FME'S native format (FFS) stores attribution and lists as it appears from the source file. So if a SOSI was read in that stored all of its attribution per point per line within a list, then the FFS would likewise store the list.

 

 

In FME's terminology, a line is a single object. It's individual points / coordinates are components of that line. Attribution of the points themselves could be stored within a list akin to how some existing FME readers have Format Attributes that are lists. e.g. some CAD formats.

 

 

Thus for your example network topology the attribution for that line could be:

 

sois_extended_data{0}.pointIndex = 1

 

sois_extended_data{0}.KURVE = 4

 

sois_extended_data{0}.OBJTYPE = Naturverngrense

 

sois_extended_data{1}.pointIndex = 2

 

sois_extended_data{1}.KURVE = 5

 

sois_extended_data{0}.OBJTYPE = Naturverngrense

 

etc.

---

I think you answer is interesting Kathy although I'm not sure I understand it fully. Lets say my datasource is not SOSI, but I am producing SOSI from eg. an ArcSDE DB. This means the 'KP' network topology tag (for connection points) will need to be established in the workspace with the use of a custom transformer or other. Does your answer suggest that I will be able to store this new-found attribution - that references a particular point in a line segment - on that line segment in FME? And let me tell you, I'm not proud of that last sentence - what a mess.. I hope you understand;)

---

I'm not 100% following your question.

 

 

If your source is an ArcSDE DB, is your KP network topology tag recorded as an attribute within ArcSDE? Is it attached to point or line geometries?

 

---

You can create that structure using fme and store it in different formats.

GML is of course very well suited for that. But so is a txt- or csv-file.

---