[Css-csts] extension-chaining
Martin Götzelmann
martin.goetzelmann at vega.de
Sat Aug 22 15:27:46 EDT 2009
Dear Tim,
My view is that extension of operations should follow the usual inheritance / derivation rules and that implies that a procedure can extend an operation defined or extended by the "grandparent" but not extended by its immediate parent procedure.
Given O is a common operation and Pn are procedures, I believe the following is legal:
P1 defines O1 by extension of O
P2 derives from P1 and adopts O1 unmodified
P3 derives from P2 and defines O2 by extension of O1
Of course P3 cannot extend O but only O2.
As discussed on the last telecon, we provide the option that the specification of a procedure can include the rule that an operation it includes must not be extended by derived procedures. This corresponds to a "final class" in java.
Kind Regards, Martin
________________________________
From: css-csts-bounces at mailman.ccsds.org on behalf of Ray, Timothy J. (GSFC-5830)
Sent: Fri 21/08/2009 11:44 PM
To: css-csts at mailman.ccsds.org
Subject: [Css-csts] extension-chaining
Dear WG Members,
The general question is this: "If a procedure does not make use of an extension provided by a lower level, does the extension-chain end at that procedure?" The next paragraph provides an example.
The generic Stop-Invocation provides an extension for use by the next higher layer (e.g. the Buffered Data Delivery procedure). The Buffered Data Delivery (BDD) procedure does not make use of that extension capability. Does that mean that the BDD procedure does not provide an extension for use by the next higher layer (i.e. procedures that are derived from the BDD procedure, and services that use the BDD procedure)?
Best regards,
Tim
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
More information about the Css-csts
mailing list