Demand more from your SGML database!
		Bringing SQL under the SGML limelight

			     Arijit Sengupta

Abstract

Have you ever been frustrated by how inadequate SGML databases are in
terms of searching or querying your documents? With the current state
of the art, you will easily be able to search for a word, phrase, or
keywords in the whole document. Some systems let you perform
approximate searches or regular expression searches. Even fewer
systems let you search for keywords or phrases in certain SGML
regions. However, there is much more information already in SGML
documents that one can utilize cleverly to design a proper SGML
database system. The current trend of modeling SGML documents with
object-oriented and object-relational databases has certainly brought
SGML closer to a complex object database model, but much research and
development remains to be done in this area. This article introduces
the popular relational database query language SQL (Structured Query
Language) and its applicability in the SGML domain.The capability of this
query language to express complex queries with a not-so-complex syntax
gives relational databases that support SQL an advantage over other
similar systems. The ability to use SQL or an SQL-like query language
with SGML has the potential of giving much more power to SGML
repositories. This article shows how we can pose complex
document-related questions easily with SQL. SQL-capable systems will
let you solve problems that would otherwise seem impossible, or at
least, tedious.