Hello friends
My intention is to have a quadrilateral surface between the two triangular panels so that it is exactly perpendicular to both panels. (Note that it is not correct to find the midpoint of each panel and then connect them, because the midpoint of each triangle is different from the opposite triangle). is there any help?
BSSS.gh (11.1 KB)
Hi! Try to see if this is what you are looking for:
Here’s the Gh file:
BSSS.gh (13.1 KB)
@user135 Tell me if you have any questions on how does the script works
1 Like
this is another option that uses brep Edge-Face topology to get edges that have 2 face neighbors
I think I’m using like double of the components needed to do that, but my brain is not wired very well this early 
BSSS_brep_EF_topology_inno.gh (23.3 KB)
I’m also a bit triggered by my definition going from mesh to brep… probably could be done working on meshes exclusively
2 Likes
thank you @inno for your efforts. and i ask @Paolo_Colombo a extra task. i fact i tried to extend the quadrilateral surface in form a shaped-knife as i showed in the figure. but i could not extract a point in the same direction. is it possible?
thanks i advance
Hi @user135 ! I would suggest to use something like a combination of Mesh Closest Point and Mesh Eval or Evaluate Surface to get the normal per each point that you have.
Hopefully this helps.
i tried several time but i cuoldnt because the direction of each panels is not same.
if possible please do it 
thanks
given the previous rectangular surfaces, you could do something like this
BSSS_brep_EF_topology_inno_Re.gh (22.7 KB)
this is taken from my definition, but I believe it might be applied to Paolo’s solution
2 Likes
thaank you so much @inno . this is the same that i want 
1 Like
Hi @inno
i have a problem in my project when i want to create polyline between four points. some points do not follow the reverse rule and they are broken to reach a polyline or surface. please consider it happens for only some complex surface although for other free form surface works fine.
i really need a help.
thanks in advance
Participants.gh (153.1 KB)
I would do something like this, which briefly takes all lists of 4 points, averages them into a central point, and builds on that center a plane that best fits the 4 point themselves → at that point a circle is drawn on each plane, and the 4 original points are sorted along that circle
this guarantees the points are “somehow sorted”, but does not guarantee each list is sorted the same way, and clockwise/counterclockwise depends on plane orientation, which we are not controlling
long story short, it’s not robust, it just does not allow points to be connected in such a way the polyline would cross itself, nothing more nothing less
I don’t know if it’s what you are looking for, if you need something more reliable I think it would be better to dig a bit more into your definition… here I just opened the file and put a fast patch to that problem 
Participants_inno.gh (158.8 KB)
thank you so much @inno for your efforts
but the some of joints have been like a butterfly.
i think this issue is very complex because of the logic of parametric
