I'm also interested in this subject. I'm wondering if the SDF reader can read a multiple geometry as an aggregate and each part of the aggregate has the geometry column name as its geometry name or a trait. However, I don't have any sample SDF dataset containing multiple geometry features, so cannot confirm that unfortunately.

 

In addition, I would like to know how to write multiple geometry features into a new SDF dataset with the SDF writer.

 

---

Thanks @david_r and @takashi. I will find a technical or strategic expert to respond to the initial questions as well as your follow up questions. Thanks!

---

Hi @david_r and @takashi. Dean replied to https://knowledge.safe.com/questions/40984/how-to-select-the-right-geometry-column-in-sdf-fil.html

---

Hi @david_r and @takashi. Dean replied to https://knowledge.safe.com/questions/40984/how-to-select-the-right-geometry-column-in-sdf-fil.html

Hi @NatalieAtSafe, yes I know that the Quick Fact describes that the SDF reader/writer doesn't support aggregates, but surprisingly, in my observation, the reader/writer can read/write aggregates in fact.

 

In addition, the SDF writer has a parameter called Geometry Column Name, and the help doc says "A comma-separated list of the geometry column names for writing". I therefore thought that the SDF writer supports multiple geometry and the user could specify the geometry column names through the parameter. However, the writer seems to write a multiple geometry as just an aggregate.

 

See also the attached workspace: sdf-vs-mutliple-geometry.fmw

 

---