Issues #656
Publish/subscribe enum with value > 127
Added by Richard Schimmel over 3 years ago. Updated 9 months ago.
Description
Item 6 from Christophe Camelis list;
Is such enum value allowed by the standard?
Files
Solution to redmine 656 and 657.docx | Solution to redmine 656 and 657.docx | 58.3 KB | Richard Schimmel, 01/09/2023 08:19 AM | ||
Solution to redmine 656 and 657 (Jan24).docx | Solution to redmine 656 and 657 (Jan24).docx | 21.2 KB | Richard Schimmel, 01/24/2023 10:01 AM | ||
Solution to redmine 656 and 657 (Feb21).docx | Solution to redmine 656 and 657 (Feb21).docx | 21.7 KB | Richard Schimmel, 02/21/2023 09:17 AM |
Related issues
Updated by Thierry Dufaure over 3 years ago
8-1 does not prevent it.
Even fixedLength encoding forsees 2 bytes for the transmission of enumeration value - allowing a range of: -32768 to 32767.
Updated by Joel Greene over 2 years ago
- Status changed from New to In Progress
- Assignee set to Richard Schimmel
- Priority changed from Normal to Low
There are no enum values > 127 currently.
Makes sense to test for forward compatibility - GOOSE, client, tool.
Updated by Joel Greene over 2 years ago
- Related to Issues #657: Publish/subscribe enum with negative value added
Updated by IEC 61850 TPWG over 2 years ago
- Due date set to 10/04/2022
- Closed Reason Test Procedure Update added
- Closed Reason deleted (
--Not Set---)
Updated by John Bruder over 2 years ago
This could really complicate the MMS mapping.
8-1 says "the MMS data type shall be the minimum size required to contain the maximum standardized value".
Currently, all Enums map to an 8-bit signed integer.
But if one attribute exceeds 127, it will need a different MMS data type than all other Enums. This can be done, but users will need to know the "maximum standardized value" for each Enum.
Updated by Thierry Dufaure about 2 years ago
To respond to John:
"but users will need to know the "maximum standardized value" for each Enum"
It is not only about maximum standardized value, but minimum extended value. Users can check the EnumerationType from SCL since SCL Ed1.
Updated by Richard Schimmel almost 2 years ago
- File Solution to redmine 656 and 657.docx Solution to redmine 656 and 657.docx added
- Test Case ID set to sGos9
Added solution to update sGos9 with fixed length ENUM value >127 and negative.
Updated by Thierry Dufaure almost 2 years ago
The issue is not only for the scope of fixed-length GOOSE tests, but also for BER-encoded GOOSE tests.
Additionally, the test procedure shall indicate that the EnumType in the SCD is indicating ord values that are negative, or bigger than 129,...
Updated by Richard Schimmel almost 2 years ago
- File Solution to redmine 656 and 657 (Jan24).docx Solution to redmine 656 and 657 (Jan24).docx added
- Test Case ID changed from sGos9 to sGos16
Updated solution with new conditional test case: sGos16
Updated by Thierry Dufaure almost 2 years ago
I wouldn't make the test conditional ENS can be subscribed.
The ENS using values outside the INT8 range, could be in a DataSet with bolean; those boolean are indeed subscribed (application) - and ping-ponged.
When a subscriber cannot subscribe to a type, I think it is meant: the application cannot receive it, but the communication can. Otherwise it is difficut to shre and configure dataSet based on what the subscriber is allowing. Am I wrong?
Updated by IEC 61850 TPWG almost 2 years ago
- Due date changed from 02/07/2023 to 02/21/2023
Test to be updated to remove ENS conditional.
Add a boolean to dataset and verify processing when ENS is not supported.
Updated by Richard Schimmel almost 2 years ago
- File Solution to redmine 656 and 657 (Feb21).docx Solution to redmine 656 and 657 (Feb21).docx added
Updated solution - removed ENS condition and added boolean to the dataset when ENS is not supported
Updated by Richard Schimmel 9 months ago
- Status changed from Resolved to Closed
- Updated Test Document set to TP1.3