Subject: Re: SDL-News: context parameters
From: Rick Reed TSE (rickreed#tseng.co.uk)
Date: Sun Oct 03 1999 - 11:35:38 GMT
The originator of this message is responsible for its content.
-----From "Rick Reed TSE" <rickreed#tseng.co.uk> to sdlnews -----
> I am doing a semantic analyzer of SDL based on the book 'Systems Engineering
> Using SDL-92'.
I think that it is only fair to inform you that this book is already out of
date, because there was a addendum to SDL-92 agreed by ITU-T in 1996.
However, this essentially mode some extensions SDL-92 and therefore did not
invalidate SDL-92 models: there may be a few exceptions - but I can't think
of any. Currently the book still applies, but describes a subset of the SDL
standard that is in force.
For a semantic analyser you should really work from the ITU-T Z.100 standard
which can be purchased on-line direct form ITU-T with a credit card if
necessary. The standard is a reference for the language, and you are wise to
use the SDL-92 book for explanation and background information. There is a
formal definition published by ITU-T (annex F to Z.100 in three parts F.1,
F.2 and F.3) that defines what is given in the main body of Z.100 by syntax
and natural language in a formal notation (MetaIV). THis could also be
useful.
The last news that I have for you is that SDL is on target to be updated
again in November this year to "SDL-2000". There are some significant
changes in this new version and SDL-92 models may need to be "translated" to
be SDL-2000 compliant. A few (troublesome or rarely used) SDL-92 features
are not supported in SDL-2000. A brief list of these are: alternate block
substructures and channel substructures (blocks in blocks are still
supported of course), graphical macros, data axioms (but there a more
practical and powerful object data scheme), view definition (but "global"
data can now be defined).
SDL-2000 has many new features. Block and processes can be used more or less
interchangeably. The powerful new data model with: object types, methods,
inheritance of algorithmic operations, choice types, extendable structures.
Links to object modelling: class-like icons, associations. Composite states.
Exception handling. Interface descriptions.
You can get access to the draft new standard by becoming a member of the SDL
Forum Society - it is available on the members only ftp site.
There is also a new formal definition in preparation. This is scheduled to
be agreed by ITU-T in spring next year. It defines SDL-2000.
> I have a question (a lot, really):
>
> In procedure formal parameters, is it posible that a parameter is neither 'in'
> nor 'in/out'?
> In afirmative case, what does it mean?
If you omit IN or IN/OUT then this is the same as specifying IN.
In SDL-2000 you can also have an parameter which is just OUT, in which case
no value is transferred to the procedure.
By the way these are parameters - not "context parameters" which are
something different.
-- Rick Reed - rickreed#tseng.co.uk Tel:+44 1455 55 96 55 Fax:+44 1455 55 96 58 Mob.:+44 7970 50 96 50-----End text from "Rick Reed TSE" <rickreed#tseng.co.uk> to sdlnews ----- Join http://www.sdl-forum.org/Society/members.htm for extra SDL Forum Society benefits For help, email "majordomo#sdl-forum.org" with the body of your email as: help or (iff this does not answer your question) email: owner-sdlnews#sdl-forum.org
This archive was generated by hypermail 2a23 : Sun Jun 16 2013 - 10:41:41 GMT