Subject: SDL-News: Data in SDL-2000
From: Rick Reed TSE (rickreed#tseng.co.uk)
Date: Fri Dec 11 1998 - 12:30:02 GMT
The originator of this message is responsible for its content.
-----From Rick Reed TSE <rickreed#tseng.co.uk> to sdlnews -----
The expert group for the SDL standard met in Heidelberg, Germany this week,
and finally agreed the basis for data in SDL-2000. It was agreed some time
ago that the data model would be changed for two reasons:
a) to overcome some of the limitations arising from the existing data model
(for example non-inheritance of DEFAULT data values);
b) to allow polymorphism and referencing.
The new model brings SDL data more in line with the other object/type
features of SDL (BLOCK TYPE, PROCESS TYPE etc.). It also makes data in SDL
easier to understand for someone who is familiar wiht data in a (so called)
object oriented programming language such as C++ or Java.
However, existing SDL data models (NEWTYPE sorts of data in the old
terminology; "value types" in the new terminology) are still valid in
SDL-2000, except where they rely on full implementation of axiomatic
descriptions. Two reasons for this change to equations are that equations
in SDL-92 have not been popular with engineers using SDL, and are difficult
to fully support with tools.
Value types in SDL-2000 are either the language defined types (as in
current SDL with Z.105: for example Integer, Boolean, enumerated types,
Octet, Character ...), or composite value types (such as Array, Struct,
String ...) constructed using other non-composite or composite types.
Operators can still be defined with a value type (as is possible in current
SDL), with the body of the operator described in algorithmic SDL or as
EXTERNAL (as allowed in current SDL).
The SDL-2000 model introduces "Object types" for data, that can refer to
"value types" and have polymorphic operators and methods.
Methods introduced with a type are applied to instances of the type using
the dot notation familiar from other languages. For example:
/*counter is a variable and increment is a method*/
Some other changes of syntax are being considered to make SDL-2000 data
easier to use (for example eliminating the keyword FPAR), but it has been
agreed that for legacy reasons the old syntax will still be valid, but
should not be used in new SDL-2000 models (it is redundant and obsolete in
SDL-2000, and may be deleted from the standard at some later date).
The Heidelberg meeting also considered several other changes to SDL,
including extensions which allow object class diagrams (as in UML) to be
directly incorporated into an SDL model.
A revised draft of the SDL standard will be prepared and input to the
February ITU study group meeting on 12 February. This draft will be for
"determination", which means that the study group determines that the draft
is in a satisfactory state for final voting on the standard to take place
at the next study group meeting (Novemeber 1999 in this case). The draft
can therefore be further elaborated and refined before the final version is
submitted for approval, but there should be no substantial changes to the
scope and content.
There was also a previous agreement that tool vendors would endeavour to
make trial implementations of SDL-2000 before it is finally agreed by ITU-T
so that the standard and tools are in alignment. Obviously, for this to
happen the February 1999 draft needs to be substantally stable.
The full report of the Heidelberg meeting, and all documents discussed will
be available to SDL Forum members as usual. If you wnat to follow the work
more closely, or participate in it then become a Forum member!
-- Rick Reed, TSE Limited 13 Weston House, 18-22 Church Street Lutterworth Leicestershire LE17 4AW United Kingdom Tel +44 14 55 55 96 55; Fax +44 14 55 55 96 58 Mob +44 79 70 50 96 50 email: rickreed#tseng.co.uk http://www.tseng.co.uk ftp://ftp.tseng.co.uk/tseng/
-----End text from Rick Reed TSE <rickreed#tseng.co.uk> to sdlnews ----- 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 - 10:41:40 GMT