| [railML 3.3] IS:netConnector and IS:netConnectorRelation [message #4068] |
Fri, 26 June 2026 14:01  |
Mathias Vanden Auweele
Messages: 143
Registered: February 2025
Location: Brussels
|
Senior Member |
|
|
Hello,
IS:netConnector and IS:netConnectorRelation have been introduced in railML 3.3 but show several modelling issues:
1) Since the @id is file specific and it is not possible to refer from file A to an element's @id in file B, the designator should be a mandatory element for the netConnector. That is the only external identification available. Without it, the netConnector is useless.
2) During many discussions on the topic of global identifiers, the coordinators have always said that topology elements shouldn't get a designator subelement. Why is the netConnector than an exception? It is also not an infrastructure asset that can appear in a register.
3) If <netConnector> can get a <designator> subelement, then there is no reason not to give this possibility to <netElement>s as well. In that case, there is no need for <netConnector>s and <netConnectRelation>s as the <netElement> and <netRelation> with <designator> are already sufficient to connect two files together. They would play the exact same role as <netConnector> and <netConnectorRelation>.
The example also only shows usage in 1 file. If this element is intended to stay in railML 3.4, please extend the example with a second file to show how to link the netConnectors.
Best case proposal:
- Best case: add <designator> or Globally unique identifiers to netElements and remove netConnectors and netConnectorRelations
Proposal if best case is rejected:
- Make <designator> mandatory for <netConnector>
- Extend the example with the <designator> and a second railML file
- Explain in the documentation why a <netConnector> can have a <designator> but not the <netelement>
Thanks
Mathias Vanden Auweele
Railway data freelancer
https://matdata.eu
Brussels, Belgium
|
|
|
|
|
|
Members
Search
Help
Login
Home
