Subject: Proposals for data structures
From: Peter Ashenden (peter@ashenden.com.au)
Date: Fri Aug 01 2003 - 05:23:57 PDT
Folks,
The TBV group has a number of proposals before it for what amount to
abstract data types: fifos, lists, sparse arrays, associative arrays. The
proposals suggest including new built-in types.
Seems to me that many of these types can be implemented using the facilities
already in the language. If there are aspects that cannot be effectively
implemented, perhaps we should seek to provide improved language features
for abstract data types in general, rather than specific solutions for
specific types.
Again, I'd like to ask for the motivating requirements. What modeling tasks
are made difficult with existing language features?
Are we just looking for a variety of predefined collection structures so
that verification engineers don't have to reinvent them for each project?
If so, we should define a standard library of collections, akin to the C++
STL, or the Java collections library.
One problem with doing this is that we can't currently make a collection
type generic with respect to element type. I think a general purpose
mechanism to solve this issue would be to provide type generics and to allow
generics in packages. This would be a solution that would work for all of
the structures proposed, plus for those we haven't thought of yet.
Comments?
PA
-- Dr. Peter J. Ashenden peter@ashenden.com.au Ashenden Designs Pty. Ltd. www.ashenden.com.au PO Box 640 Ph: +61 8 8339 7532 Stirling, SA 5152 Fax: +61 8 8339 2616 Australia Mobile: +61 414 70 9106
This archive was generated by hypermail 2b28 : Fri Aug 01 2003 - 05:24:34 PDT