Today I encountered a problem with the new Solve Adjacencies component (or at least I belive that it is the source of the problem).
I tried to solve adjacencies and it gave strange results - internal, horizontal surfaces were recognized as roofs.
I used somewhat complex, geometry with small differencies level-by-level but I made sure that all the faces are planar in each brep. I recreated same process using legacy 0.0.66 version and it turned out that the results are different and all ceilings/floors/roofs seem to be solved as expected.
I experimented with rhino file tolerance as well but it did not help. At this stage I ran out of ideas and decided to ask you for help.
I kindly ask you to check whether you could recreate the issue. What could be the possible source of the problem encountered?
I attach a screenshot of results (green surfaces are roofs in legacy, red are roofs in new version) and a gh file (with both - legacy and new version solutions to compare).
Thank you for taking time. I think this may be bacause i saved the file in Rhino WIP. I saved another one in Rhino 6 here. It should have data internalised. In case it does not, I attach *.3dm file as well. I also changed the files in the original post.
I did some more testing and got some other problems with v1.1.1 Solve Adjacencies. An example is below.
Please note different face/vertices count on output mesh of Interior_floors and Ceilings. There seems to be one floor missing (second picture).
Legacy component seems to solve everything in order on the same geometry.
It’s definitely a bug and I’m working on fixing it now. For future reference, it took me a while to find this topic since it was not categorized as being for LBT honeybee. If you do categorize the issue under “honeybee” next time, I’ll be able to find and fix it faster.
You should be able to get it on your end with the “LB Versioner” component in a few hours after the change makes it through all of our tests.
For anyone who might be concerned about whether this issue affects their workflows, the bug only affects the solving of adjacency for specific types of CONCAVE faces.
Thanks again for reporting, @Wujo, and for the clear example that allowed me to fix it quickly. You’re example is running as intended now:
