Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!rutgers!ames!ucbcad!zen!xanadu!larry
From: larry@xanadu.uucp (Larry Rowe)
Newsgroups: comp.databases
Subject: Re: Call Interfaces to RDBMSs
Message-ID: <3209@zen.berkeley.edu>
Date: Thu, 6-Aug-87 12:36:56 EDT
Article-I.D.: zen.3209
Posted: Thu Aug  6 12:36:56 1987
Date-Received: Sat, 8-Aug-87 11:09:10 EDT
References: <168@bernina.UUCP>
Sender: news@zen.berkeley.edu
Reply-To: larry@xanadu.UUCP (Larry Rowe)
Organization: Postgres Research Group, UC Berkeley
Lines: 36

In article <168@bernina.UUCP> marti@ethz.UUCP (Robert Marti) writes:
>The prevalent approach to offer set-oriented access to a relational
>database from an application written in a langauge such as C seems
>to be pre-compilation.   For example, embedded SQL uses database
>statements bracketed by EXEC SQL and END-EXEC keywords while embedded
>QUEL uses the ## marker on the beginning of each line.  The pre-
>processor strips these statements and replaces them by procedure
>calls to the DBMS.
>	.....

i think you could define such an interface for any of the major
DBMS's that are offered commercially.  the preprocessor that
you mention just translates from a ``user friendly'' notation
to an ``ugly'' subroutine call interface.  you could define
your own subroutine call interface that in turn did calls on the
vendor supplied routines.

a couple of things to look out for though:
1. most vendors don't guarantee compatibility at the subroutine
call level between releases.  (it doesn't change often, but changes
are needed periodically to add functionality and fix bugs.)

2. this solution assumes that the subroutine call interface can be
linked into your program.  on the vax and 68K/unix this is possible
because all languages use a standard subroutine call interface.  i don't
recall what machine you are using or what modula compiler you are
using, but this might present a problem.

finally, as to whether any commercial database vendors publish the
protocol between the application program and the dbms program (typically
an IPC protocol), i don't know of anyone who is doing this yet, but
i expect there will bbe many people arguing over this issue as 
distributed heterogenous dbms's become more widely used.  there is
an ISO standards committee looking into this, but i don't get the
sense that anyone is paying any attention to them.
	larry