Project

General

Profile

CIM Issues #5237

Eliminate compounds

Added by Herbert Falk over 2 years ago. Updated about 1 month ago.

Status:
Review
Priority:
High
Author/Contact Info:
Jim Hosrtman / jim.horstman@strateture.com
Base Release:
CIM 101
Solution to be Applied To:
CIM 101
Solution Version:
Solution Applied By:
Completion Date:
CIM Keywords:
61968-Common
Breaking Change:
Yes
Breaking Change Description:
The refactored classes are used in WG14 Part 3, 6, and 8 profiles.
CIM Impacted Groups:
WG14, WG16
Requestor:
Jim Horstman
Standard(s):
Version:
Clause:
Sub-Clause:
Paragraph:
Table:
Originally Closed in Version:
Origination Date:
Origination ID:
Originally Assigned To:

Description

Eliminate use of compound classes street address, street detail, electronic address, telephone number (2)

Proposed Solution

Remove attributes using compound class replacing with 0..* associations to relevant classes

Related issues

Related to WG14 Issues - CIM Issues #5318: Address modeling is too complexReviewActions
#1

Updated by Henry Dotson 6 months ago

  • Status changed from New to Open
  • Author/Contact Info set to Jim Hosrtman / jim.horstman@strateture.com
  • Base Release set to CIM 101
  • Solution to be Applied To set to CIM 101
  • Breaking Change changed from No to Yes
  • Breaking Change Description set to The refactored classes are used in WG14 Part 3, 6, and 8 profiles.
  • CIM Keywords 61968-Common added
  • CIM Impacted Groups WG14, WG16 added
  • CIM Impacted Groups deleted (WG13)
#2

Updated by Henry Dotson 6 months ago

#3

Updated by Henry Dotson 6 months ago

  • Status changed from Open to Review
#4

Updated by Todd Viegut about 2 months ago

This issue officially was revisited on 11-Mar-2024 at the IEC/UCA Joint TF13/14/16/21 Meeting @ Siemens:

Originally, there was a proposal to remove all <<Compound>>s from the CIM. At the time of the creation of this Redmine issue the topic was debated extensively and both extremes were explored. Here was the outcome at that time with corresponding decisions:

Restating once again the decisions made:

1. Compounds are acceptable but should be used judiciously. Not all serialization formats work well with Compounds (e.g. RDF CIM/XML). The idea is to lean towards not creating new <<Compound>> classes as we move forward (and if needed only on rare occasions and only upon joint TF review and mutual consensus).
2. The key concern is that there must not be multiple levels of Compound within a Compound class. This was a primary point of concern.

This is should be closed as we are not planning to remove all <<Compound>> classes.

#5

Updated by Todd Viegut about 1 month ago

An additional comment from the 11-Mar-2024 meetings:

Should be noted that a verification was done on the latest CIM Modeling Guidelines document. It was confirmed that a Rule283 had been introduced that reflects the decisions previously made:

Rule283
A Compound datatype may not contain an attribute with a Compound datatype.

Also available in: Atom PDF