1766   RSYN Rel mandatory?

Created: 05 Mar 2021

Status: Solution Accepted

Part: Part 7-4 (2020; Edition 2.1)

Links:

Page: 294

Clause: 6.13.12

Paragraph: Table 151

Issue

Presence condition for DO "Rel" is "AtLeastOne(1)/F" but there is no second DO marked AtLeastOne(1).
Does this mean that "Rel" is actually mandatory?

Note that CSYN has both DO "Cmd" and Rel marked AtLeastOne(1)

Proposal

Make RSYN.Rel mandatory?

Discussion Created Status
Ballot period expired without comment.
Move to solution accepted.
08 Jan 24 Solution Accepted
Starting ballot period 29 Nov 23 Ballot Period
No TP change, NSD update will bring this into testing. 28 Nov 23 Conformance Test Verification
Change to state: Conformance Test Preparation 17 Oct 23 Conformance Test Preparation
The technical error is a correction caused by an UML modelling/auto-generation issue. Only the expression of the presence condition is affected. No weaker or stronger presence condition is specified.

In summary: no compatibility issues.
10 Oct 23 Analysis Of Compatibility
Draft is attached. Changes marked in yellow.
Change state to verify draft implementation.
26 Sep 23 Verify Draft Implementation
final proposal:move DO Rel to RSYN (M/F).
In CSYN DO Rel and Cmd will have presence condition AtLeastOne(1)/F.

--> drafting implementation
22 Apr 21 Drafting Implementation
I agree with the proposal to move DO Rel to RSYN (M/F).
In CSYN DO Rel and Cmd will have presence condition AtLeastOne(1)/F.
The note will not be necessary.
08 Apr 21 Discussion (red)
The easiest and clearest way to solve the issue: Remove Rel from Abstract Class; add Rel to CSYN with AtLeastOne(n)/F and to RSYN with M/F
We have other cases for instance @ Protection where Str or Op could not be inherited from ProtectionLN Abstract class because they have sometimes with presence condition M and sometimes O. The definition of Str resp. Op was moved to the abstractClasses with the proper presence condition (M or O); this was inherated further on.

08 Apr 21 Discussion (red)
Presence conditions are normative specified in part 7-1, section 5.6 (i.e. in part 7-4 they are informative).

Proposal for part 7-4:
Add a note to section 6.13.12.:
The presense condition "AtLeastOne(1)/F" of data object Rel should be interpreted as mandatory/forbidden.
07 Apr 21 Discussion (red)
That is an interesting point. I guess that might happen on multiple cases, not only related to RSYN.

First of all, it seems to be a weakness of our UML model. It should be possible, in a specialized LN to make presence conditions stronger than in the abstract one.

But to handle it, as it may happen at other places, I would rather make a clarification where the presence conditions are explained; i.e. clearly stating that if an "AtLeastOne(n)" condition is applied to only one element, that element is mandatory.
07 Apr 21 Discussion (red)
Change to Discussion 07 Apr 21 Discussion (red)
My proposal:
Add a note to section 6.13.12, below table 151:
The presense condition "AtLeastOne(1)/F" of data object Rel should be interpreted as mandatory/forbidden.
16 Mar 21 Accepted

 

Privacy | Contact | Disclaimer

Tissue DB v. 23.12.13.1