Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!linus!philabs!cmcl2!harvard!seismo!umcp-cs!nbs-amrf!libes
From: libes@nbs-amrf.UUCP (Don Libes)
Newsgroups: net.lan
Subject: BBN's implementation of X.409 (ANSI/ISO 8824, ASN.1)
Message-ID: <310@nbs-amrf.UUCP>
Date: Tue, 20-May-86 12:50:43 EDT
Article-I.D.: nbs-amrf.310
Posted: Tue May 20 12:50:43 1986
Date-Received: Sat, 24-May-86 00:20:28 EDT
Distribution: net
Organization: National Bureau of Standards
Lines: 25

Upon obtaining BBN's implementation of X.409, I discovered that
ASN.1 is not X.409, although they are very close (ASN.1 is based on
X.409).

I have made the modifications to BBN's X.409, to bring it into line
with ASN.1 (as well as fixing one bug).  If anyone is interested,
let me know.

To my dismay, the BBN implementation, while complete, did not
implement the proposed ENCODE and DECODE extensions to the grammar
compiler.  These extensions would make it so much more useful.
If anyone has done this (hey, this time I already asked around NBS
first!), let me know, otherwise I will try and do it.

Don Libes        {seismo,umcp-cs}!nbs-amrf!libes

For your information, the ENCODE and DECODE extensions allow
associating C-source actions to the grammar.  It resembles yacc.
The result is that you can generate the machine independent
byte-stream directly from a C-structure by encode() with an
extended ASN type spec.  Similarly, a byte stream can be read,
parsed and stored directly into a C structure.

Without this feature, you must write a conversion routine for each
C-to-ASN structure and another to go the other way.  Ugh.