local documents and GILS records with a Z39.50 client (you could use a web browser-enabled client such as the Z39.50 test client maintained at the Library of Congress, or you could obtain BookWhere 2000 for Windows 95/NT.)
Provide searching functions against databases of the following types:

ASF GILS records (ASCII files containing XML markup)
HTML files
ASCII text files

ASFserv and Isearch support the following search features

Simple keyword search
Boolean search (e.g. water and table)
Fielded search (e.g. search for all documents containing "power" in the Title)
Fielded search within any available GILS field
User-selectable search results formats (brief or full)

Automatically create GILS records for each document in a document set

Using global document set metadata template
For certain document types, using "fields" already present in the document (e.g. Title in HTML documents)

GILS records are editable (ASCII files containing XML markup)

Additional Features

Search available databases with a Web-to-Z39.50 client (Zap)
Zap supports the following search features

Plain text search
Boolean search (e.g. water and table)
Fielded search (e.g. search for all documents containing "power" in the Title)
Fielded search within any available GILS field
User-selectable search results formats (brief or full)
Optional thesaurus expansion
Proximity search (e.g. water within 1 word of table)
GRS-1 record syntax
SUTRS record syntax
Customizable search result formats

"Crawl" remote document location and build a list of documents for indexing