Area in area problem for overlay analysis?

Just trying to reproduce and understand what you are doing.

• Are any of these polygn features collections ?
• What order is the two layers in the intersect form ?
• What Polygon operation did you use ? Or did you try all of them ?

Usually if you have 2 of the same ploygon feature types, one instance inside of another, you have a 'hole' or donut situation. I would need to understand why you would want a feautre inside of another of the same feature type ( possibly one is a subset of another ??, not sure).

Thanks !

Jerry

Hi Jerry,

I really appreciate your fast reaction.I will try to explain my idea and what I need a bit more but also will try to be brief :-)

I have a feature "Area", there can be more other areas in a bigger one ... I think you can think about the conditions as about parcels or zones also, all these head towards to "shapes in shape" situation. My original idea was to create simple polygon feature and if there will be smaller area in the bigger one, to create the instance as as a grouped holes cell, where the hole will be filled" by another area. I didn't use collections. It is drawn at the attached picture.

It works quite fine except overlay: The results was created correctly, but for areas in the hole no properties were copied to the resulting shape, which makes it unusable as I need to export properties from all ovelrayed areas. I can send you example project including macro I use to create features with groupe holes cells.

Based on that I returned back to more simple situation and I have tried what the overlay analysis will produce if there will two simple area features placed on top of and I was surprised the inside polygon is treated as the hole. This case I tried with the standard geo_example_designer project, in the attached file is small example.

HTH Jan

Jan,

If the "Area 1" feature instance contains the 2 holes as indicated and "Area 2" and "Area 3" fill those holes the intersection spatial overlay should work as expected. In the sample design file you provided, there does not appear to be a hole in the outer polygon. Attached is a design file which illustrates I think what you are attempting to accomplish. It contains 2 polygon feature classes (e.g. FeatureClass1 and FeatureClass2) with a resulting intersection overlay.

Regards,

Jeff Bielefeld [Bentley]

Hi Jeff,

you are righ and it is my mistake I mixed a bit two questions / problems in one thread. So short summary:

• Question 1: Why if a polygon is inside another, the inner is treated as a hole if I use intersection tool? I assumed they are independent and both "solid", so also the inner should be processed as solid.
• Question 2 (the original and more important issue): The situation you created in jan1.dgn is exactly what I tried in the past. The problem with this analysis is that for areas, that cover holes, properties are not copied to the result shapes, only geometric attributes are recorded. In the attached zip file is example dgn and xml for GSA for the project I used to create the example.My target is to have grafically exactly what you create, but with XFM properties also.

With regards,

  Jan

Jan,

I offer the following feedback:

• Response 1 - During construction of the underlying topological graph, the inner polygon is seen as hole since it is contained within the boundary of another polygon. If the 2 overlapping polygons represent instances of the same feature class there must be a hole for the spatial overlay to process the result as expected. In other words, the solid (faces) of the graph should not overlap and occupy the same space (e.g. there should be a hole in the outer polygon). 
• Response 2 - Using your provided example I was able to replicate the problem you observed with the propagation of the business properties to the overlay result. I've logged an issue to have this problem resolved in an upcoming Bentley Map V8i (SELECTseries 1) maintenance release.

Regards,

Jeff Bielefeld [Bentley]

Jeff,

thanks a lot for your explanation of 1st question, it is good to know about this rule and to take it into a consideration. In fact I think such situation is usually the result of not well designed project, but I think it can happen from time to time :-)

Regarding to 2nd issue: Thanks for the confirmation and logging of the problem. As I discovered it during the development and implementation the running project, I am looking forward to the maintenance version you mentioned as this feature is quite valuable and important for us.

With regards,

  Jan 

Jeff,

thanks a lot for your explanation of 1st question, it is good to know about this rule and to take it into a consideration. In fact I think such situation is usually the result of not well designed project, but I think it can happen from time to time :-)

Regarding to 2nd issue: Thanks for the confirmation and logging of the problem. As I discovered it during the development and implementation the running project, I am looking forward to the maintenance version you mentioned as this feature is quite valuable and important for us.

With regards,

  Jan