Why are shape and complex shape elements being imported as linear features?

Hi. I think this has to do with the way the Dynamic Feature Scoring of Bentley Map works. If I understand properly you are working with mostly "plain" dgn elements, you may not have an XFM schema to work with that data. The DFS system will automatically create classes for you to work with your data based on a few criteria. When there are cells involved, it will use the cell name and group all instances on that cell into a single class. In your case with lines and polygons it will use the level name. Bentley Map will create one class per level on which there are plain elements. Since we do not have a class definition we will infer one based on the first instance we encounter. If by any chance you have a line on that level and that is the first element we see then we will think the class uses line geometry. That's a limitation of that inference system. I think to fix your problem you would need to make sure you have only polygons on that particular level.
Hope this helps.

Does the data on all levels in different files need to be clean?

For example, I am referencing in others files - Level SOME-AREA in used but has polygons and open elements on it.

I copy only the polygons to my file from the reference file. Per our standards, these polygons belong on the Level SOME-AREA so that is where I leave them.

Is there no way for map to work with the data on that level in my file without the same level from a reference file confusing it?

Is there a way to pre-create a Polygon Class to be based upon a Level and then, only Polygons from that level get processed?

The strength of many of these tools is the ability to take data that is being used for one purpose to re-use it for another with minimal modifications.

If we always have to modify it from its original format before we can use it elsewhere, that usefulness is reduced and will discourage the workflow from being adopted.

Hi. I think the level names are shared between the master dgn model and its references so unfortunately even if there are only polygons in the master file on that level, elements from the references may cause confusion.
I am under the impression that if you perform the import of your .shp file with Bentley Map, a class will be created for those elements instead of inferring from the level name. Not 100% sure here, I will let others confirm.
You should also be able to attach the .shp file as a reference and work with it that way instead of copying it into your main model. Maybe that would be even easier ?

I need to union three sets of Polygons: 2 are from an online agencies web pages and are shape files. The remaining are drawn by engineers using DGN files. Sometimes they draw all closed shapes. Other times they draw shapes for existing conditions but then draw linestrings where proposed conditions will divide the existing shapes.
Currently, they are manually computing all of the subdivided area and is seemed to be an ideal application for automation. But to prove the concept, I needed to get something to work. So I tried a few example areas and kept crashing and having issues like getting linear features instead of closed shape features.
At this point, I think I have a handle on the limitations. But until I can make it work consistently and reasonably easily, there will be push-back from the masses.
Thanks for all your efforts and information.

Unknown said:

and is seemed to be an ideal application for automation.

Yes, it's probably true.

Unknown said:

At this point, I think I have a handle on the limitations.

I don't think so. Maybe there is no way how to achieve the full automation, because GIS environment is far far more strict than CAD, so conditions like "it's shape, but maybe it would be also line string sometime" are not acceptable, but often some workaround can be implemented, depending on local conditions and workflow.

But because you have not provided any real data (in a form you mentioned: two shape files and one dgn file at least), it's completely imposible to investigate anything. There are different tools available in Bentley Map, different workflows can be used and also DFS technology mentioned by Martin can be configured, so without real data the discussion only is time wasting effort. And if everything will fail, there are tools like VBA and MDL that can solve many problems.

Regards,

  Jan

Unknown said:

and is seemed to be an ideal application for automation.

Yes, it's probably true.

Unknown said:

At this point, I think I have a handle on the limitations.

I don't think so. Maybe there is no way how to achieve the full automation, because GIS environment is far far more strict than CAD, so conditions like "it's shape, but maybe it would be also line string sometime" are not acceptable, but often some workaround can be implemented, depending on local conditions and workflow.

But because you have not provided any real data (in a form you mentioned: two shape files and one dgn file at least), it's completely imposible to investigate anything. There are different tools available in Bentley Map, different workflows can be used and also DFS technology mentioned by Martin can be configured, so without real data the discussion only is time wasting effort. And if everything will fail, there are tools like VBA and MDL that can solve many problems.

Regards,

  Jan

I am quite pleased with what this exercise has shown me. From here, I know that in order to move forward, specific workflows will be needed. The general approach is not flawed, only my initial efforts, due to my unfamiliarity with Map.
We have multiple seats of Power InRoads and a similar number of seats of Civil 3D. Both have a Map product that is part of their installation and licensing that can work with GIS data and perform GIS tasks. On the other hand, we have only one seat of ESRI and that seats is almost always in use for other tasks.
There are a lot of activities that go on in an office like ours that can benefit from automation. But if often takes a fresh set of eyes to see the possibilities.

Chuck,

First of all I have to comment on your forum signature, specifically the IGDS reference and image containing the Intergraph Dual-Color Interact and InterPro workstations. That takes me back many moons. I remember those days well.

Next depending on your specific output needs you may be able to define a Dynamic Feature Scoring (DFS) rules XML file that instructs Bentley Map on how to infer feature classes from your data.

In addition you may be able to produce the desired set of output polygons using the Bentley Map scripting engine which has extensive spatial data processing tools, including the ability to process a mixture of polygon and linear data, including segmentation of overlapping linework or boundaries.

Please contact me offline via direct message if you are interested in learning more.